This is obviously a nice article, but a practical example, let's say with 4-5 players at the table, instead of just a formula would be nice.

There actually are two examples in the article, one with 4 players and the other with 10 players.

Twoplustwo's web team is dealing with making the second example's table readable, sorry for the garbled format at this stage.

However, this is mostly useful when there are more than 10 players left and ICM computations are not available. The examples are given for comparison purposes.

Ok, I didn't make myself clear enough. Sorry for that.

What I meant was instead of

prob(p) = so / CP * so/ (CP - so]) * ... * so / (CP - i * so) * ... * s / (CP - (j - 1) * so) * ... * 1/2 * 1 : if we finish in jth place in p

having something like prob(p) = 2300/18000 * 2300/(18000-2300) *.....etc.

Just doing it with real numbers to show how much effort it takes to do it at the table. Also it is quite confusing to use a formular where the reader has to scroll up to see what so or CP means.

Alix,

Thanks for a very interesting and important article. I think it addresses one of the most important questions in tournament poker; namely, mapping chip equity to dollar equity. Unfortunately, I am also unable to adequately parse the formulas you have given, and as a result, I cannot even derive the ECM values you provided in the simple 4-player-remaining scenario.

I would like to "mess around" with the ECM model in a great number of scenarios, varying the stack sizes, players remaining, and payout schedules. Could you show us how to calculate the ECM numbers? I think if you showed the derivations for the 4-player example, I'd probably "get it" from there. I know that for non-trivial examples, the ICM is not implementable in a spreadsheet; does this also hold true for the ECM? If possible, a quick spreadsheet would also be illustrative; if not, perhaps some algorithmic pseudocode?

Thanks.