User talk:66.16.144.18

Welcome!
Hello, and welcome to Wikipedia! Thank you for your contributions&#32;so far. I hope you like the place and decide to stay.

Here are some links to pages you may find useful:
 * Contributing to Wikipedia
 * Tutorial
 * How to edit a page and How to develop articles
 * Simplified Manual of Style

You don't have to log in to read or edit articles on Wikipedia, but if you wish to acquire additional privileges, you can simply create an account. It's free, requires no personal information, and lets you:
 * Create new pages and rename pages
 * Edit semi-protected pages
 * Upload images
 * Have your own watchlist, which shows when articles you are interested in have changed

If you edit without an account, your IP address (66.16.144.18) is used to identify you instead.

We hope that you choose to become a Wikipedian and  [ create an account] . If you need help, check out Questions, ask me on my talk page, or ask your question and then place  before the question on this page. We also have an intuitive guide on editing if you're interested. By the way, please make sure to sign and date your talk page comments with four tildes (&#126;&#126;&#126;&#126;).

Happy editing! I dream of horses If you reply here, please ping me by adding to your message. (talk to me) (contributions) @ 03:04, 10 May 2015 (UTC)

List of football stadiums in England
Thanks for the update on this - I've done the 'big summer update' the last couple of years and it really is quite a cumbersome chore, so much appreciated! Super Nintendo Chalmers (talk) 10:32, 26 May 2015 (UTC)


 * LMAO ... well, believe me, if I had fully known what I was getting into before I started my edits I don't think I would have done so; I would have just left the seasonal updates to some other sucker person such as yourself. :) It wasn't as if I intended to update all the teams when I started; I thought I would just update the Premier League and Championship club stadia info.; and leave all the other clubs for someone more familiar with the lower leagues than myself to complete.  Which is why I ended my first edit after updating only the largest 70 stadia. When I realized I had made my Notts County error and had to make a further edit to correct it, I thought perhaps I should go the extra distance and complete the League One club stadia too - having updated the 70 largest stadia I was only 7 clubs short of achieving that. So I pushed on with a further edit without realizing how far down the list Fleetwood Town's stadium fell - hence my 107 stadia cutoff for my third edit. When I realized I was now only half a dozen stadia short of completing the main list (the ones prior to "Other Listed Stadiums") I thought "what the hell - let's be done with it"! Talk about pulling a loose thread and seeing things unravel... by the end, it had indeed became a most "cumbersome chore" and all I could think about was how silly I was not to have quit (after just the first 70) when I had more than achieved my initial goal. :(
 * Anyway, thanks for taking the courtesy and trouble to drop me your line of appreciation. You're very welcome. It is my experience that most Wikipedia editors are simply egotistically confrontational and boorish, which is why I stopped editing from my user accounts and now just do occasional updates as an anonymous IP user when I think I can do a quick fix (which, as I learned to my cost, this article update was most definitely NOT!). Your unsolicited courtesy was very refreshing and much appreciated. And I'll now happily leave the "summer updating" of all the other lists of English football stadia on Wikipedia in your very capable hands. :) 66.16.144.18 (talk) 18:45, 26 May 2015 (UTC)

List of European stadiums by capacity
Hey, So I think that you're right in identifying it as a page that needed some attention! If I was putting this page together my inclination with the 'retractable seats' etc would be to just give the highest maximum regular capacity of a stadium: it saves the need for annotations. However I think that I have a slight editing bias for removing info in favour of neatness, and I can see the merit of your fix (which works).
 * I have no problem with removing irrelevant info., especially if I know of a more appropriate location in Wikipedia to go include it instead. But I have little time for the "form over substance" types that would have Wikipedia articles read like those flashcards they use in remedial education. It's an encyclopedia. It's meant to contain facts. Lots of them; sometimes in excruciating detail. But they must be pertinent facts and presented in a well-structured and accessible manner. And that's the tricky bit! But ease of editing should NEVER take preference over being an accurate and useful source of information.

The two paragraphs "Also note that stadiums...rather than "regular", "permanent" or "primary" use" could be formatted into a footnote rather than included in the lead.
 * The problem with footnotes is few, if any, people actually read them. The problem with the text you referenced is that it is really instructions to other editors and should not be visible to the users of Wikipedia (as opposed to other editors). The correct place for it is as a comment in the text so that editors see it rather than users.  The problem with that is there is no way to make such guidance stand out from all the other text - such as being in a red or bold font - so it gets lost in amongst all the other textual clutter. Or purposely ignored ... but a more readily identifiable "guidance to editors" font won't ever solve the problem of other editors choosing to ignore the instructions.

With regards to tenant - one way of solving that is to replace it with 'owner', whether that's FC Barcelona, The FA, Consortium Stade de France, Milan Municipality.
 * I disagree. Not all clubs are owners (e.g., Man. City do not own the Etihad Stadium).  However, even less clubs are tenants - that is, most of them DO own their stadium. I submit that the current column title should really be, "Owner or tenant sports team(s) calling this stadium home" or perhaps something equivalent but pithier. However, all of the event/venue stuff isn't in there because prior editors didn't know the meaning of the word "tenant". It is (and this is just a hypothetical example) because instead of someone entering "Consortium Spyros Louis Olympic Stadium" (or whoever the owner really is) for that stadium, they instead entered something along the lines of: "Greek national athletics and 2004 Summer Olympics team". Then a subsequent monkey-see, monkey-do editor interpreted the word "Olympics" as being an event rather than the name of a tenant athletics team, and so he entered World Cup info. or some such in the entry he was changing, and then the whole thing just escalated over time, as I tried to explain to you in my previous message. IOW, whatever that column is currently called is completely irrelevant while all that other venue/event stuff is also included in the table for all the other entries.
 * Even if/when it is removed there is nothing to stop the whole process from starting all over again. Because the column name is really a sort of instruction to other editors (e.g., please enter tenants in this column) but because the table is so humungously long, for 98% of the entries being edited that column name is off the top of the visible screen out of sight (and thus out of mind) for editors updating specific entries. So instead they will look at the collocated entries nearby for a reminder of what they should be entering in that column (I do that a lot myself). If the nearby entries say things like "Olympics" and "World Cups" they go, "Oh, right," and proceed to add a Euro or Confederations Cup competition for the stadium they are updating. It's just another case of the editorial guidance being lost, this time not due to being hard to spot in amongst other textual clutter, but because the table column headings are off the screen and nowhere near for most of the entries being edited. Once again, if the reason the editor is editing the entry in the first place is because he's read the table and seen all the events listed for other stadia and he wants to make sure the stadium of his preferred club isn't left wanting in comparison, then NO amount of editorial guidance - no matter how bold, nor how red, nor how frequently and emphatically stated - is going to stop him!

This would then create scope to strip-out all the events material.
 * We don't need to change the column title to "Owners" in order to create such scope - we can do that right now based on its "Tenant" title. Because events and venues are no more "tenants" than they are "owners". What I need from you is a commitment to help sort this table out in a planned fashion with me. I don't just want to delete this info. cold turkey and completely reset the table because that will very likely trigger adverse reactions from past editors that have placed that info. in the table.  We need to strategise a plan to systematically remove the info. in an organised fashion.  For instance, by placing a link to an article such as this one plus some verbiage into the lead-in text while at the same time removing all references to the Olympics from those stadia that contain it, most/all of the editors that might otherwise have had conniptions if that info. just disappeared would probably see that as a better and more efficient way to link that info. to the table (rather than just trying to cram it in the "Tenants" column).
 * References to World Cups and Euros, etc. can be similarly removed by placing a link to an article containing that info. elsewhere. That is the approach I wish to first strategise with you; and then we can divvy the actual execution of the work up. Also, I need someone to cover my back; to back me up when I do my part in some of this removal because I'm editing from an IP address. Most people don't get upset when you add stuff from an IP address, but they tend to view removal of anything they are emotionally attached to as vandalism, especially if done from just an IP address. IP addresses are actually less anonymous than made-up monikers but most Wikipedia editors don't appear to understand that fact!

If someone then wants to find out about the various events and uses of Twickenham, or wherever, then they can read the article! We could throw in a 'sports' column but this does create the problem of arbitrary definition. Finally - perhaps killing the 'UEFA stars' column is appropriate, given that this is not a football article.
 * Now you are getting me upset because I just spent a chunk of time improving that column! So I'm emotionally attached to keeping that column based on the amount of time I've already got invested in it. LMAO. The problem with the UEFA stadium designation is that UEFA doesn't ever print any kind of list of all its 4-star, 3-star, 2-star, etc. stadia so that all you have to do is copy it and then reference the UEFA list as your source and then you are done. If only life was that easy! What UEFA does publish is the criteria that each stadium has to pass in order to get those designations, and that subject matter is covered here. It was that article that was my point of entry to the list article we are discussing. IMO, the authors of that article were quite right not to reinvent the wheel (and come up with an almost identical list article for their purposes) but to adapt the existing article instead. By adding that additional column. So don't be too quick to remove it - pull one piece of thread and another one unravels!
 * Once all the extraneous stuff is removed from the "Tenants" column we can probably restrict it in width for what it has to contain thus freeing up more space for an additional column or two. I don't need to put the text, "This stadium uses whatever seating" into that column - I just need something to attach my ref notes to (which is the main purpose of that string). We could codify all stadia into one of the four classes (that I identified in my previous message to you) upfront in the lead-in text, and then introduce a new very thin column next to the "Capacity" column that contains the stadium's 1-4 codification. I could attach my ref notes to the codified number in that column. That's how my strings can be removed from the "Tenants" column. What we need to come up with is similarly creative ways to remove and codify the other event/venue info that is already there, or place a link in the article to where else on Wikipedia that particular info. is handled in all its glory so we can eliminate the info. in the "Tenants" column that way.

Removing this and a shorter tenants column would created space for a thumbnail image - this works well at List of football stadiums in Scotland. --Super Nintendo Chalmers (talk) 08:31, 26 June 2015 (UTC)
 * Yes, that is exactly the clean look I would want to get this article list to end up having. I'm surprised you were able to find images for so many of the stadia. That is one thing we would have to investigate for the Euro table - there is no point in planning to go in that direction if there only exists images in Wikimedia Commons for 20% of the stadia. But one thing I must point out is that there are only 85 stadia in that Scottish list - there must be closer to 300 in the Euro list, so it will be a whole level of effort above and beyond. 66.16.144.18 (talk) 15:58, 27 June 2015 (UTC)

Player lists
Thanks for your efforts on the player lists.
 * You're very welcome.

The "main" list suffers a bit from having started out as a lightly sourced list back when a couple of general non-specific references was viewed as enough. Its easier for the odd thing to fall through the cracks when the sourcing is being done retroactively.


 * It doesn't "suffer a bit" by "being lightly sourced"; the article always was - and still is - inadequately sourced by Wikipedia standards. Hopefully we can correct that rather than having to delete the article instead. Nor was there ever a time on Wikipedia when "a couple of general non-specific references was viewed as enough" - except perhaps by some of the more "tribal football" editors who hoped they could rationalise (as much to themselves as well as to others) their recently inserted POV data which couldn't be justified by a more acceptable defence! But they don't count.
 * If only "the odd thing fell through the crack" regarding this article then it would have taken me only a few odd edits to fix them, wouldn't it? Go count my edits ... even discounting the ones that add new info. or re-style or reformat things in a better manner, I think the remaining corrective ones still total way more than what one could normally describe as, "just the odd thing or two that has slipped through the cracks"! I can see you are clearly a gentleman with a preference for euphemism rather than stating things plainly.

Positions are always going to be open to a certain degree of interpretation. Lee Crooks, for example. He certainly did play midfield at times, yet his longest run in the team was at right back, including the 1999 playoff final. I generally prefer having fewer categories to reduce the overlap, but its largely a matter of personal preference.


 * I don't disagree; my change to Lee Crooks' position was my mistake. I will revert it back to DF next time I take a look at that article if you don't beat me to it. Or maybe, in his case, the proper designation should be "DF/MF"? Likewise, my change of George Livingstone's position from FW to MF was merely an anachronistic gaffe on my part. But HB is still a midfield role while FW was clearly wrong for a HB; I merely applied the post-70s designation instead of the pre-70s one. However, it was a very time-consuming edit requiring multiple passes of the list, and GL's change came near the end of it, by which time I may have gotten a little sloppy because I simply wanted to be done with the long edit. BTW, I'm still not done. I'm just taking time to first respond to your message before continuing ...

Centre half is a particularly nebulous one, with its gradual movement backwards over time. Sam Cowan, for example. He would have started his City career in the 2-3-5 centre half position i.e. the middle half back, but he would have ended it as a W-M centre half - from the middle of the park to the back without ever changing position.
 * I don't believe much, if any, of the "centre half" information I've changed was due to CH being a "nebulous" designation, or even to it being later vandalism that nobody noticed. The changes I made were mostly due to the original data entered being poorly researched or misinformed. Perhaps some of the HB positions that were entered as "CH" could fall under the "nebulous" argument, but entering "CH" for someone who was a FW or FB simply smacks of an editor having shoved some designation, any old designation, into the position field rather than take the necessary effort to at least get the position close to being right.
 * WRT to CH being a position that has transitioned over time, I quite agree, but that still doesn't explain the misidentification of FBs or FWs as CHs. I also disagree with your assertion that the CH migration happened within the playing span of a single player such as Cowan. The full transition of the CH role has happened over a period spanning more like 80 or 90 years, rather than the ten year playing career of Cowan that you just claimed. But I fully agree it has transitioned, and that it has moved from being a single player central midfield role to a dual player deeply defencive one. Cowan, Ewing, Heslop, Booth, Watson, McCarthy and Kompany are all centre halves (if you prefer the "half back" terminology) or centre backs (if you prefer the alternative "wing back" terminology) and although the responsibilities and focus of these players has been different at the various eras in which they played, the commonalities of the role in each of the different eras far outweighs the differences. When he was still playing, Tommy Booth was considered to be both a "centre half" and a "midfielder" despite the fact that he wasn't doing very much in the middle of the field, and the other "half backs" (Doyle and Oakes) alongside him were doing a lot more than what had been earlier expected of a traditional 2-3-5 formation wing back. In fact, the term "2-3-5 formation" is pure revisionism - when teams were playing that formation they never referred to it as such (mostly because at that juncture it was the ONLY known formation and didn't need to be distinguished from any other one). It was only AFTER the advent of the 4-2-4 formation that earlier formations were referred to by similar numeric designations; likewise it was this same numeric designation that finally had people now correctly calling the centre back role a defencive one instead of a midfield one, despite the fact that it had become a purely defencive role twenty or more years prior to that.
 * The bottom line is a CH is a CH is a CH, whether he played in 1933, 1973 or 2013. The focus of the CH role may have transitioned between those years but at no point has a CH been a GK, a FB or a FW. Which is why I corrected those false designations, and your point about how "nebulous" the CH position is nothing more than a complete red herring!

Looking over the article history I think the origin of many of the misidentified centre halfs is here. Thanks for clearing them up. Oldelpaso (talk) 12:41, 5 July 2015 (UTC)


 * The way the article was sourced - and still is (because I haven't changed that area of the sourcing) - the designated player positions are implicitly sourced to James by default just like the appearances and goals data are. The reader is led to believe that the whole of the table comes from James except for the later player entries whose information is sourced from Soccerbase instead.  The implication is that the positional info. for the Soccerbase sourced players similarly comes from Soccerbase.  If the position info. was obtained separately from the MCIVTA web site then it should have been clearly sourced as coming from there. 66.16.144.18 (talk) 12:27, 6 July 2015 (UTC)


 * Re Oakes league appearances. Soccerbase says 565. Gary James' books say 564 (561 starts, 3 sub). So does the Official Handbook series. Andrew Ward's The Manchester City Story says 565 (562+3). Don't much fancy a manual count, but it may prove necessary. One possible source for discrepancy might be the abandoned match against Newcastle in 1969? Oldelpaso (talk) 13:04, 5 July 2015 (UTC)


 * Other than an overlooked sub. appearance that is most likely it. Abandoned matches are pretty rare so I'm unsure how they are normally handled. But I would be very surprised if they and their goals were included in total appearances and goals scored tallies by any of the reliable sources that maintain such stats. The article's sourcing information was currently missing any statement regarding how abandoned matches were handled, so I went ahead and added a statement to the statistics sourcing information to the effect that any goals and appearances in abandoned matches are excluded. 66.16.144.18 (talk) 12:27, 6 July 2015 (UTC)


 * Eric Brook / Tommy Johnson - Brook scored a goal in the abandoned 1939–40 season. Officially, these matches were expunged from the records but some sources still include them. This is a common discrepancy. As for 1939 or 1940 playing date, his last game was in the 1939–40 season, but it came in November 1939. Oldelpaso (talk) 13:28, 5 July 2015 (UTC)


 * Yes, but there were discrepancies at both ends of the date range. The lead text originally stated "scored between 1928 and 1939" while the table stated his club career span as 1927–1940. I wasn't sure if this difference was telling me something - i.e., that he made appearances for the club in 1927 but didn't score his first goal until 1928, and likewise that he played his final games in 1940 but also didn't score any goals that year - or was due instead to simply a sloppy inconsistency. Which is why I tagged it for attention. Given all the sloppy mistakes in the position column I strongly suspected that the differences were simply due to inconsistency rather than Brooks having a goal drought in both his first and last years at the club. Your changes and comments have explained and corrected the discrepancy in the last year of his career span, but there still remains one for the beginning of his career span (1928 in the lead text versus 1927 in the table). Is this due to additional inconsistency oversight or did he really not score any goals at all in 1927? For such a prolific goal scorer this seems highly unlikely to me. 66.16.144.18 (talk) 12:27, 6 July 2015 (UTC)


 * The article was first created (by me) in 2006. No doubt there are errors in it that are my fault. I was a lot less experienced. Wikipedia was a lot more relaxed about referencing; for example the Template:cite web used for inline references had only just been created. You really could use "a couple of general non-specific references" back then. The first ever featured article about football looked like this. Standards have rightly risen since then. There's no way I'd have used the MCIVTA source in more modern times. We've both spent plenty of time on the list, and we both want it to be accurate. I'm sure we can agree on that much.
 * Well, I didn't come here to rag on you. I didn't even access the article to edit it. I initially accessed the article to look something up ... you know, actually using Wikipedia as an encyclopedia sort of thing. Some people do actually use Wikipedia that way! But every time I've accessed this article in the past the screwed-up format (with the Nasri/Aguero image from one section intruding into the section below, which contains the main table, so that the intrusion caused the table columns to be squished up) has always bothered me, but I refrained from doing anything about it and left it for someone else to fix (which you have to do if you are in reading mode and on a time schedule). This time around I decided to add the missing 'clear' command needed to fix the problem, but as I accessed the edit function it occurred to me that I could also add more text to the lead so that the photo no longer finished below the bottom level of the text. So that's what I did, and now what started as a mere 10 second edit has escalated into a multi-week project!


 * Brook made his debut in March 1928. Complete Record's player summaries are listed season-wise, meaning it lists 1927 as the start as his debut was in the 1927–28 season. Is it your preference to change them to exact date where this differs? It can still be sourced by the book that way, but will need to be to two separate locations each time. Once to the alphabetical player list with totals, and once to the season page with the debut on it.
 * You claim to be doing that already in the "25-99" table. I "upgraded" the text in the stats. sourcing and key section by first copying over the bulleted text from the corresponding section in the other article and then comparing what each comparable bullet stated and choosing the best of the pair. I may have reworded the result a little for greater clarity and succinctness but NOT to change meanings. However, in the case of the text bullet describing the "Club career" column - which had the mouthful of a title "Manchester City career" requiring two lines of heading before I changed it - there was no defining description for that column, so I simply adopted and used the copied text bullet from the other article. I realized at the time I might have changed the original intended meaning of the date range in the way you just pointed out, but I felt that having some definition and clarity was more important than the vagueness caused by the prior lack of definition. These tables are about the appearances and goal-scoring feats of the City players, not about their contractual dealings, so I particularly liked the "irrespective of how long the player was contracted to the club" caveat you had for the comparable column in the other table.
 * Contractual dealings are private and personal issues between the club and player. First team debuts and last appearances, however, are a matter of public record because they are documented by a long journalistic history of match reports from multiple sources. So the article should stick to only those factual dates. Otherwise it will get into problems with players such as Paul Lake who was employed by the club for nearly 30 years but was only a first team player until 1996 (I think - see how nebulous this gets!) when the team got relegated, but never played for the first team again after 1992 when he suffered his knee injury. The true contractual status of Lampard would be another source of uncertainty, not that any of the tables distinguish between loaned and directly contracted players anyway. But it's all irrelevant. All that matters is that Lampard wore City colours and played and scored goals between 2014 and 2015.


 * I don't think we'll ever agree with regards to books / online sources. Web sources regularly turn to dead links. Sometimes they've been archived by the Wayback Machine and can be rescued. Often they have been lost forever. Books, on the other hand, stick around for centuries. Libraries may not be as convenient as the web but they're a whole lot less ephemeral. Even in the most extreme cases with a tiny print run, any title with an ISBN will be at the British Library or Library of Congress. And with things like WikiProject Resource Exchange/Resource Request, geography isn't necessarily a barrier. Oldelpaso (talk) 20:57, 6 July 2015 (UTC)


 * I have no problems with books. I have over 5,000 of them on my bookshelves and I normally have my head buried in one for 2 or more hours of every day. I much prefer and respect books to most of the crap I have to sift through on the web. And if I didn't like or trust books I wouldn't have asked you to source my "Prolific goalscorers" table from one, now would I? But Wikipedia isn't built to handle books; or perhaps more accurately, because it has grown so popular since its introduction, book cited references can't hold their own versus web cited references.
 * Here's the problem. After I did my first round of edits and tidied the article up a little plus added new material, somebody who obviously watches this article but thinks it a waste of time contributing to it, decided to update the stats. of all the current players from the matchday programme for the club's last game of the season, despite the fact that someone - you, I believe - had already updated them from Soccerbase. In fact, he said he was correcting them. But he didn't change the sourcing; he just changed the numbers that were being reported as being sourced to Soccerbase. So I reverted his edit for not properly being sourced - Soccerbase did not validate his numbers; it only validated the data he overwrote. But I made a note of his changes and started to investigate them. His changes were the cause of my adding the footnotes. Every number he entered that I've investigated so far (I still haven't finished yet; I just wanted to come up with a scheme that solved the issue and see if it worked) has been correct.
 * Ignoring the fact that he didn't do the update properly and a matchday programme is probably a less reliable source than Soccerbase - although that's a contentious claim since the true source of all the stats. is the club itself who also publish the programmes, so coming directly from the club the programmes are a primary source of the data, while Soccerbase as a third party gatherer of such data is only a secondary source - what if that editor had instead made some changes to the numbers (let's assume not correct ones) and sourced his changes to a more recently published book about City than James' tome. As long as his ISBN checked out (and wasn't bogus) how am I to determine whether to accept or revert his edits as the next editor looking at the article? Given nothing else to go on I would probably choose to accept his correction of your data since it came from a more recent and ostensibly just as reliable a source. When you saw that editor's changes you would probably revert them, but if he was persistent and kept re-entering his changes and it got as far as becoming an edit war, how would the adjudicating moderator know in who's favour to arbitrate? He would probably use the same criterion as me (and presumably a lot of other interim editors that had also seen his changes and had likewise adjudged them to be copacetic too).
 * With a pretty rare (i.e., almost inaccessible) book source such as James or Baskcomb what you are effectively saying to other editors is, "This data is true, trust me." In your dreams, exactly how many copies of their books do you believe exist outside of Manchester or the U.K.? Baskcomb is just an annual handbook and thus doesn't have an ISBN, so it is hardly likely to show up in the Library of Congress. Which completely undermines your whole argument because Baskcomb is the one I replaced by a web RS citation, not James, so your argument that it is a superior reference to the one I used because publications with ISBNs are theoretically available in all major national libraries is completely irrelevant since it doesn't have an ISBN. And even if it did and was in the national library in Washington D.C., do you really believe I, or any other editor in California, is going to take time off work and spend the money on a 6,000 mile round trip to the national library simply to dispute or defend one of the numbers in your table?
 * Furthermore, how is your saying "trust me" regarding your table data any different from the guy with the matchday programme who just wants to see the correct data in the article? I could have simply gone ahead and stuck those other 21 M.E.N. article citations in and changed the existing data to be consistent with them accordingly, and from my perspective be done with the update, and there would not be too much you could do from your end about that (for all the reasons I've stated). My intent is not to boorishly push my own agenda or get my own way, but to ensure that the article is as accurate and truthful as it can be, and that the data in it is as defencible as possible by being web based. In not adding those other 21 M.E.N. citations I am showing you that I respect your editorial integrity and only wish to use any data conflicts as a means to focus in on where the current data may be erroneous. If the errors turn out to be in Scott's articles then we just ignore those articles; there's no intent to use all 50 anyway, only those that support James' data.
 * As for cited web links becoming deadlinks, only the laziest editors allow that to happen. If you take any pride in your article then you will check them regularly - say once a year - and replace any that have gone dead. Also what you choose for a reference source has a big determining factor on how long the web link will be good for. The material on some web sites is notoriously transitory and shouldn't be used no matter how pertinently it supports your information. Citing a less pertinent RS from a more permanent source, such as a national newspaper that archives and keeps old articles web available, will always be preferable than a web site like the Bluemoon nor Happy to be Blue! The likely useful lifespan of that Happy to be Blue! link was always my major concern regarding it, rather than whether the site as a whole was objective and neutral enough to meet WP:RS. And you also need to take the trouble to cite your source material fully, such as giving it a title, because in most cases deadlinks are caused simply because the referenced source gets a new path name when it is archived, thus breaking the old saved URL, but a Google search on the article's exact title will immediately locate the article again with its current URL. I had to toss away one of the deadlinks I was trying to rescue in this article simply because whoever had created it had skipped over giving it a title, so I had no idea what I was searching for. Making your own title up, rather than using the source's actual headline (misspellings, mispunctuation, warts and all) will cause the same problem once the link dies, although the made-up title might still give sufficient clues to someone regarding what to search for. 66.16.144.18 (talk) 04:20, 9 July 2015 (UTC)

I was aware that Ric from Bluemoon had been working with Gary on getting his stats online but hadn't realised it was now complete. I shall have to look at some test cases to establish exactly what is and isn't included. If they're consistent, links to that could replace a lot of references to Complete Record. Related to that, Gillespie's tally of 30 goals in 33 in 1902–03 in the book matches up with its counterpart here, but there's no convenient season summary cite. His next best total was 21 in 1903–04, so there's no other candidate season.

With regard to formatting and widths, there are so many different screen sizes and resolutions in use these days that I have no idea what would be recommended any more, or what is or isn't preserved in the layout for mobile devices. It was a lot simpler when all you had to worry about was 1024x768. Choosing a column layout that follows the Soccerbase splitting makes sense. Thankfully the majority of pre-Soccerbase players have no need for an entry in the "other" column, and its only really the 1970s where there is scope for ambiguity. The flags / nationalities thing gets very heated. There are highly vociferous factions in more or less every aspect of flag use in articles. It makes little difference to me, though it was put in that format because of Manual_of_Style/Icons. The arguments that go on about flag-related matters were one of the reasons I chose the international recognition format in the <100 lists. That and avoiding the dangers of assigning a nationality to someone born on the island of Ireland. Given the difficulty I had in sourcing some of the international cap numbers, I'd consider leaving out the cap number column. Another example of something that seemed like an easy change but turned into days of work.

I understand your position re web sources. Its forever a balance between source quality and source accessibility. Soccerbase being an apt example. Occasional inexplicable omissions, practically impossible to get them to correct errors and so on, but most of the alternatives are more flawed still. I'd take issue with how easy it is to detect deadlinks though. I must have added hundreds, maybe even thousands of links in citations over the years. Given that I don't often have that much time to edit these days, if I checked them all annually my Wikipedia editing would resemble the painting of the Forth Bridge - by the time I'd finished it would be time to start again! Oldelpaso (talk) 18:24, 12 July 2015 (UTC)

References for new Most prolific goalscorers table
BTW, are you able to source from James my individual line entries in the "Most prolific goalscorers" section that I added? Ideally, I would like to source that table from the web, but at this point any reliable sourcing would help. Thanks. 66.16.144.18 (talk) 12:27, 6 July 2015 (UTC)


 * Goalscorer refs done, except two. I have 43 apps for Browell (42 League, 1 FA Cup), and 33 apps 30 goals for Gillespie (League 32 apps 30 goals, FA Cup 1 app 0 goals). Oldelpaso (talk) 20:57, 6 July 2015 (UTC)


 * Thanks for that. Regarding Browell, just go ahead and change the 42 to 43, recalculate the goals/game ratio, and source it. Whoever compiled that information either was unaware of the FA Cup or purposely ignored it because it makes his goals/game stat. slightly better. Or I could have just fat-fingered what I originally copied down and hit 3 instead of 2. I've no idea where I got this info. from, and I've had the data for everyone up to Agüero for quite awhile, otherwise if I'd first run across it during the Wikipedia era I would have also noted down the source on the basis I could later use it somewhere on Wikipedia. As for Gillespie the error is more complicated. 30 goals from 33 appearances yields a much better goals/game ratio and should be an entry in its own right. Meaning that, in his case, I wonder if the error lies in the season selected and if he should also have a second entry in the table for the 33 goals in 38 games, and that pairing comes from a different season? Could you please investigate? If he doesn't come close to scoring 30 or more goals in any other season then I have no idea why that data is so way off (I've double-checked that I've transcribed it O.K. from my records), but I guess we just have to correct and source the entry that's there. 66.16.144.18 (talk) 04:20, 9 July 2015 (UTC)

Player of the Year awards sourcing
Last one, I think - not sure the "Happy to be Blue" website meets WP:RS. Oldelpaso (talk) 13:28, 5 July 2015 (UTC)


 * Coming from someone who just justified using the MCIVTA site as a valid citable source for the various player positions that comment is really very rich! It never ceases to amaze me that editors that are too damn lazy to properly source their own data input always immediately find the time and energy to nit pick someone else who takes the trouble to actually find and add sources for their unsourced material rather than simply nix it as being improperly sourced (which I was well within my rights to do). I'm fully aware of the dangers inherent in using that web site. In addition to possible WP:RS issues, my biggest concern is that the site is not going to be around for too much longer, in which case all but the latest two years of the PotY awards will be left sourced to a deadlink. However, what data is to be found on that site for the few years up to 2012 is actually pretty accurate and well presented, which is more than I can say for the poorly presented decade-or-more old crapola to be found on the MCIVTA site. So with one link I was able to source all of the PotY award data up to 2012, rather than have the data up to 1995 sourced from a book no other editor has access to, with the remaining 17 years of award data sporadically sourced with mostly deadlinks. If you look at my edits immediately prior to my making "Happy to be Blue!" the main source of the PotY list data you will see that I was systematically fixing what few years were individually sourced with deadlinks and finding and adding individual RS citations for the vast majority of those 17 unsourced years between 1995 and 2014. I was 3 players (Tiatto, Wiekens and Brown) away from completing this process when my Google search for Tiatto returned the page I'm now using. I was completely unaware of the "Happy to be Blue!" web site until I got that hit.
 * It may be a bit of a dubious RS but it is still decidedly superior to a cited book which is inaccessible to other editors for the purposes of entered data veracity verification plus 17 years of either unsourced data or data sourced with deadlinks! IMO book citations are little better than deadlinks and, although not forbidden by Wikipedia, they run contrary to the whole philosophy behind the way an online encyclopedia is meant to work - which is that true information that is entered ultimately sticks (because it can be supported by RS) while blatantly false and dubious information eventually gets removed (because it cannot be supported by adequate RS). If other editors (the potential removers or correctors of false or dubious information) cannot access the supporting cited sources to determine veracity then the whole process breaks down. BTW, I could just as easily have used this web page as my supporting RS (as it is even more up to date than the "Happy to be Blue!" one) but a quick glance at the wording in it and the flag icons used (including the old Georgian flag for Kinkladze) clearly indicates that this site has copied its data from the article, so to cite it as the supporting RS for the article would be to set up a circular reference loop which I'm pretty certain, without even checking, violates WP:RS. I don't believe we have that problem with the web page I used. But if you are that concerned why don't you do some Googling to see if you can't locate an alternative single web page source for the data.
 * What's really annoying (ironic?) about the PotY sourcing situation is that, for the most part, it is completely uncontentious information. You, I, or another dozen or so editors will most likely all agree that it is correct, and in that context, it really does not need to be supported by a RS because no one is threatening to remove it as being inaccurate or false. We wish to support it with RS citations merely as a matter of following good Wikipedia editorial process. Meanwhile, the article is still rife with contentious incorrect statistics and fallacious player position information all of which remains unsourced or sourced to a non-web accessible (thus effectively non-verifiable) book source. Your whole take on this topic seems a little unfocused. 66.16.144.18 (talk) 12:27, 6 July 2015 (UTC)


 * As it happens I have done some research on the subject of Player of the Year. I was unable to satisfactorily resolve one particular anomaly, and for that reason I left it alone pending further research. For the Player of the Year in 2003, different sources list different players for that year. The original reference (long since a deadlink) said Distin. There are other references out there that say Berkovic. I've looked through the Manchester Evening News archives for that period and unusually for the MEN there wasn't a single article about the awards. There is a solitary reference to "reigning player-of-the-season Sylvain Distin" in the October. The award used to be administered by the Official Supporters' Club but is now done by MCFC themselves. My suspicion, that I have been unable to confirm, is that 2003 may have been the transition between the two. There may have been two different awards that year, one by the OSC, one by the club. Should I get to the bottom of that, I'll make edits / footnote accordingly. Oldelpaso (talk) 20:57, 6 July 2015 (UTC)


 * Interesting. 2003 as the transition year makes sense. The transition may explain why neither the club (MCFC OWS) nor the OSC has deigned to publish the complete list of award winners to date on either of their servers. Meaning, neither of them owns the complete data. OSC only lays claim to the awards given up to 2003 (or some other year of transition) so could only truthfully print the list up to that point, while MCFC only lays claim to the awards made after 2003, and could only print a list starting at that point. Rather than either web site printing a partial list and drawing attention to the issue, they may simply prefer not to go there. However, they are clearly working together on the heavy promotion of this year's award (and did so on last year's award too) both of which have become massive web voting enterprises that last all summer long. Usually, the award was all done and dusted by late May or early June when the club had its end of season gala and handed out all the other awards such as Young Player of the Year and Players' Player of the Year, etc.
 * Go take a look at this version of the article before I found the Happy to be Blue! web site and had all the awards but three since 1995 individually sourced. Look at the "Main table references" section at the end. Tevez received his award on 7 May 2010, Kompany received his award around the 25 May 2011, Agüero on 3 May 2012, Zabaleta on 4 June 2013 (looks like I entered the wrong year for that citation), and then I had a real problem finding any sort of citable source from anyone for Yaya's award last season, despite the fact that it was similarly hyped all summer long too. In the end I had to cite a passing reference to it in the later announcement of his CAF Player of the Year Award. I thought that was curious at the time. Being the most recent, and Yaya having such a large worldwide fan base, I thought I would find no end of hits for that award, with plenty of them outside the usual M.E.N. and MCFC OWS sources. But nothing at all from those last two. Also take a look at reference "6" for Distin from Goal.com. It's only a passing reference but it is one more vote for Distin having won PotY in 2003 rather than Berkovic.


 * BTW, I'll retract what I said about the MCIVTA web site. In researching something for my updates yesterday I had cause to spend some time reading some of its newsletters from the early 90s which I thought were pretty good. So it's not totally crapola. But that's the nature of these fanzine web sites, parts of them are factually very accurate and objective - or at least, if not totally neutral, not too unreasonably partisan given the tribal nature of their subject matter - while other parts are as abysmal as that DOS era page you sourced the playing positions from. In the past, I've sourced stuff from the Bluemoon history pages - because they claim affiliation with Gary James - but I wouldn't source anything from its forum, which is the main reason for the web site's continued existence. The same applies to tabloid news sources such as the Daily Mail, Daily Mirror or the Sun - some articles in isolation can be O.K., but the body of articles taken as a whole are abysmally jaundiced and unreliable, so I just avoid them as a possible source of useful citations. 66.16.144.18 (talk) 04:20, 9 July 2015 (UTC)


 * One minor aside, I think MCIVTA was the very first MCFC specific web content. Not so much use as a Wikipedia source, but an interesting "social history" document for its 1990s heyday. The editions of its newsletter from just after Wembley 99 are still in my email archives. I used to regularly go back and read them when I needed a smile putting on my face. Oldelpaso (talk) 18:24, 12 July 2015 (UTC)

List of players table restructuring
There is currently a problem concerning the main table. The sourcing info. was presumably first written to address the original data sourced from James circa 2006. As players post 2006 achieved >100 appearances and were added to the list they had to be sourced from Soccerbase. But Soccerbase only tracks domestic league and cup games, Community Shield games, UEFA club competition games, and UEFA and FIFA national team games (both competitive and friendly). It doesn't track any of the closed season "pseudo" competition games such as the Audi Cup, Joan Gamper Trophy, Barclays Asia Trophy, Dublin Super Cup, World Football Challenge, etc. Likewise, neither does the BBC and most of the other sporting media write match reports for such closed season friendlies, no matter how grandiose their titles. These games are the modern day equivalent of past closed season competitions such as the Anglo-Italian Cup, Texaco Cup, Anglo-Scottish Cup, and the one I really hate, the Intertoto Cup which was universally ignored until UEFA took it over and made it a cheap backdoor entry into the UEFA Cup for teams such as Newcastle, West Ham and Aston Villa who would never have qualified otherwise.

According to the sourcing information currently in the article, all the stats. in the main table include such games, because the only games that are explicitly identified as being excluded are war time related fixtures - which probably have more merit for being considered real fixtures than Texaco Cup or Watney Cup games! I'm assuming this sourcing info. applies to the data sourced from James because it cannot possibly apply to Soccerbase related data - first, because Soccerbase wasn't in existence back then; and second, because even if it had existed, it would have ignored such games the same way it ignores the modern day equivalents such as the Barclays Asia Trophy or the World Football Challenge. So what the main table is currently listing (and comparing when you use the various column sort options) is apples and oranges. For the data to make any real sense (and to be either all apples or all oranges) the James-sourced data and the Soccerbase data should only tally the stats. from the same mix of identifiable competitions. Since neither you nor I have any control over what Soccerbase chooses to tally, this means that the only option open to us as editors is to adjust the numbers quoted from James to only include the same fixtures that Soccerbase covers. Those James sourced numbers really should not have included any of the closed season fixtures anyway because virtually no other reputable source, such as the BBC, include them in their stats. records.

However, since I (like 99.99% of other Wikipedia editors) don't own a copy of James, only you can make those adjustments to bring the James sourced stats. into alignment with the Soccerbase sourced numbers. 66.16.144.18 (talk) 12:27, 6 July 2015 (UTC)


 * What is sourced to James' Complete Record in the table and what is not? In short, the appearance / goal figures are sourced to it. Nationalities and positions are not. Birthplaces are listed in the book, but of course that does not equate to nationality in cases like Terry Phelan. At no stage has the book been used to reference nationality. If your former user account is the one I think it is, we previously had a dialogue on sourcing for the pre-2006 players. The totals were all checked using both James and Baskombe. The types of matches included are those counted as official by the Association of Football Statisticians. We could possibly separate out the competitions. Due to the scale of changes required this would probably need to be done in a sandbox, so that it wouldn't all have to be done in a single sitting. Oldelpaso (talk) 20:57, 6 July 2015 (UTC)


 * I saw what had been done in the "25-99" article and my immediate reaction was, "Why are these two very similar topic articles so different?" I just assumed they had been created by different editors. Nevertheless, the inconsistency between two list articles that should be identically structured bothered me. I felt that the stats. sourcing and table column definition info. were presented much better in the other article and decided it was something I could fairly easily bring over to this article to improve it, while also making the two articles more consistent. Thereby making the migration of players from the other article to this one easier for other editors to handle. It bothers me that information actually gets lost (e.g., international caps, number of league appearances and goals, etc.) the more significant - in terms of number of appearances - a current player becomes at the club, causing him to migrate from the top of the other article to this one. Giving more detail to subject matter that is relatively less notable and significant than other comparable subject matter is bizarrely topsy-turvy, and I'm quite certain it runs contrary to Wikipedia guidance somewhere, though not being a Wikilawyer myself, I couldn't quote you chapter and verse. "The more notable a player is the less information we are going to tell you about him" effect resulting from the way the two articles currently interact most certainly appears to violate the standard structure and spirit of almost all reference material (not just encyclopedias), not to mention running contrary to plain common sense.
 * Regarding your "the types of matches included are those counted as official by the Association of Football Statisticians" statement, that may well be. I wasn't questioning your original decision of what to include in the totals; I'm quite sure back in 2006 when the table was first created and was filled with data that was 100% sourced on James, your choice of what to include and what not to include was the best possible one. But that was then and this is now. There are now over 30 players in the table sourced from Soccerbase who no longer conform to the inclusion criteria (i.e., which games are included and which are excluded) currently stated in the section prior to the list table, and each season going forward that number is only going to increase in quantity (at the rate of about three players per season) and as a relative percentage of the whole. If we change the prior section inclusion criteria to properly define the Soccerbase entries, then all the entries currently sourced from James will become equally wrong.
 * This situation permits only three possible solutions. (1) We split all the Soccerbase sourced entries off into a separate but identically structured table that has new inclusion criteria written for it that matches the actual games for which Soccerbase captures data. This table will continue to grow as current players make more than one hundred appearances and migrate over from the "25-99" article. The other table will remain frozen in size and contain only the oldest players that are sourced on James. (2) We modify the inclusion criteria for the table to match the inclusion criteria that Soccerbase is currently using and then manually adjust all the "Appearances" and "Goals" tallies for the entries sourced off of James so that they are consistent with this new inclusion criteria. This is the option I was suggesting to you in my previous posting. (3) We leave things pretty much as they are but we make the "Appearances" and "Goals" columns unsortable (since it is highly misleading to allow the user to sort apples against oranges as if it had any meaning. The "Refs" column should also be made unsortable as sorting arbitrarily named reference groupings makes no sense either - once again the "25-99" article handles that aspect correctly while the main list article doesn't. All the other sortable columns can remain so. Additionally, the sourcing definition text in the previous section will need to be amended to state that the table contains two sorts of player statistics: players with their stats. sourced from James for which the match inclusion criteria is the one currently stated, and players with their stats. sourced from Soccerbase for which the alternative match inclusion criteria will have to be added.
 * My own personal view is that option (2) is the preferable one but it puts an onus on you for a large portion of the conversion work. The advantage of option (1) is that it can be used as an interim step to achieving option (2) at a leisurely and undaunting pace and does away with any need for sandboxes, etc. Let's call the James sourced table "Table A" and the Soccerbase sourced table "Table B". Entries in Table A can be modified to reflect the match inclusion criteria of Table B and then moved between the two tables accordingly on as leisurely a schedule as can be handled. Eventually Table A will shrink down to nothing leaving only the fully integrated Table B that is the immediate goal of option (2). No need for sandboxes and massive workload commitments; at any juncture the two tables will be internally consistent and valid according to their respective stated match inclusion criteria. The migration progress would be no different than the current one that occurs between the "25-99 appearances" list table and the ">100 appearances" table - although the goal would be to migrate Table A into Table B at a faster rate than 2-3 players per year! If option (1) is embarked upon with the agreed goal, once the two separate tables are created, of achieving a single integrated Table B over a more manageable time frame than the more immediate and work intensive time frame of option (2) then, of course, my own preference would be option (1).
 * There would also be no need to lose any of the current data. As you suggest a good way to resolving the differences and anomalies between the two sources of stats. data would be to break out the individual competitions into new columns. I would envisage the final version of Table B to be able to display appearances and goals for the league games; ditto for the domestic cup games; ditto for all other "official competitive games" (which equates to all Community Shield, Champions League and Europa League fixtures and their natural predecessors such as the Charity Shield, European Cup, UEFA Cup, and UEFA Cup Winners' Cup); ditto for the totals of those three columns; and ditto for all the other (mostly closed season "friendly" competition) games not covered by those other columns. So I'm not suggesting that you expunge all mention of appearances and goals in competitions that are approved by the Association of Football Statisticians, only that you re-categorise them so that they don't muddy the totals data that most folk universally recognise and which is also the same totals number you get when you add together the corresponding tallies in the four Soccerbase columns. International appearances and goals should also be separately shown.
 * That's an awful lot of new columns needed and they certainly can't be introduced while you keep trying to shove photo images down the side of a table that needs to grow laterally! Which is why I've moved them. There is plenty of "white space" available for photo images down the side of the "Hall of Fame" tables which have no requirement to grow any wider, although they will eventually grow longer when more players are inducted, which in turn will create more white space and opportunities for adding further images in the future. And photo images can always be added to the text areas as the text grows.
 * I have already created more new column space by shrinking the width of the "Player" name column and by replacing all instances of flags and countries with just the flags. The vast majority of people know which flags belong to which nations. If they don't, and for instance they have no idea what Shaun Goater's nationality is, they only have to run their cursor over the flag and it will tell them Bermuda. If and when we reach the point of adding an international appearances column (just like you have in the "25-99" table) the "Nat." column can be deleted and the "flagicon" info. preserved and copied over to it only for those players that won any international caps.
 * Other areas of gaining width for new columns when required is to reduce the "Position" column data to just the GK, DF, MF, and FW acronyms (as you already have done in the "25-99" table), and to change the reference grouping tags "PotY" and "stat" to simply "P" and "S" in the same manner as "A" and "G" are currently used for the footnotes. No need for mammoth editing sessions and sandboxes, etc. to change all this stuff if we take a manageable phased approach to doing it. In fact, the new columns only need to be added to Table B and part of the migration process of moving a player's modified stats. figures row from Table A would be to add the additional column information it requires in Table B. 66.16.144.18 (talk) 04:20, 9 July 2015 (UTC)


 * Do you agree with me that option (1) is the best route to go - that is, initially splitting the table into two tables with the goal of consolidating them again by modifying then migrating the entries in the James sourced table (Table A) into the Soccerbase sourced table (Table B) in a more leisurely and less work intensive manner versus trying to immediately fix and make consistent everything in the current table - option (2). Or did you prefer option (3)? 66.16.144.18 (talk) 23:33, 12 July 2015 (UTC)


 * As mentioned before, if I was doing everything solo, I'd make the changes in a user subpage and port them over when done – I'm wary of giving the 150 or so daily viewers a work in progress to look at. Well, Wikipedia is by definition a work in progress, but you know what I mean. But who knows how long I'd take for that. I'm not editing much now, in the close season, never mind when I spending most of my weekends watching the names at the bottom of the current list. To be clear, is your proposed format as now but apps / goals becoming "league apps, league goals, domestic cup apps, cup goals, other apps, other goals, total apps, total goals"? With appropriate abbreviations, naturally. Given space concerns, do we need the player of the year column if that is now listed in another table? Oldelpaso (talk) 19:36, 13 July 2015 (UTC)


 * I would consider the decision to reduce or remove anything that is currently in the table that I've already reduced as a bridge to be crossed when we come to it. I quite like the "PotY" column but if it has to go to make room for a column that we consider more essential then so be it. Ditto the other suggested space-savers I mentioned above - I prefer "stat" to "S" and "centre half" to "DF" but not if either of those prevents a more essential column being added if and when the time comes to do so. For right now we have plenty of room to add new columns so let's defer such decisions to later.
 * The table migration I'm suggesting is much more easily achieved if we just concentrate on doing one incremental change at a time. When you try and take care of a few different issues in the same edit, not only does the edit potentially become a chore, but you also significantly increase your chances of overlooking one or other of the multiple issues here or there as you process, say, each new player row, thus introducing errors. The edit I did to change all the player nationalities to just the flag icons was pretty mind-numbing given the length of the table, but because I was focused on just that one repeated change in every row I did not make any errors, which I probably would have done if I tried to that and at the same time catch and remove all the post-flagging asterisks and maybe also change any player positions that I felt were still incorrect, and so on. The more diverse things you try and do in a single edit the more chance you have of screwing things up. It's also much easier for other editors to understand (and more importantly, accept) what you are changing if you do your edits in a systematic and iterative fashion.
 * Of course, you already know that, which is why you are suggesting first doing these changes in a sandbox since they will require the addition of multiple new columns and the changing of all the existing numbers. That's sounds like a hell of a lot to do in a single edit, or even a few back-to-back edits, plus a lot of the number changes may require much number crunching and rechecking behind the scenes in order to simply modify a single tally, which to the casual observer might appear as though it only required a ten second edit. But that is because you are looking at what you envisage the finished product will be like and assessing that it will require many edit changes to get there from here ... and, of course, it will. However, to give you an analogy, if I compare the current version of the article to what it looked like before I made my first edit on 25 June to simply fix the Nasri/Agüero photo, and if back then my vision of what I wanted the article to ultimately look like was the current version, I probably wouldn't have made my first edit because it would have seemed like a humungous task to get from there to here and I wouldn't have wanted to embark on such an undertaking lest I broke the article and left it in an "under construction" state for a large portion of the time that it took for me to transform it. But I didn't have any such vision to deter me from making that edit; every change that I've done since has simply been a small iterative improvement on the previous one that only occurred to me after I viewed what I had already done. Sometimes my next edit was a "ce" edit to fix something I had broken, and in other cases it was done to make things consistent after having changed something significantly - such as putting the Positions tactical evolution table back in the "25-99" article where I had originally pinched your similar much smaller table from in the first place. At no point in the changes I've made to the article between my first edit and now have I left it in a "under construction" state. So don't focus too intently on exactly how many columns the final table will have, but rather on the systematic improvement process that I outline below in order to get us there. As we get nearer to achieving the final goal, we may have to adjust that goal based on lessons learned along the way.
 * The Soccerbase records we source from have four pairs of column totals (apps and goals) for each of league games (any division), FA Cup games, League Cup games and "Other games". I don't believe they define what "other" covers - at least I've not discovered any definition on the Soccerbase web site of what it is meant to cover - but a process of removing from their total records everything included in the other three suggests it is all UEFA club games plus any Community Shield fixtures (if you know of any other type please let me know). Clearly, the numbers in our table would become a lot more transparent to both users and other editors of Wikipedia alike if our table contained the same four pairs of columns that make up the current totals pair that we do have. That is five column pairs where we currently have only one pair and we are still not done yet, so I think we can rationalize things a bit and combine the FA Cup and League Cup column pairs into a single domestic cup pairing, just as you suggest, without much loss of transparency, and regain a column pair. The same data breakdown of the totals can be similarly done for the James book sourced data (I'm assuming; it most certainly can be done from the Bluemoon web site data presented at the bottom of each player's "Player details" page).
 * Breaking out the league and domestic cup data into two pairs of new columns merely adds transparency to the numbers. The disconnect between the two sets of data comes in the "other" column, where "other" for the Soccerbase totals is what I defined it to be just above, while "other" for the James book data includes all of that and then some, with it adding in the Anglo-Italian Cup and all the other such competitions listed in the preceding "Key" section. What I am suggesting we have is two pairs of "other" columns, one to the left of the "Totals" column pair and one to the right of it (think of that column pair as the one we regained by merging the the two cups into a single domestic cup column pair). We would have to give them better names than just "Other 1" and "Other 2" but that's mere detail for later. The "Other 1" column pair to the left will only contain (for the James book sourced players) the tallies for the same mix of games as covered by the Soccerbase "other" totals and, of course, that is the column those "other" numbers are entered in for the Soccerbase sourced players.
 * The "Other 2" column is inapplicable to the Soccerbase sourced players and will be blank for all of them; it only applies to the James book sourced players, and it will only have an entry for any one of them if that particular player appeared in a competition other than a UEFA club game or the Community / Charity Shield competition (a.k.a. the English Super Cup *rolls eyes*). As you said earlier, this applies mostly to the players who played in the 1970s when all those competitions were being experimented with. So to take a hypothetical example, a player such as Tommy Booth - who played three games in the closed season Anglo-Scottish Cup competition (and possibly even scored a goal in it!) - will have 8 not 11 entered into the "Other 1" column, his total appearances tally will be reduced by 3 from 491 to 488, and his entry in the "Other 2" column will be 3 (representing that he played in three more games that season than we are including in our "official total" tally for him). This way, none of the games you have included in your current totals are lost, they are merely redistributed to the "other 2" column, while the "Other 1" column tallies will be normalised to be consistent with the Soccerbase "other" column tallies, and likewise, the tallies in the "Totals" column pair will be similarly normalised to account for exactly the same mix of competitions regardless of whether the player's stats are sourced from James or Soccerbase, thus allowing apples to apples comparisons.
 * I think we could add in the league and domestic cup column pairs quite quickly to the current table. In fact, now that I know about the Bluemoon statistics base, I could do all those columns myself fairly quickly using the data on the Bluemoon web site for all the James sourced players. You could then go through those same players and adjust it, if necessary, to the book values should any numbers differ. Spotting those discrepancies will also immediately identify which numbers need footnotes - if you have to adjust one the Bluemoon sourced numbers that I've entered then you know it will also need an explanatory footnote added. So that method of dividing the work will be quite productive, I think. But you could take as long to do that verification and footnoting process as you want, because at this point the numbers in the new league and cup columns don't have to be perfectly correct anyway. By that I mean that no user reading the article while we are scrubbing the numbers in this fashion is going to notice that player A is shown as appearing in 43 domestic cup games when he has really only appeared in 42 or 44, or that player B is still lacking a footnote (because you haven't got to him yet). If a user reading the table knows that sort of detail off the top of his head then he probably has no need to read the article in the first place! The table is potentially rife with such tally anomalies right now as the process we are going through with Book, Booth, Power, et al is demonstrating, so I don't consider that level of ongoing flux to be anything different than the table for the last nine years or so listing 50% more centre halves than have ever played for the club!
 * However, the situation starts to get trickier when we get to the point of adding in the "other" columns. These account for the areas where all the discrepancies between the two sets of source data lie. Both of the "other" column pairs will have to be added at the same time - you can't have one without the other (pun intended) - and there will now be a requirement that the numbers in the three column pairs to the left of the "Totals" column pair actually sum to the same amount that is shown in that "Totals" column, because such a discrepancy would be immediately obvious to a Wikipedia user with even the most basic numeracy skills, plus some Wikipedia editors might also notice such a disparity! This is the point at which the table may need to be split into two - Tables A and B as I have previously referred to them.
 * If you work your way up from Clichy at the bottom of the current list as far as David James, all those players are sourced from Soccerbase, while Sibierski is still sourced from James. So if I was splitting the table tomorrow that's where the split would be; Sibierski and all the players above him would be in Table A (primarily sourced off of James but also containing some Soccerbase sourced players intermittently spread out near the bottom of it) while James and all the players below him would be in Table B (100% sourced off of Soccerbase). However, we don't have to split the table tomorrow, and by the time we do I or you will have had plenty of time to investigate, for example, Sibierski's stats record in Soccerbase to see if it matches his current numbers in the table (sourced from James) and if they are the same then Sibierski can be sourced from Soccerbase instead. I have already done that for a few players but I can't remember exactly which ones now - but it's documented in my edit summaries if you care to search for it. Kinkladze is one I remember, and to be honest I was quite surprised to find a player from his era having his City stats complete; the comparable records for Rösler, who played alongside him for the three seasons 1995-98, are a mess in comparison. It was mostly because Kinkladze could be sourced from Soccerbase that I spent so much time messing with Rösler's records at the weekend in the hope that he could also be sourced from there.
 * However, the next two players down from Kinkladze, Kit Symons and Michael Brown, cannot be sourced from Soccerbase, while the one after them, Paul Dickov, can be and is. So there is a grey area between Symons and Sibierski where some players can be sourced from Soccerbase while the others cannot, although it might be possible to source more than are done so currently. Anyway, all the players from the grey area that can be sourced from Soccerbase can be immediately migrated from the initial Table A to the initial Table B (but we would have to remember their relative position in the table which is determined by their debut dates for the first team, a piece of information that we don't actually keep anywhere). The "Other 2" column entries for the Table B players are a doddle as they are all, by definition, blank. The "Other 1" column entries for the Table B players comes straight out of the "Other" column in the corresponding Soccerbase record for those players and can be populated almost as quickly because there will only be about 30 odd rows in Table B at this point. So the initial Table B can very quickly become the final version Table B that we seek, with all five column pairs populated and properly totaled and every row individually sourced to Soccerbase.
 * The real effort in this phase will all fall on the Table A players and, as they would all be sourced from James, only you could do the necessary adjustments to the numbers so that the two "Other" columns are properly populated. Except the Table A entries won't have any "Other" columns; those were only created for Table B once we did the split. Once you have calculated what the numbers should be for those two column pairs, you migrate that player from Table A to Table B so that you can enter those numbers into the two "Other" columns that the player record is required to have in that table. This is no different than the process that a player record in the "25-99" article table, such as the entry for Jesus Navas, has to go through in order to be transferred to the main article list once he has played his 100th game, because the two article tables don't have exactly the same structure. Although Table A will contain only three column pairs and Table B will have all five column pairs, at any point the two tables will be internally consistent with what they are defined to contain, in the same way that the two article tables are internally consistent despite the fact the player records are structured differently in each article. At no point will this migration to the final Table B format require the article to be left in an "under construction" state. It will simply contain two tables structured slightly differently, one (Table B) slowly expanding over time while the other (Table A) is slowly shrinking.
 * The rate at which these two tables expanded and shrank would be solely dependent on how quickly you were able to convert the current James numbers to the "Other 1" / "Other 2" column format requirements. I could probably help in this process using the Bluemoon statistics base as a surrogate for actually owning the book. We would probably be checking each other's edits anyway, so in the cases where Bluemoon and the book differed you would probably catch and adjust something that needed correcting. When I first suggested this process to you a few days ago I didn't know about the pertinent statistics records on the Bluemoon web site so I assumed that only you could do the heavy lifting for this portion of the table migration. The fact that I could also participate in this phase only just occurred to me as I wrote this. There are approximately 200 players in the main table list by my calculations, and Table B would start off with 31 players if we did the split tomorrow. The other 170 or so players would all have to be touched and processed (i.e., migrated). That should give you a feel for the scope of the undertaking if you are up for it.
 * I don't view the decision to embark on the foregoing as an all-or-nothing decision. Each one of the iterative changes I described above improves the article anyway. Adding the league and cup columns to the current table is an improvement whether or not we decide to push on and tackle the normalisation of the "other" stats. Splitting the table in two is a relative trivial edit and it achieves the option (1) goal of separating out the two types of sourced data into their own specific tables so that which appearances and goals are included for the players in either table can be more clearly indicated in each case. Adding in the "Other 1" column pair for Table B is also a doddle since it can be populated directly from the "Other" column pair in Soccerbase. Adding in the "Other 2" column pair to Table B is even more of a doddle as the entries in it are by definition blank for the current Soccerbase sourced players. In fact, it doesn't need to be, and shouldn't be, added until we are at the point of migrating the first "1970s era" player who has some data to go in it. Again, terminating the process before that juncture still leaves the article in an improved state.
 * Having two separate tables with slightly different structures in the "List of players" section is no more arbitrary nor any different than the past decisions made to split the complete list of players into three separate articles, with the tables in each article structured slightly differently. Migrating players between the two tables is also no different than the process that currently occurs for the current players as they reach and cross the 25 and 100 appearances thresholds. The fact that you won't be able to sort the players in Table B in amongst the players of Table A as you can now is no different than not being able to sort the <100 appearances players in amongst the >100 players in the current article divide. Except the Table A / Table B divide is only intended to be an interim divide with Table A eventually shrinking away to nothing leaving only a fully normalised Table B.
 * The interim period is of an unspecified duration so that editors - initially just you (since you appear to be the only editor with James' book), but now that we know about the Bluemoon stats I can probably help too - would be under no pressure to complete a rather tedious and complicated normalisation process within some arbitrarily short timeframe. Should either of us run out of time, energy or motivation to actually complete the whole migration process, at whatever point we run out of steam the article will still be left in a consistent and improved state, with all the players more correctly grouped together, and each player entry in either table containing more stats info. that is better verified and possibly better sourced (from a web based perspective) than those entries currently contain today. The actual divide between the two tables will be one based on normalised versus yet-to-be-normalised "other" stats data, but as far as the user reading the article would be concerned, the divide between the two tables would be a purely chronological one, with all players who made their first team debut before a certain year listed in Table A and all those who made their debut after that year listed in Table B. The year of that divide would, of course, be gradually changing, but not in any manner that would affect any of the current 150-250 daily readers of the article in any detrimental manner. I don't see any need at all to add one of those "work in progress" or "article under construction" banners at the top of the article while the table migration process was occurring. The editorial impact on the article to move one or a few players from Table A to Table B in a single edit update would be less intrusive from a reader's perspective than many of the edits that I've done over the last couple of weeks. 66.16.144.18 (talk) 05:47, 14 July 2015 (UTC)

I'm going to knock up a few example formats just to see what they look like in terms of width etc. Haven't got time to do all the experimenting now, but will endeavour to do so over the next couple of days. Oldelpaso (talk) 17:17, 14 July 2015 (UTC)


 * Things to consider when you do. I view the current "Apps" and "Goals" columns to be the precursor of the "Totals" column pair in Tables A and B. When additional column pairs are added, particularly the "Other 2" column pair to the right hand side of it, I think the "Totals" column pair should be made to stand out from the subordinate column pairs either side of it in some manner, either by putting its entries in bold font or giving it a different background colour. Normal practice would suggest a darker grey. We might give those two columns a sky blue background like I gave the current players but that will probably look too funky. ETA: The more I think about this issue, bold font is the way to go IMO. Using small font in the subsidiary columns is also something to be considered if necessary. The reason I prefer the bold font solution is that it also makes the tally number stand out clearer from its footnote superscript. Additionally, even two grey columns running down the middle of the tables will still look a bit too quirky IMO.
 * More importantly, the current width of these two columns cannot be shrunk any further than they currently are because of the necessity to add the "A" and "G" footnotes to them. Without the footnotes they could indeed be made a little thinner by reducing the length of their already short heading titles. The new subordinate "Apps" and "Goals" column pairs for the league, domestic cups and "Other 1" games can, however, be made to be much thinner if we give them a single "G" or "A" title heading and restrict footnoting of appearances and goals to only the current "Totals" column pair. This means that we will need to add an additional line to the top of the current table header to identify "League", "Cups", "Other", "Totals" and "Addit." for the various column pairs only identified by "A" and "G" on the second line. The full "Apps" and "Goals" headings probably could remain for the "Totals" pair due to the footnotes requirement, which would also serve to clue the reader in as to what all the other "G"s and "A"s stood for too. ETA: Not only will the current "Totals" column pair not get any thinner, these two columns are actually going to get a little wider due to the fact that after the ninth footnote the footnote superscripts become 2 digits long instead of being just the single digit we have in the current table. Even if we didn't restrict all "Apps" and "Goals" footnoting to just these two columns they were bound to do that anyway. There is actually a very high likelihood that we may require more than 99 superscripted footnotes in one or both of those columns! Anyway, that's something else to consider.
 * Therefore, if a player's appearance or goal scoring record in a league, cup or UEFA club game needs footnoting it only gets footnoted in the appropriate "Totals" column, NOT in the subordinate column where perhaps the footnote might be considered to be more pertinent. If we restrict footnoting in this manner, and keep the column headings minimalistic, the new subordinate columns to the left of the current "Totals" column will all be thinner than the current pair. The new "Other 2" column pair - let's call it "Additional" for now since it will document additional appearances and goals scored not included in the "Totals" column pair to the left of it - will probably also have to permit footnotes. At least the "Apps" one will. That way, if you take a player such as Frank Swift, his "Apps" figure in the "Totals" column pair will be 375, but we could also include his three 1939 expunged league appearances in the "Additional Apps" column appropriately footnoted. That would simply mean moving the footnote from the "375" tally in the current "Totals" column pair and applying instead to the new "3" appearances tally in the "Additional" column pair (as it would be redundant to footnote the same discrepancy in both side-by-side columns). Thus the new "Additional" column pair possibly allows us to more elegantly show and document the wartime appearances and goals of the players who played in those 3 expunged 1939 league games which continually show up on other sources for those players' stats - including, most recently, the Bluemoon site - and which are the most common cause of the discrepancies that arise for the stats that are currently shown in the table for those WW2 era players. ETA: I've not investigated this, but similar common discrepancy issues may also apply to the WW1 era players that the new "Additional" column pair may also help to more elegantly document. The column has to be there to handle the common "1970s era" discrepancies so we might as well make use of it anywhere we can to handle other commonly recurring disconnects in the numbers too.
 * If you stick with those basic rules and stuff maximum possible tallies in the various columns - such as 99 goals and 999 appearances - I think you will find that we can accommodate all four new column pairs required by a fully fleshed out Table B without us having to make any further compromises with what is currently showing in the current table. However, we still always have the option of reducing player positions to two letter acronyms and losing flag icons, etc. should that not be the case. I already know without doing any sandboxing that a fully fleshed out Table A - with four less new columns - will comfortably fit in the "white space" to the right of the current table. Table B is the only issue IMO. 66.16.144.18 (talk) 19:51, 14 July 2015 (UTC)


 * A few examples, in no way intended to be exhaustive, are at User:Oldelpaso/sandbox. The double line header used to separate the appearance and goal types means it is generally the length of the longest word in a title that determines the width, as titles are free to take up two lines. The data is of course nonsense, I just used Shaun Wright-Phillips as a base due to his long name and gave the rest long but feasible values. Feel free to add any further modifications you had in mind. Oldelpaso (talk) 18:01, 16 July 2015 (UTC)


 * Example of envisaged Table B format


 * Example of envisaged Table A format - with footnoted columns left-aligned

Player lists - data resolution
I have been correcting these dates according to the new definition of the column that I added as I've been adding in the supporting RS citations from M.E.N. because, once corrected, the M.E,N. articles also supported the dates as well as the total appearances and goals tallies. In some cases, your dates were correct IMO, while the M.E.N. dates were incorrect, in which case I suppressed them from my quoted text in the cited reference since, in those particular instances, the M.E.N. article only supported the stats. but not the career dates we are currently listing. At present, only the two dozen or so entries that I've sourced from M.E.N. are the only players whose dates I've spent any time verifying and you might wish to double-check the current career dates for those of the two dozen plus entries I've modified against James and let me know if you find any differences granted the new definition of the column. Even if James is not used as the primary sourced reference for a table entry - and he isn't in the case of those 24+ players who are now sourced from the M.E.N. articles - he should still be used as a ghost corroborative reference source for such entries. It is not my intent to clobber the James data, only to source it from the web where other editors can use it to revert vandalism and POV changes to what is currently there. You may also wish to check and adjust against James the career span dates of all the other players (besides those 24+) according to the new definition of that column. 66.16.144.18 (talk) 04:20, 9 July 2015 (UTC)


 * Let me now discuss those M.E.N. cited RS. All 50 players in James Scott's "top 50" are in the table - I would have had a big issue with his whole selection process if any had not have been! - but twelve ten of them are already sourced from Soccerbase so we don't need another web source for those players. Five of those players (Kompany, Silva, Hart, Agüero, Touré) have to be sourced to Soccerbase because they are current players and they have played more games since Scott's 2013 closed season articles, rendering them old news, while two others (Tevez and Barry) had technically played their last games for the club when he wrote his series of articles - although Scott may not have known that at the time - but I would prefer them sourced from Soccerbase because I trust it more than Scott's own attention to details (he's OK as a secondary source for what is ultimately James' data, but there is a big difference between copying other people's stuff and being an expert in gathering the original data, although Barry and Tevez hardly require much research). In addition to those seven players, three others currently have their stats. sourced to Soccerbase - Dunne, SWP and Goater - with the last having been sourced to there by me, prior to my finding the M.E.N. articles, as part of my earliest efforts to source more of the data from the web rather than an inaccessible book.
 * Dunne's stats. can be consistently sourced from either Scott or Soccerbase; in fact, sourcing him from Scott does away with his current goals tally footnote, and I'm considering sourcing him from M.E.N. instead, but since I've processed his stats. and already added the footnote I'm leaving him be for right now. Goater however cannot be sourced from his M.E.N. article as it only states "189 games" instead of "212 appearances" (189 being the tally of his times in a starting line-up, the other missing 23 appearances being the tally of his substitute appearances). Likewise, SWP cannot be sourced from his M.E.N. article for reasons I address below.
 * Of the other 38 40 articles, seven are internally inconsistent with the Bluemoon data sources they are derived from so I cannot use those M.E.N. articles even if I wanted to (which I did until I discovered that fact - see below). That leaves only 33 articles that can be cited as RS in support of individual player data from the whole batch of fifty, and all of those 33 citations have now been added. 20 21 of the ones I've used completely matched statistics-wise (see entries in the "Data resolution status table" below tagged with a "1", "2" or "3" explanatory footnote in the sortable "Notes" column) but with three of them differing on career dates once I had adjusted them to the new definition (these are tagged with a "3" footnote). Seven of those 20 (21-3=) 18 matched the M.E.N. dates as is (these are tagged with a "1" footnote), and 11 of the 20 18 I had to adjust due to the player's debut and/or final appearances for the club, or, as in the case of Dennis Tueart, it was just plain wrong (these are tagged with a "2" footnote). The three players whose career span dates I believe were/are still correct, but Scott's article has the dates wrong, are Paul Lake (for the reasons mentioned above), Dave Ewing and Peter Barnes. Of the other (33-21=) twelve M.E.N. articles cited, eight supported the existing or adjusted table data only after footnotes were added - Swift, Brook, Herd, Doherty (these are tagged with a "4" footnote), Book, Booth (these are tagged with a "6" footnote), Power, Rösler (these are tagged with a "5" footnote) - and four (Barnes, Young, Oakes and Quinn) will require processing your end in the same manner as Book, Booth, Power and Rösler did, so I've added new section headings for them below. The other 18 articles each differed slightly regarding the statistics for the player and I would like to resolve those with you. The error could lie with Scott's article (like the dates for Lake, Ewing and Barnes) but could also mean you are totaling up goals and/or appearances that most other sources do not recognise. 66.16.144.18 (talk) 20:49, 16 July 2015 (UTC)


 * Having now completed my processing of all fifty players covered by James Scott in his M.E.N. articles, I need to revisit and update this section yet again because it is still wrong, plus I wish to add some further information. I will use strike-through indications again to clearly indicate what has now changed and bold font for newly added text and then, once I know you've read it, I'll remove the bold font and old strike-through figures and text as if they had never happened. IMO this is a lot less confusing for either of us than writing the same thing three times over (I've already been through this process once), and most certainly a lot less confusing to any third parties that may later read this text. Nothing is conceptually affected by this update, just the actual numbers of how I categorise things. 66.16.144.18 (talk) 20:49, 16 July 2015 (UTC)


 * The Scott series of articles in the summer of 2013 is clearly derived from this portion of the Bluemoon web site. As the disclaimer on that page clearly states, this web site content is derived from an article James wrote for The Times sometime in 2009. This James article had to have been added to Bluemoon prior to June 2013 when Scott started using it, and of course after 2009, since it could not have been copied from an article in The Times before it had even been published there. The apology re the selection of the fifty players not including any of the more recent ones strongly suggests it was only added to Bluemoon near the end of that four year window (2009-2013); James or Turner would not have felt any need for such a caveat if it had been added there soon after appearing in The Times in 2009 before most of the seven players who we have to source from Soccerbase had even joined the club.
 * All that Scott did to create his article was replace seven players on James' original 2009 list (Cowan, Crossan, White, Owen, Law, Woosnam, Deyna) with Tevez, Barry, Hart, Kompany, Silva, Agüero and Touré. For those players he would have to go get his own appearances and goals stats. because he wouldn't be able to use James' book or 2009 article for those, and he doesn't even bother trying in the cases of Hart, Kompany and Silva, while the stats. he uses for Barry are at least one season out of date (as of summer 2013) because they should be the same numbers back then as they are today since Barry's last game for the club was in the game against Reading in May 2013. And that is the reason I don't trust Scott over Soccerbase as a citable RS for either Tevez's or Barry's stats. All that Scott then did was reorder James' ranking of this new and improved list of fifty players, add some different photos in many cases, and perhaps write some new bio text, but much of his verbiage is derivative of James' original text too. I have compared all three pieces of data (career dates, appearances and goals) for the 43 players in common between the two lists and in all cases Scott used 100% James' data from his 2009 article as hosted on Bluemoon - even the errors in the dates for Ewing, Barnes and Lake come from that article.
 * Which now brings me to the seven players plus SWP (referenced above) whose three pieces of stats. data is internally inconsistent on the Bluemoon web site. If you click through the five pages of ten players each in James' 2009 article, and go to any of the pages at the next level showing the individual "player details" for that player, you'll find internal inconsistencies for those eight players between the total appearances and goals stats. (I'm not worried about career date discrepancies because they can simply be suppressed in the manner that I have already done so for Lake, Ewing, Barnes and Young) listed in their 2009 article headings - which is exactly the same values Scott uses in his M.E.N. articles and I also quote in the cited reference text - and the same data that appears at the bottom of their "player details" pages. So if, for example, we take Franny Lee, according to the 2009 article entry for him his essential stats. are 340 / 144, but when you click on his name in that entry and go to his "player details" page, the same stats. derived from the totals at the bottom of that page are 330 / 147. So which is correct? One, possibly both, has to be wrong. Neither pair agree with the same data currently in our table sourced from James' book, which is 330 / 148. It is most likely the one in the article entry heading that is wrong because it is "frozen" or "static" data from 2009, while the tallies at the bottom of the players' pages are derived from the team sheet data that makes up the core Bluemoon database.
 * For example, if Ric or Gary corrects that Honved game team sheet and replaces Booth by Book at RB, both Booth's and Book's appearances tallies in their individual "player details" pages will automatically adjust themselves too (just like column totals in a spreadsheet). So given that the data in the various "player details" pages is "dynamic" it is most likely reflective of the latest corrections Ric or Gary has done to the team sheet data in the years since that 2009 article got the numbers wrong. Consequently, since the M.E.N. articles for those eight players in question align themselves with the (most likely) erroneous "static" values from the 2009 article, it seems inappropriate to use them to source corrections to the data currently in the "List of players" table, even if they are wrong. Replacing an error (if that is the case) sourced on a book with a web sourced error doesn't get us very far, and if the current data is correct, then replacing it with a web sourced error would be even more stupid. Therefore, IMO those eight M.E.N. articles are of little use to us.
 * The eight players affected are Bell, Lee, Summerbee (but of course!), Meredith, Browell, Horace Barnes, Roy Paul and SWP. Whether we should cite the individual "player details" pages for these players is an issue we need to discuss. In the case of three of them - Barnes, Paul and Browell - these pages actually support the data currently in the table, but in the other five cases they simply muddy the water and present a third data view (to the M.E.N. articles and what's already in the table sourced from James' book). Plus we can't pick and choose. If we use the Bluemoon "player details" pages for some players (such as the three I suggested) then we really should also use it for all of the other players we cannot source from M.E.N. On a tiny sample of eight players, five (62.5%) being different from the current table entries doesn't bode very well for adopting that approach to the remaining individually unsourced players in the "List of players" table. Actually, the situation is somewhat worse than that. Two further players - Tilson and Ewing - also have disconnects between the stats. data shown in their entries in the 2009 article list and the totals at the bottom of their individual "player details" pages, but in their cases the data already in our table aligned with the former numbers (which is why I cited their respective M.E.N. articles in support of it) rather than the latter numbers (in the case of Barnes, Paul and Browell) or a three-way mismatch in the case of the other five players (Bell, Lee, Summerbee, Meredith and SWP). 66.16.144.18 (talk) 00:14, 17 July 2015 (UTC)


 * I've updated the above couple of paragraphs because the numbers that I originally included in them were mostly wrong. Of the 18 articles I mentioned I wished to resolve I have already gone ahead and updated five of them, so that number is now 13 (= 50 - 12 - 20 - 5). The table entries for first 20 players I sourced to M.E.N. differed only in regard to career date spans, if they differed at all (6 didn't), but the final 18 all of which require possible changes to the appearances and/or goals tallies (plus possibly career dates too). The five resolved so far are:
 * Power (apps-1)
 * Rosler (apps-1)
 * Swift (final year 1950->1949, added footnote A1)
 * Book (final year 1974->1973, apps-1)
 * Booth (debut year 1967->1968, apps+1, goals-1)
 * I'm going to continue on processing the other remaining 13 players by counting the appearances and goals from scratch (thus these edits require a lot more data crunching than just checking when the player made his first team debut and played his last game, which is mostly all the first 20 entry updates required). 66.16.144.18 (talk) 06:15, 12 July 2015 (UTC)


 * James figures:
 * Power as your edit summary except League 358 starts + 7 sub
 * Rosler as your edit summary except League Cup (10 starts 1 sub 5 goals) 3+1 (2) in 1994–95, 3 (2) in 1995–96, 2 (1) in 1996–97, 2 (0) in 1997–98. Soccerbase entirely different (and questionable).
 * Swift as your figure.
 * Book as your edit summary except Europe 17. All 9 games in the 1969–70 CWC. 7 out of 9 in 1970–71 CWC (he missed the 2nd leg and neutral ground replay against Gornik) 1 app in 1972–73 UEFA away to Valencia.
 * Booth as your edit summary except 1 "other" goal in Anglo-Scottish Cup vs Sheffield Utd. 25 Euro apps. All 9 games in the 1969–70 CWC. 5 in 1970–71 CWC (Linfield h, all 3 vs Gornik, Chelsea a), 11 out of 14 UEFA games between 1972 and 1977 (games missed Juventus h, and both legs vs Twente) Oldelpaso (talk) 19:07, 12 July 2015 (UTC)

OK, I assume we are both in agreement over my career dates changes and M.E.N. sourcing of the first 20 players plus the footnoted Swift, and we can now just focus on the remaining 17 players, the first four of which are the ones bulleted above (the others 13 still tba).

Uwe Rösler stats
The disconnect lies in the 1994–95 League Cup appearances (total goals was always fine). That season City played a 2-legged tie against Barnet in their first round (that would be Round 2, I believe) and Rösler did not play in either of those games. In the next round, City played and beat QPR 4-3 at home in which he also did not feature and no subs were used. In Round 4, City drew with Newcastle 1-1 at home; "Der Bomber" came on as a sub and scored the goal. In the replay at St. James' Park, City won 0-2. This time "Der Bomber" was in the starting line-up and he and Walsh scored the 2 goals. In the QFs Round, City lost 4-0 away at Crystal Palace, and that was the end of their League Cup run that season. "Der Bomber" was in the starting line-up for that game. That's 2 starts and a sub appearance plus 2 goals, not 3 starts and a sub appearance. That's your error unless I'm missing something. I will address the other 3 players as and when I can get to them. 66.16.144.18 (talk) 20:44, 12 July 2015 (UTC)


 * Complete Record has him down as playing in the first leg away to Barnet. Team: Coton, Edghill, I. Brightwell, McMahon, Hill, Vonk (Foster), Summerbee, Walsh, Rosler (Quinn), Lomas, Beagrie. After searching through some boxes I found a programme from the same season (Newcastle h, League Cup 4th round). That lists the same team, as does mcfcstats.com. Using my library card to log in to NewsBank, the reports in the Times and Independent from the following day also list the same team. Think this discrepancy is definitively solved. Oldelpaso (talk) 21:16, 12 July 2015 (UTC)


 * Ric's (whoever Ric is?) stats side of the Bluemoon web site that you just pointed me to (in the section above re Gillespie's 30 goal tally) says he definitely didn't play! LMAO.  You can't pick and choose your sources to suit whatever agenda you have. If Bluemoon is a good enough source to support Gillespie then it's a good enough source to support Rösler. What you are offering as an alternative version is just a bunch of "trust me" sources - it's in a book I once read; it's in some archived news articles that only I have access to; and it's in a matchday programme I own in my attic. Unfortunately, editors in Malaysia, Brazil, Australia, Canada and the U.S.A. don't have access to any of those. You can't talk out of both sides of your mouth and say Bluemoon is a valid accurate source over here (for Gillespie) but it is a badly unreliable source over there (for Rösler). You also can't champion the accuracy and validity of James' stats one moment (re his Complete Record book), then ten minutes later claim his stats are a bunch of Crapola because you now disagree with him (because, as I understand it from what you wrote above, Ric is working with James to source his stats on the Bluemoon web site, so effectively the contradiction here has James supporting both sides of the argument!).
 * OK, to be fair, you did also mention the MCFCstats web site. The problem there is that that particular site hasn't been updated in over a decade. That shouldn't matter for the statistics of games that occurred over two decades ago - after all, history doesn't change - but it's my understanding that the site is a frozen "work in progress" effort. The person who created and maintained that web site unfortunately died before he had completed his basic goal of adding all the various statistics associated with the club's past seasons. I can live with the funky DOS-era eccentricities of the site such as the outrageous choice of fonts and the way the pages navigate, but my main problem with it is that I have no idea what is complete and what was in a state of flux when he unfortunately passed away. His choice to include almost every game played in a season - including practice kickabouts in Wythenshawe Park, I think! - in the appearances and goals tallies unfortunately renders those totals mostly useless for our purposes. In almost all cases his stats for the players in the main table will differ from those of James due to his highly inclusivist approach. If you wish to start citing that site as a supporting RS in this particular instance then you open the door for his stats to be similarly used as the main supporting RS elsewhere in this article, which, to put it quite frankly, would be shooting yourself in the foot. His stats rarely agree with the comparable ones from James, but his are web citable while James' stats are not, so guess how that situation is going to play out if you start using MCFCstats to support your case only when it suits your agenda.
 * My advice to you is to stay well clear of MCFCstats if you wish this article to preserve most of the numbers currently sourced from James. We are not talking a few +1 or -1 tweaks to the stats as we are doing with the M.E.N. and Bluemoon sources, we are talking instead of much more radical disconnects. Here's just a very simple example of what I'm talking about. Go to the MCFCstats "Most appearances" page ... I can't give you the URL so that you can go straight there yourself because all pages on that legacy site have the same http://www.mcfcstats.com/ URL, so I'll have to navigate you there. From the "Home" page select "Players". On the next page select "Appearance Statistics". On the next page select "Most appearances". Finally, on the last page select "Overall". Now compare the top ten players in his top 20 (which not only contains Swift, Meredith and Bray instead of Summerbee, Power and Donachie, but also note the calendar years and appearances tallies!) with the Bluemoon's "Top 10 Overall Appearances" and the top ten appearances according to our article when the "Appearances" column is sorted. So tell me, would you prefer that we reconcile the table data to the MCFCstats view of the world, or to the M.E.N. view of the world (which I believe is sourced off of Bluemoon, but I was unaware of that until you gave me the Gillespie link)?
 * To return to the matter at hand, I have no idea what the "actual truth" is in the particular case of the Barnet tie because I wasn't at the game in question ... I wasn't even in the country when it was played. I assume you weren't at the game either. So we are both reliant on third party information here. I have two citable sources for my version of things - M.E.N. and Bluemoon (thank you, for that!). And you? You just have a bunch of "trust me" sources and a highly unreliable web source. For all I (and any other editors) know, you don't really own a matchday programme (or if you do, it doesn't say what you claim it does); you didn't really access Newsbank (you just said you did instead); and you have never possessed a copy of the Complete Record in your life - and when I asked you for goal scoring references you simply made up page numbers to a book you don't own! I'm not saying that you did. I wouldn't be wasting my time trying to work with you if I even suspected that was the case. But on Wikipedia citable sources trump "trust me" sources, hearsay, gossip, fantasy and, yes, even the truth! What really matters is citable "truthiness" NOT uncitable "truth". Surely you must know that by now. :)
 * So let's not waste any more time over this one. I suggest we go with the number I can cite two web sources to support (i.e., we leave things as they are now) and that you/we add an "A" footnote to Uwe's appearances tally regarding the discrepancy for the Barnet game, so that the reader sees both alternative versions. I suspect that this may be the first of many similar discrepancies that we are going to come across as I plough on with analysing the stats of the other 13 players, and I suggest that we need to apply this rule/priority to all of them when we come across them - i.e., the value that can be best supported with citable web sources is the value that is used for the tally in the table, and the alternative versions get footnoted. Just like we do for DGC reassigned goals and Community Shield match appearances that are fully documented by Soccerbase on a per season match basis but somehow get overlooked when the various totals are tallied by Soccerbase.
 * Regarding Soccerbase stats for "Der Bomber", the reason why he was the latest player whose stats I sourced yesterday is because I must have wasted a couple of hours on Saturday trying to make sense of the Soccerbase stats for him with the hope that he might be a 13th player that could be sourced from there instead of James. But his stats are very incomplete. There is no record at all in Soccerbase for season 1994–95; he played only twelve PL games in the previous season but just two of those games show up in Soccerbase; and while his PL stats for season 1995–96 are correct, his domestic cup records are incomplete throwing the tallies completely off for that season too. The Soccerbase records for his final two seasons at City are completely correct IMO. I have no idea whether any of the stats for the clubs he played for after he left City are correct, but I have no intention of finding out. Once the Soccerbase records can be shown to be incomplete in some manner for ANY of the seasons for which we wish to use it as a citable source it becomes worthless. Missing some records for each of Rösler's first three seasons at Maine Road makes it as useless a supporting source as its records for Franny Lee and Brian Kidd are in supporting any of the table's stats for those two players (go check them out). Just be glad that Rösler's records for the whole of the 1994–95 season are omitted by Soccerbase because otherwise I would have three citable web sources for the Barnet game compared to your zilch, instead of just the two I currently have! :) 66.16.144.18 (talk) 23:33, 12 July 2015 (UTC)


 * As you picked up on before I did, the Bluemoon stats are indeed identical to the ones in the MEN cites, at least for the players considered above, and I agree that they are the likely source. That means 1939–40 matches are included in the Bluemoon stats. So to my surprise the differences to the book version are numerous. They could be "good" differences (revisions in light of new information) or "bad" differences (transcription errors), but either way the differences exist so they cannot be regarded as the same corpus. By Ric further above I was referring to Ric Turner, who runs Bluemoon, per the footer on each page.
 * I sympathise with your Soccerbase travails, I've been there myself. The column they put playoff appearances in might as well be a lottery, and that's if they are included at all. What I'd give for football to have a statistics site that is as reliable as Cricinfo is for cricket.


 * BTW, I totally agree with you that despite also being sourced from James, the Bluemoon records database should be treated as a separate corpus of data to James' book. Because it embodies, as you say, both "good" and "bad" differences, and neither of us is necessarily going to know which (of any discrepancies that may arise between James' book reported data and the same data as now represented on the Bluemoon web site) is the correct version. As I've checked through the stats. of all the Soccerbase sourced players in the table over the last week or so I have found numerous disconnects between the Soccerbase numbers and the comparable Bluemoon teamsheet database tallies and in almost all cases the discrepancy is caused by a stupid error in the Bluemoon data (along similar lines to the Book-Booth swap). In the few cases where Soccerbase is at fault it is always for one of the two reasons I have already documented in the article - either a player's goals tally is off because Soccerbase never readjusted to a later DGC assignment of a goal to or from another player, or a player's appearances tally is off because he has participated in one or more of the Community Shield finals (and all of the players with >100 appearances that have such an error caused by having participated in one of those three games have now been footnoted).
 * Consequently, this makes me leery of using the Bluemoon data as a directly cited source of any of the stats. numbers in the article in the same manner that I am wary of citing comparable data from the MCIVTA, MCFCstats and Happy to be Blue! web sites. Being correct for 80% or even 90% of the players is just not good enough IMO. The Bluemoon data is probably the best of the lot but it still cannot hold a candle to Soccerbase in which, despite its two really annoying quirks mentioned above, I have still yet to find any fundamental errors. In contrast, Bluemoon has teamsheets showing players subbing in for players that were not even in the starting eleven, and you don't have to be an avid City fan - or even a casual follower of football - to identify such basic logical contradictions.
 * Of course, I can only spot such errors in Bluemoon WRT the most contemporary players whose stats. are also available elsewhere on the web (such as Soccerbase). I assume you are finding the same level of disconnects between the data in James' book and the comparable Bluemoon numbers for the more historic players (i.e., any player playing in the old millennium!). 66.16.144.18 (talk) 12:02, 26 July 2015 (UTC)


 * Verifiability not truth is indeed the mantra (I could offer to email photos of print sources as proof, but to do so would be to spectacularly miss the point). I just like to get to the bottom of these things when they occur, to try to understand why they happened, where an error first propagated from. The "patient zero", if you like. Irrespective of whether the "true" answer can be used on Wikipedia or not, I can at least make a note in my own records. To that end, I have some ideas about the origin of some of the differences, but there's no need to bother you with them here. Web cite with a footnote about sources differing is the way forward, I agree. Oldelpaso (talk) 19:36, 13 July 2015 (UTC)

Paul Power stats
According to the James sourced data on Bluemoon Paul made 444 appearances due to only 6 sub appearance in the league, not 7. Once again we have a discrepancy that traces back only to James himself. Whether Gary or Ric (if he is the person coordinating the Bluemoon statistics effort) has made an error in the web site data entry, or whether there was a publishing error in James' 2006 book and he has taken advantage of this latest Bluemoon effort to now correct the record, neither you nor I are in a position to determine. Unlike the Rösler situation where we were able to very quickly isolate the Barnet game as the source of the discrepancy, I have absolutely no idea which league game is causing the problem in Paul's case. If it was a domestic cup or Euro game I might be prepared to search through his Bluemoon record, but I'm not willing to do so for the league games! I wasted over two hours analyzing Uwe's Soccerbase records at the weekend and he only played 4 and a bit seasons for the club. Paul's career is almost three times as long as Rösler's and I don't have 6-7 hours to spare similarly going through his Bluemoon records. If you are willing to do so and are able to isolate the game causing the discrepancy then we can footnote it. 66.16.144.18 (talk) 19:00, 13 July 2015 (UTC)
 * Found it. 13 December 1975, Coventry (h). Listed as replacing Booth in the book (p. 419), but not on the web. Oldelpaso (talk) 19:51, 13 July 2015 (UTC)

Tony Book stats
Because of the info. you gave me I was able to very quickly isolate the cause of the discrepancy in Book's stats. Once again using the Bluemoon records, they too have him only playing 16 Euro games, not 17. Based on your info. the cause of the discrepancy appears to be the away game against Honved. They show Tommy Booth playing in the #2 shirt, not Book. Book played in the home league game against Southampton four days prior, and he also played in the away league match against Wolves three days later, so it appears to be a data entry error with Booth being entered as the right back instead of Book. This wouldn't have been possible if he had been playing as the centre half (#5) position as he normally did - because someone would have spotted two Tommy Booths in the team by now! - but Heslop was playing centre half against Honved. Checking into this a little further reveals that Booth appeared to get injured in, or just after, the home league game against Stoke City in mid-September. For a period of three months after the Stoke game until the derby match at Old Trafford in mid-December, Booth is out of the side and Heslop becomes the first choice CH for the duration. I hardly think they would have patched Tommy Booth up and played him as a right back in the middle of his injury period even if Book had to be out for one away European fixture. They would have played Arthur Mann - because, after all, 'arf a man is better than no man at RB at all! - or moved Pardoe (who was right-footed despite normally playing left back) over there and brought someone else in as left back - Dave Connor, Willie Donachie? So this one looks like a Bluemoon error, and now I'm convinced more than ever that M.E.N. journalist James Scott simply used the Bluemoon site as the source for his series of 50 articles. Suggest we still keep, rather than lose, the web citation but footnote this one too, and that the table number be 315 rather than 314. 66.16.144.18 (talk) 20:24, 13 July 2015 (UTC)

Tommy Booth stats
Because of the Book-Booth swap error identified above we now immediately know the source of the discrepancy in Booth's appearances stats. It's Bluemoon once again. Because they have him listed as the right back in the Honved game his appearances are one too many and your original number (491) is the correct one (based on 25 not 26 Euro games). I think we should still source this to M.E.N. but put 491 not 492 in the table and footnote the discrepancy from the cited source. Regarding his goals tally which I decreased by one, as you correctly identified, this is the goal attributed to "Other" by James in his book, but which Bluemoon attribute to the young substitute Leman (I'm assuming, otherwise Royle's goal tally is going to be one off - perhaps you could investigate? - since we know Marsh's tally is in agreement with your original total because I've already sourced him to M.E.N. as one of the original 20 players with only career date issues). Once again the discrepancy is due solely to later web James versus earlier book James as the source of both numbers. This needs to be footnoted but I'm not sure which number should appear in the table. I think we can be pretty sure the Book-Booth swap is an error so we should correct that in the table, but I have no idea who scored that goal (Booth or Leman or Royle) so we should probably align the table with the M.E.N. (ultimately Bluemoon) source and just footnote the discrepancy with the earlier James book attribution. 66.16.144.18 (talk) 22:51, 13 July 2015 (UTC)
 * Second time of asking: do you know which player is being wrongly attributed by Bluemoon with the goal that James claims in his book Tommy Booth scored in the Anglo-Scottish cup game against Sheffield United? Can you also give me the page reference so that I can include it in a footnote? I would also like a page reference for the Honved UCWC cup tie team line-up so I can include that in the footnote I've already written which you need to validate against James and let me know that you agree with it. I got a little ahead of myself in writing that particular footnote. I claim in it that James has Book playing in the game with Booth taking no part in the match. I happen to know that to be true because I saw City play every home game that season. That obviously doesn't help me with the away fixture at Honved but I definitely do remember him being out during one of his main seasons during the period I was a regular in the Kippax due to his having an operation on his knee to fix his cruciate ligament. Booth, Doyle and Oakes were almost regular fixtures in the team at that time, playing every game, so his absence from that mighty midfield triumvirate (and Heslop taking over the CH role) is something that I do still remember 45 years later, although until this issue came up to jog my memory I would not have been able to pin it down to the autumn of 1970. I'm surprised I even recall it after so long a period, but Google is your friend once you know what you are looking for, so I have now found a few articles out there that prove he could not possibly have played in that game. But that's all by-the-by; the footnote isn't about my own uncitable knowledge and fond memories. It addresses what Bluemoon states versus what James states in the book, and I have no idea what James states until you tell me. But if it is not in line with what I have already written then we have a big problem! Finally, to round things out, could you also send me a page reference for "Der Bomber"'s appearance in the Barnet game and I'll add that to his footnote in the same manner as I included your page reference for the Coventry City match (which I was also at) in Paul Power's footnote. Thanks. 66.16.144.18 (talk) 20:36, 15 July 2015 (UTC)


 * The scorers are listed as Booth, Marsh, Leman. So its Royle, and that checks out when Royle is cross-referenced. 31 goals in the book, 32 on Bluemoon, Other being the only column with a difference. Page 538 for Royle totals, p. 515 for the Sheffield United goal being credited to Booth.
 * The book has Booth playing 19 September against Stoke then absent until the derby on 12 December. Nothing to contradict the footnote. Page ref added for Rosler. Oldelpaso (talk) 17:06, 16 July 2015 (UTC)

Ken Barnes stats
tba

Neil Young stats
tba

Alan Oakes stats
tba

Niall Quinn stats
tba - BTW, note that the "443" total apps number in the article edit summary I wrote for his update is a typo and should really be "243".

Data resolution status
Notes:
 * Individual player statistics data web sourcing effort – Status of James Scott's "Top 50" MCFC players
 * 1 – No change to data; M.E.N. RS citation added
 * 2 – Career dates span updated; M.E.N. RS citation added
 * 3 – No change to data; M.E.N. RS citation added (but does not support career dates span as that aspect incorrect in cited RS)
 * 4 – Career dates span updated; M.E.N. RS citation added; footnote(s) added
 * 5 – Appearances and/or Goals tallies updated; M.E.N. RS citation added; footnote(s) added
 * 6 – Career dates span updated; Appearances and/or Goals tallies updated; M.E.N. RS citation added; footnote(s) added
 * 7 – Career dates span updated; Appearances and/or Goals tallies updated; M.E.N. RS citation added (but does not support dates) ; footnote(s) added
 * 8 – Appearances tally updated; footnote added documenting Soccerbase error
 * 9 – Goals tally updated; footnote added documenting Soccerbase error
 * 10 – Originally sourced erroneously to Soccerbase on 2 July 2015
 * 11 – Career dates span updated; no supporting RS citation added

66.16.144.18 (talk) 04:25, 22 July 2015 (UTC)