Deluxe ICM Equity + Deal Calculator for 2-10 Players

[SmallPotJeff]

A little while back, I posted a 3-person tournament equity/deal calculating spreadsheet. Here's a link to that post: Spreadsheet to make a Fair 3-way Deal (inspired by a bad deal) .
I later verified that its results were equivalent to ICM calculations. However, it had some extra features lacking in most ICM calculators, in that it allowed you to calculate a fair deal in which money was left on the table (thus letting you devise any new prize structure you with), and it also allowed you to adjust for player skill.

Anyway, shortly thereafter I put together a new spreadsheet with the same features which wasn't limited to just 3 players. I hadn't planned on posting it, but I got a request, so I thought I'd post it for all to see.

Here's a link to the new spreadsheet, which can handle up to 10 players: <a href="http://students.washington.edu/jeffgmh/2-10_player_icm-deal_calculator.xls" target="_blank">http://students.washington.edu/jeffgmh/2-10_player_icm-deal_calculator.xls </a>

Here's a list of features on this spreadsheet:

* Results are equivalent to ICM calculations, which are relatively well accepted as a way to calculate equity
* Lets you calculate tournament equities for all players with anywhere from 2-10 players
* Allows you to adjust for different players' skill levels
* Lets you calculate a fair deal, based on players' skill (if so desired), with money left on the table
* Allows you to define a new prize structure (with deal amounts to match), leaving money on the table not just for first place, but for any places desired.

There is one major difference between using the last spreadsheet and using this new one: this 10-player spreadsheet works using a visual basic macro that I programmed, meaning that in order to use it you have to enable macros and then run the macro. I included instructions for use in the spreadsheet, so hopefully it's pretty self-explanatory. Hopefully the macros don't keep anyone from using the spreadsheet.

Let me know what you all think, and if you have any questions or suggestions for additions.

[EDIT]
I should also mention that I had been working on putting together a standalone executable windows program that performs the same calculations as this spreadsheet. But I stopped working on it because I got bored. The beta version is about 2/3 done, so if anyone IS interested in it, let me know, and I might finish working on it.
[/EDIT]

Jurullo -

Haven't read through all this yet, but can this be moved to MTT where I think it will get the right exposure?

On a side note, this looks kind of interesting. I'll dig in later tonight when drunk.

[SmallPotJeff]

No prob - wasn't quite sure the best place to post it. Here's looking forward to your drunken comments.

[sheetsworld]

Great Job Jeff

sheets

[La Brujita]

thanks for posting this!

edit to say i answered my own basic question.

[SmallPotJeff]

First of all - thanks to everyone that's given some encouragement. it's much appreciated [img]/images/graemlins/smile.gif[/img]

Anyway I just received a pm asking about how I adjust equities to arrive at deal amounts:

[ QUOTE ]
Hi Jeff,

I was looking at your ICM spreadsheet and had a question. How is the amount left out of the deal treated? It looks like you calculate what the equities would be without taking the $30k out, and then take out the $30k from each player in proportion to their chip stacks. Is that correct, or does the result of doing a brute force calc with that money not in the prize pool end up the same? If so, how do you adjust the payouts?

Great tool, BTW.

[/ QUOTE ]

I figured I'd quick explain it here in case anyone else has the same question.

As part of the equity calculation, I first calculate each player's probability of coming in each place (so if there are 5 players, for each of those players their chances of coming in each place from 1st through 5th is calculated).
If money is left on the table, in order to find a fair deal amount for each player (i.e. a deal that does not change their equity), they should all receive a little less than their total equity up front, because they also have some equity in the prizes left on the table. In order to calculate each player's deal amount so that their total equity doesn't change, the calculations subtract a fraction of the prize left on the table for each place proportional to that player's chance of coming in that place.

So if player B has a 75% chance of coming in first, and $1000 are left on the table for 1st place, then $750 are subtracted from that player's equity to find their deal amount (b/c they have $750 additional equity from the money left on the table).
This is done for each player and for each place for which money is left on the table. It ends up working out perfectly so that the total adjusted deal amounts combined with the money on the table add up to exactly the total money at stake.
Furthermore, as a side note, you can actually reproduce the original prize structure just by leaving the right amounts on the table (which indicates that everything is working properly with the deal adjusting calculations). You can even generate a fair deal in which all of the money goes to first place (but if the player with the most equity loses, he or she actually Owes money to the other players).

Hopefully this explanation made some sense. Let me know if any of you have any other questions or thoughts to add.

[La Brujita]

im kinda dumb but do i understand what you wrote to mean the following two would give equivilant numbers:

final table of stars million gtd with exactly 5000 entrants, 20k left on table:

1. pretend the prize pool is 980,000

2. keep prize pool as 1M and put 20,000 as left on table for prize one

[rabbitlover]

Jeff, would it be possible to use these spreadsheets or the standalone program (if you pursue it) with a Palm Handheld device?
Would be real handy to punch numbers at the table of a live tournament if a chop is desired.

Thanks for your efforts.

[SmallPotJeff]

[ QUOTE ]
im kinda dumb but do i understand what you wrote to mean the following two would give equivilant numbers:

final table of stars million gtd with exactly 5000 entrants, 20k left on table:

1. pretend the prize pool is 980,000

2. keep prize pool as 1M and put 20,000 as left on table for prize one

[/ QUOTE ]

No worries - I'm not quite sure I understand your question, but I'll try.
If you wanted to find a deal for the final table of the 1M guaranteed tourney in which 20k was left on the table for 1st, then you would simply type in the original payouts for the players still on the table (even though it wouldn't add up to 1M because some of the money has already gone to players who've busted). You would then type 20,000 in the 1st place row under "New Prize Structure - $ Left on Table."

What the spreadsheet would then do is calculate each player's equity at the moment (based on the original payout structure) and then determine how much they should each receive immediately for the deal with the new payout structure (20k on the table for 1st) so that they still have exactly the same equity.

However, to answer your question, this result wouldn't necessarily be the same as if there was simply 20k less in the prize pool to begin with, because that may affect different players' equities unequally, whereas with 20k left on the table for the deal, the adjustments are designed not to impact players' equities.
(As a further example, if one player had a 100% chance of coming in 1st place [which won't actually happen], then the entire 20k would come out of their equity when calculating the deal amount, whereas if there were 20k less in the prize pool, everyone would be affected.)

[La Brujita]

cool. thanks for the explanation.