Talk:List of biggest box-office bombs/Archive 2

The problem with original research
This was brought up before, notably here but this entire list is the result of original research. There are no sources provided which attest to the estimated losses claimed in this article. The formula which arrives at them is based on original research and a synthesis of sources not related to their ultimate conclusion. The calculations are not routine, as was previously suggested, because they are not correct and they are not obvious (a section of this article is devoted to explaining them and they are still only estimates). But to illustrate the problem more directly, and thus highlight the reason articles shouldn't be built this way, here is an article from Business Insider which directly contradicts this article's estimated losses for Tomorrowland (This article estimates ~$75m, BI and others estimate ~$140m). Wikipedia is not a place for publishing original calculations, regardless of the reasoning behind them, and only reliable third-party sources should be used to make claims in an article. 2001:57A:400B:101:54D3:FEC6:58:1E4F (talk) 19:25, 12 October 2015 (UTC)
 * I agree this list has many problems but the biggest problem is the lack of potential solutions. It survived an AfD so that means we actually have to do something with it. I proposed ranking films by published writedowns in that discussion you linked to, but it was never followed up. Part of the problem is that there are hugely different methodologies for calculating loss. Let's take John Carter for example: Disney took a writedown of $160 million according to The Economist, while according to The Numbers it lost $128 million, while Filmsite give a figure of $109 milion. The other problems is that you can't really rank losses that are calculated by different methodologies because that would constitute WP:SYNTHESIS (i.e. say one source factors in marketing costs and the other doesn't). Maybe one approach would be to list films alphabetically to avoid synthesizing a non-level comparision and just list all loss estimates we can source? Betty Logan (talk) 20:08, 12 October 2015 (UTC)
 * Yes, while its easy to find fault in this list, I was also at a loss for ideas on fixing it. Granted, it can always be re-nominated for deletion, I don't think that's necessary or even the best solution. Still, something must be done. The claims in this article are almost certainly inaccurate and they are propagating across the internet. BuzzFeed ran an article which seems to use this Wikipedia page as its only source (a rare miss for BuzzFeed's crack fact-checking team). But I do like your idea. My biggest objection is simply to the "Estimated Losses" column, the absence of which would bring this article more in line with your suggestion while retaining its valuable information. Unfortunately, as much as we'd love to see them ranked, there just doesn't seem to be any feasible way to calculate the losses nor to sort the films based on their losses. Other publications don't seem to even try to calculate losses,  they simply list the budget and gross earnings objectively (Business Insider, again) (LA Times) and some of their declared 'bombs' had slim profits. If the films were organized alphabetically, it would still be possible to include sourced estimations on their losses, where applicable, even as a range of figures. Otherwise it may simply be overreaching to try to sort them all based on that without the sources. 2001:57A:400B:101:54D3:FEC6:58:1E4F (talk) 21:57, 12 October 2015 (UTC)
 * I am especially keen to avoid sources that rely purely on hyperbole. It is very easy to call something a flop without providing any evidence of financial analysis, so I would prefer to keep the list quantifiable i.e. go with sources that provide figures. I think an encyclopedia should stick to facts as much as possible. Case in point: there are numerous sources that list films such as Cleopatra and Waterworld as bombs when in fact neither were: Waterworld underperformed a bit and broke even on video while Cleopatra was actually the biggest film of the year, but just cost an exorbitant amount of money (but still broke even thanks to a TV sale in 1966). I think one way of approaching this would be to have a table with films listed alphabetically, but perhaps have sortable columns for The-numbers estimates, the Filmsite estimates, writedowns etc. I will knock up a small table later as a working example and if it is acceptable we can start to model a new article around it. Betty Logan (talk) 22:41, 12 October 2015 (UTC)

I think we may be hitting a wall here. We've acknowledged that there is a problem with varying methodologies As I understand it, a writedown is a decrease in the estimated value of an asset on balance sheets. Make a film for $50 million, project a profit of $75 million above that from gross revenue of $160 million. Actual gross revenue is $50 million. After theaters take their cut and anyone else who gets a bite, the studio brings in, say, $40 million, for a loss of $10 million. The writedown would be from the original estimated value of the film from $75 million to -$10 million, a writedown $85 million, but an actual loss of $10 million. One site reports a writedown of $85 million, two other sites report estimated losses of $5 million and $15 million. To readers, it looks like three completely different outcomes: "Lost $85/15/5 million on a $50 million film? That's a (major bomb)/(disappointment)/(rounding error).

At the same time, we acknowledge a strong desire to list "box office bombs" and rank them. Two problems: what in the hell is a "bomb" and how do we rank them? Mega Studio Corporation spends $150 million to make and market a film, with tie-ins at fast food places, stars on talk shows, "Now a major motion picture" versions of the novel and everything from stuffed animals to shower curtains in the stores in time for Christmas. After untangling who gets what from the inevitable Lego-Angry Birds-(movie name here) XBOX ONE game sold at Target (and who pays for the copies that don't sell), different sources say the studio lost $5 million, broke even or made $5 million. Joe's Basement Film Co. makes a movie using $5,000,025 from his brother-in-law's hedge fund fortune. He discusses a deal over coffee and the owner of a local theater picks up the tab and runs the film for one night. No one goes to the film's only showing. He lost exactly $5 million. The first film lost 3.75%, spread out over 2 years, multiple subsidiaries and such. With a bit of advanced accounting, they end up in the black. Sort of. Maybe someone lost a bit of credibility, maybe not.

The second film lost 100%. Joe is ruined. His brother-in-law really thinks he's an idiot now.

Both lost $5 million. There's no sense in comparing them. Maybe we go on percentage lost? Joe's the winner for sure, but the comparison is still meaningless. Extreme, absurd hypotheticals? Absolutely, but similar judgement calls will inevitably skew whatever list we synthesize out of our desire to have a list.

We really, really want a list. That's swell, but it doesn't mean a list is anything more than us pretending we have a way to synthesize something verifiable.

Presumably, there are independent reliable sources that spit out lists of the "10 Most Biggest Box Office Bombs EVRRRRRRR!" Do various lists agree? If so, great: Tape them all together here and we're done. If not, take the hint. - Sum mer PhD v2.0 15:20, 23 October 2015 (UTC)
 * Plenty of editors have criticized the form the article takes which I agree with to an extent, but the way forward is not clear. The "hint" I should be taking is not obvious to me. Washing my hands of it? Deleting it? It has already been tossed out of AfD since there are many sources covering the topic of "box office bombs" thus establishing its notability. We could send it there again but I don't see why the outcome would be different this time around which is why I am trying to be pro-active. There are plenty of sources covering the topic, but the problem is there is a huge disparity between the figures they give and very little consistency. There is no valid reason for picking one figure over another so whichever solution we adopt is going to have to embrace this fact rather than resist it. Betty Logan (talk) 16:31, 23 October 2015 (UTC)


 * Well done to the editors involved here so far thinking about this problem - even if we haven't found a solution yet, this is noble work! I don't have much useful to add, I'm afraid, but here are some thoughts...
 * The prototype looks good, and is thorough, but I do have one objection to it: it's visually busy and therefore hard to read. I was confused about the Writedown, Numbers and Filmsite columns, because it wasn't immediately obvious to me that The Numbers and Filmsite are websites and not methods of calculating money somehow (unlike Writedown). I don't know if there's any way to avoid this, though, and the prototype is probably an improvement over the current article at least.
 * Maybe a stupid idea, but... could you simplify this to one column labelled "Estimated losses" and then simply put the lowest AND highest estimates in there, sourcing both? For example, "$10 million - $20 million", ignoring any estimates that fell between them?
 * My other worry is what constitutes a box office bomb when most movies eventually make their money back from TV, DVD licensing etc (as in the Waterworld and Cleopatra examples). Does this mean films will be added to this list until the day comes when they make their money back? Were Waterworld and Cleopatra bombs at one point at time, if not now? Popcornduff (talk) 17:12, 23 October 2015 (UTC)
 * Having a loss "range" instead of each individual estimate would work for me. It is not immediately clear what to do about films like Cleopatra and Waterworld. As Filmsite states: "Some films are unjustly labeled flops, such as Cleopatra (1963) and Waterworld (1995), although both are included in the descriptions. In recent years, some of these low-income producing films have become profitable (after initial box-office failures) with strong international grosses, and further profits from the sales of movies to TV syndication and to home video/DVD releases (or re-releases)." Calling Cleopatra "low income" seems a bit dubious too since it was still the highest-grossing film of the year. My view on this is that if it can verifiable proven that these films eventually broke even they should probably not be on list, if we are limiting the list to films which actually lost money. If sources contradict each other about whether the film broke even or not then I guess that could still be documented in the article. Betty Logan (talk) 20:30, 23 October 2015 (UTC)


 * {u|Betty Logan}}: The prior AfD, to my reading, seems to have been a draw between two options: 1) Delete and 2) Rewrite as a list of notable box office bombs. I don't see much support there for any kind of ranking. What we have currently is rankings based on our synthesis as explained at List_of_box_office_bombs. What you seem to be proposing is selecting several sources and giving rankings based on them. "Notable box office bombs", OTOH, would seem to be an article with sections summarizing what reliable sources have had to say about, for example, Heaven's Gate. Currently not listed here, Heaven's_Gate_(film) has a good bit of content discussing its sizable loss, repeated claims about the losses impacts, etc. As the loss, adjusted for inflation, was "only" $100 million or so, it didn't make it here. Heck, from what we have in this article, it seems that only one film prior to 1995 bombed at the box office. I don't have reason to believe that any amount of adjusting for inflation and such would put Heaven's Gate in a top 10 list of box office bombs, but the coverage at Heaven's_Gate_(film) seems to imply its bomb-itude is notable.
 * I do not seem an objective, encyclopedic ranking coming out of this, though I understand the desire for one. (Much as I like to believe Battlefield Earth is somehow objectively the worst big budget Hollywood film, I secretly know it's my bias and desire to say, "This is THE worst.") I can see room for nuanced, well-sourced prose discussing various films' bombings. - Sum mer PhD v2.0 20:04, 23 October 2015 (UTC)
 * What I have proposed purposefully avoids rankings altogether (at least as defined by Wikipedia editors). It may not be clear from the prototype table below since only a few films are listed, but they are in alphabetical order. There is an implicit ranking by permitting each column to be sortable but that comes directly from the source itself and only ranks in relation to that source. Betty Logan (talk) 20:12, 23 October 2015 (UTC)
 * While that may be your intention, the perception that it is a ranking is inevitable. Currently, this article is used as a de facto source for various film articles' claims that (film x) is the (n)th biggest box office bomb of all times. There's an underlying problem, I think, with your proposed list that I expect would perpetuate this. Yeah, it's sortable, but more to the point, what are the selection criteria? Thousands of notable films have been called "box office bombs" or have been said to have "bombed". We certainly won't list all of them here. The list will have to be narrowed down to either 1) films where a certain number of reliable sources (movie reviews? "In theaters this week" in the Hobokan Gazette?) or 2) the "biggest bombs" ("Gee, I dunno, it certainly bombed, but it was no Pluto Nash...").
 * Side note: Right now we have two films included that are still in theaters. While I don't expect Fantastic Four to suddenly become a hot property, I'm not comfortable with us judging it as one of the biggest money losers ever. - Sum mer PhD v2.0 20:34, 23 October 2015 (UTC)
 * If we're going strictly by the letter of the (wikipedia) law, creating a list for this article may be an infeasibility, due in part to the issue of coming up with reliable figures and then finding a way to organize them. On the other hand, this is certainly a noteworthy subject and one readers would like to see (and like to see ranked). I am still in favor of the middle ground: an alphabetical list with a variety of reliable sources estimating losses (Specifically, Prototype 1). This allows us to include the films objectively, allows users to rank them only by one particular source (thus keeping the ranking uniform to its parent source and not a synthesis), and best of all it gives the reader enough pause to see that there is no objective way to rank losses. When faced with the list and it's multiple columns, they get what they are looking for, but Wikipedia makes no claim as to which was the worst. The reader is thus forced to come to the conclusion that there is no simple answer. This should hopefully prevent another Buzzfeed-esque parroting of data. I'd personally be satisfied with that middle ground, so long as we can find a way to narrow down the definition of what a biggest bomb is. Naturally, we don't want to include every movie ever referred to as a bomb, but we need some metric for exclusion. If we can find that, I think alphabetical+multiple ranking columns is the way to go, personally. 2001:57A:400B:101:B0E9:9A6D:4F8B:FA31 (talk) 17:48, 7 December 2015 (UTC)

Prototype 1
This is what I have knocked up with five test cases. You will notice there are four "loss" columns: Writedowns, The Numbers, Filmsite and "Other". I have put "writedowns" first since this is the closest to an offical loss figure that we will generally get. It is a projected figure based on the studio's own estimates to clear any anticipated losses in a single quarter. Next come the estimates from The-Numbers and Filmsite, and then finally "Others": "Others" is basically a catch-all for any other estimates. "Others" is also the only column that is not sortable since different sources will probably use different methodologies which means they are not directly comparable. All sources are given in the "ref" column except those for The-Numbers and Filmsite. In those cases I just stuck them at the top of each column since the source is the same in each case. Anyway, this should be enough to give other interested editors the basic idea. If it's a goer I suggest constructing the table here and then we can just copy it into the main article. Betty Logan (talk) 04:00, 14 October 2015 (UTC)


 * Even better and more thorough than I imagined, I really appreciate the hard work.. Obviously calculating movie losses is complicated and I think the above chart is a good reflection of that complication, while still being readable. Do you think that this chart style will be difficult for new editors to add to? Is there any way it can be templated for users who may not be familiar with the CSS involved? Thanks again 2001:57A:400B:101:A802:F82E:67D6:E203 (talk) 18:27, 14 October 2015 (UTC)


 * It should be easy enough to knock up a template to simplify data entry. It's basically an afternoon's work though so I will have a go at it this weekend. Betty Logan (talk) 18:35, 14 October 2015 (UTC)


 * I have knocked up a template, and replaced the table above. If you click on "edit" you will be able to see the markup. While it is still not simple (i.e. a complete novice would still struggle with adding entries) the average editor should be able to cope. There may be glitches in the template (which is still residing in my sandbox) but those can be ironed out. I have also switched around the Filmsite and The Numbers columns since Filmsite seems to be more comprehensive, so should probably come first. Before we move any further with this we also need to decide what to do about the "inflation" chart also in the article. Betty Logan (talk) 02:19, 18 October 2015 (UTC)
 * Looks good to me. I would have thought that Box Office Mojo and Box Office magazine would keep track of this sort of thing, but I can't find anything. NinjaRobotPirate (talk) 18:30, 23 October 2015 (UTC)
 * What do you think about Popcornduff's suggestion above about having a loss range? That would effectively consolidate the conflicting estimates into one column. Betty Logan (talk) 21:54, 29 October 2015 (UTC)


 * - Your table looks good to me. (I have confidence in you knowing what you're doing with film article, I almost always agree with you). I do have a question; when the actual table is completed, will there be a way of ranking the films? By this I mean, determining which films lost the most money overall, then the second most, third, and so on. I ask because I couldn't tell from your mock-up. Also, thanks for all your work. - the WOLF  child  23:01, 28 October 2015 (UTC)
 * Like I said, "While that may be your intention, the perception that it is a ranking is inevitable." - Sum mer PhD v2.0 02:47, 29 October 2015 (UTC)
 * Presumably if the article remains we have to list the films in some order though? I don't really see a way around that. I still think an alphabetical list would be the best approach, but an alternative solution would be a chronological listing perhaps, with years in the first column. At least then the formative impression that the reader forms is one of a chronological list. Betty Logan (talk) 22:04, 29 October 2015 (UTC)
 * The crux of the problem is that different sources have different estimates of loss which makes it impossible to have a singularly ranked list. If you rank the films by loss whose figures do you use? My proposal is to rank them alphabetically and have sortable columns for the figures. It seems like the best compromise to me. Betty Logan (talk) 21:59, 29 October 2015 (UTC)
 * I agree with using a sortable table. You you can rank them in either alpha- or chrono- order, but if you add columns for total overall loss per source, then they can also be ranked in order of most to least losses (and vice-versa), by using the sort function. How many different sources do we use anyways?- the WOLF  child  00:59, 31 October 2015 (UTC)

This conversation is getting really complicated, and it's not even obvious anymore where to reply. I prefer the original table, but I think users would prefer a less complicated version, as suggested by Popcornduff. So, I guess it's probably best to go with a range. NinjaRobotPirate (talk) 22:38, 29 October 2015 (UTC)

Prototype 2
Here is the second prototype converting the table to a loss range per. In each case we take the lowest and highest verifiable estimate for the loss. Several points: If anyone has any further suggestions I am happy to hear them out and incorporate them if possible. However, I will begin the process of replacing the chart in the table hopefully in the next couple of weeks. Ultimately the prototype we have here is more consistent with Wikipedia policy than the current version we have in the article, so objections to the prototype are not really sufficient to not push ahead with this. I won't begin the transition for at least a week, so if anyone wants to put forward a viable alternative then there is still time for them to do so. Betty Logan (talk) 03:29, 6 November 2015 (UTC)
 * The chart needs to be ordered some way and despite the objections by above I think an alphabetical order is the most neutral way of doing this. I don't see how we can avoid having a list that isn't ordered in some way, unless we don't have a list, and that puts the article into the realm of AfD.
 * The table will be sortable, so readers will be able to order the list on loss if they wish to (however this will require an active decision on their part).
 * I have retained the "writedown" column since this is the closest thing there is to an official declaration of loss by the studio.
 * I have taken out the "adjusted" column since I think this would work better as a separate table
 * I think this looks like a major improvement. It's much less busy and listing the loss as a range looks way better than I thought it would. (I didn't have much confidence in the idea at first!) Popcornduff (talk) 03:26, 14 November 2015 (UTC)
 * I find the table very well-done. My only suggestion is to perhaps put "Year" next to "Title"? I find titles and years often paired together and think it would be good to maintain that here as well. Erik (talk &#124; contrib) (ping me) 23:22, 24 November 2015 (UTC)
 * That was briefly discussed above, but didn't really receive any support or opposition. I am fine with doing that though if there are no objections forthcoming. Copying the data over will take a few more days at my current rate and if nobody opposes I will switch the columns over at the end of the process. Betty Logan (talk) 23:29, 24 November 2015 (UTC)
 * - Between this table and the one above, this one would be my choice. One question; how did you determine the order in which they're listed? (Or are they random?) By the way, this looks like a lot of effort, I would like to say 'thanks' for all your hard work here. - the WOLF  child  16:16, 4 January 2016 (UTC)
 * It is certainly taking me longer than anticipated, but xmas got in the way a bit. The chart below is ordered alphabetically. The reason for this is the losses are sourced from different places and there are substantial variations in estimates so I felt it was the most neutral approach. My main aim at the moment is to replace all the OR calculations with figures that actually come from sources (rather then editors' calculators) and then if editors feel that there is a more appropriate way to order the chart then that can be decided in due course. Betty Logan (talk) 20:55, 4 January 2016 (UTC)


 * Couple questions; what's the criteria for inclusion? A minimum amount of loss? Have you consider setting a static number? ('top 50' or 'top 100' box office bombs). And, could you add a number column? - the WOLF  child  03:56, 11 January 2016 (UTC)
 * At the moment I have just been copying them in from the main article and adding sources. The inclusion criteria can be decided in due course. My primary objective at the moment though is to have a table that is fully sourced. Betty Logan (talk) 05:42, 21 January 2016 (UTC)


 * Notes



- Hey, the table looks really good with your latest update. Impressive work! Cheers - the WOLF  child  19:53, 19 February 2016 (UTC)
 * Yeah, I binned the writedown column since there were only 8 amounts in it. I decided to replace it with an "adjusted column" since that probably has more value. After I've filled this in we need to start thinking about inclusion criteria. Betty Logan (talk) 20:04, 19 February 2016 (UTC)
 * Give any thought to adding a column on the far left to number/rank the films, similar to the table at "Highest Grossing Films"? - the WOLF  child  20:26, 19 February 2016 (UTC)
 * Well, I will just tell you the problem, the potential solution, and the resulting problem. We can't just number the chart above because that would be original research i.e. the 20th film on the chart may not be the 20th biggest money loser because some films may be missing, some losses may be compiled according to different methodologies etc. We already know this is true in some cases (just compare the "loss" estimates for Tomorrowland for instance). So basically we need a source that actually ranks losses. The only source I know of which does anything like that is the one at The Numbers (bottom chart) which lists the top 20 money-losers. As you can see, there is very little correllation between that chart and the one above, since the one above is compiled from lots of different sources. If you really want a ranked chart in the list, the only way I can see how that would work is if we used The Numbers top 20 for a ranked chart, and then had a second unranked table with all the other estimates. I honestly don't see how else it could be done. Betty Logan (talk) 22:08, 21 February 2016 (UTC)
 * I see what you mean, thanks anyway. That aside, this table here looks just about ready, when do you think you'll add it to the article page? - the WOLF  child  22:29, 21 February 2016 (UTC)
 * There are some "losses" I am still trying to track down, but in theory there is nothing to prevent us adding those at a later date. The above table is ready to be transferred to the article as it stands. Betty Logan (talk) 23:09, 21 February 2016 (UTC)
 * That's quite a list, up to 91 entries now. That's almost double the number the in the article table. I know I asked before, but I'm gonna ask again; what is the criteria for inclusion? Is it any film described as a "box office bomb in a reliable source? Any film that lost a minimum of X amount of dollars at the end of it's initial release? Any film that has lost a minimum of X amount of dollars when adjusted for inflation? Just curious... -  the WOLF  child  00:30, 22 February 2016 (UTC)
 * I am just working my way through the source at the moment. The criteria can be set once we have all the information. At the moment there is nothing under $40 million adjusted so that may be a sensible threshold for the chart. Betty Logan (talk) 01:53, 22 February 2016 (UTC)
 * Well we seem to have ended up with 114 films on the list in addition to those listed below that I cannot find figures for. We could just cut it off at 40 mil adjusted, but the problem with that is that adjusted figures will rise through inflation, so it may be better to cut off the bottom 14 films and have a round 100 on the list, at least for the time-being. Betty Logan (talk) 03:21, 22 February 2016 (UTC)