Talk:Shapley–Shubik power index

I have an issue with one part. If you have 2n+1 members and you subtract a single powerful member, wouldn't the remaining members be only 2n? Rosen643 (talk) 14:11, 3 June 2009 (UTC)


 * I had the same problem. The entire last paragraph makes no sense.Pdronsard (talk) 08:04, 2 November 2010 (UTC)


 * Also, it would be nice to have a contrast with the Banzhaf power index.Pdronsard (talk) 08:04, 2 November 2010 (UTC)


 * This article is confusing because it lacks proper definitions of the function. I'm almost certain the 2n+1 members is probably not including the strong member himself. In other words, we're talking about an even number of chairmen in the vote. Subtracting the strong players power we get (2n+1)+1-k=2n+2-k. I am not confident enough to feel comfortable editing the article itself. --עוז גבריאל (talk) 15:51, 4 January 2017 (UTC)

The example with the 2n weak members and 1 strong member was simply incorrect. It conflicted with the example given in the Shapley 1954 citation, and could easily be disproven using a simple monte carlo simulation (e.g. a 50 line python script). I updated the example to correspond to the citation, fixed the mathematical expression, and added a mathematical derivation of this expression. It might need the hand of an experienced editor to improve the formatting and/or make the language a bit clearer. Joopie99 (talk) 07:45, 12 February 2021 (UTC)


 * I changed it to n weak members (instead of 2n weak members), to make the example a bit simpler.Joopie99 (talk) 01:08, 13 February 2021 (UTC)