ICM win rate stat possible?

[Zangeeph]

I was thinking about making two custom stats:

ICM adjusted $ won per 100: It does an ICM calculation and looks at the net won in terms of ICM, not in terms of actual chips.
Big stack factor: Divides VPIP when chip leader by VPIP when not chip leader

Am I right to believe that these stats are not possible in PT3 due to each players stats not being aware of other players stack sizes? If not, are they possible in PT4?

[kraada]

Your first stat is not possible in PT3 but is not just possible but available by default in PT4.

Big Stack Factor might be able to be done if you played only a very narrow range of tournaments - in MTTs it will certainly be impossible since we don't know the chip stacks of other players in the tournament.

[Zangeeph]

Well I have PT4 now but I can't find the ICM adjusted $ won per 100 you said that would be there. What is it called in PT4?

Also, any tips of how to build my Big Stack Factor? It does only need the chip counts of the players at the table. This is because I want to know how much a player bullies other players when he can knock them out and has no chance of being knocked out in the hand. I don't know where to start as I'm not sure how to use the stack sizes of all the players at the table.

[kraada]

My Currency Net Adjusted is your adjusted win total and ROI Adjusted is your adjusted return on investment. If you want it as win rate per 100 tournaments you would need to build that stat from the columns in those stats as a custom stat.

For Big Stack Factor - do you play all and only single table SnGs with the same starting stack size? Otherwise I'm not sure how we would be able to put this together for you.

[Zangeeph]

For Big Stack Factor - do you play all and only single table SnGs with the same starting stack size?

Yes, I do. Not really sure why starting stack comes into it but I would truly appreciate your help.

[kraada]

We need to figure out if you're the biggest stack but we can't compare your stack directly to the other stack sizes. We would know, for example, that you're the biggest stack if your stack size is more than half of the chips that would be remaining in the tournament - nobody else can have more. But if you have 40% of the chips and there are 3 players someone else could have 45% with the third person having 15% - we can do some work with effective stack size I think to be a little more accurate - but this won't be 100% perfect, so I'm not sure if it's something you'd want given these limitations.

The best I can think to do is to take the total # of chips, subtract out effective stack size * number of players - this is the amount we know everyone must have since effective stack size is based on the smallest stack size in play. Then if we have more than half of the remainder of the chips (after subtracting ourselves as well), we must be the chip leader. But we still can be chip leader with less, we just can't be certain that we are in a stat.

[Zangeeph]

We need to figure out if you're the biggest stack but we can't compare your stack directly to the other stack sizes.

Why not? Surely, stack sizes of all the players are stored in the database, so why can't they be compared directly? I'm very disappointed.

I thought that knowing stack sizes was required to make correct ICM calculations with regards to the Adjusted amount. Are you sure PT4 is doing the adjusted calculations correctly? They do take a very long time to load (about an hour to generate the adjusted graph), whereas on holdem manager they were pretty much instant.

We would know, for example, that you're the biggest stack if your stack size is more than half of the chips that would be remaining in the tournament - nobody else can have more. But if you have 40% of the chips and there are 3 players someone else could have 45% with the third person having 15% - we can do some work with effective stack size I think to be a little more accurate - but this won't be 100% perfect, so I'm not sure if it's something you'd want given these limitations.

The best I can think to do is to take the total # of chips, subtract out effective stack size * number of players - this is the amount we know everyone must have since effective stack size is based on the smallest stack size in play. Then if we have more than half of the remainder of the chips (after subtracting ourselves as well), we must be the chip leader. But we still can be chip leader with less, we just can't be certain that we are in a stat.

That's a good way to do it actually, as that's an acceptable level of accuracy.

[Zangeeph]

I think I actually would be able to build that statistic on my own, without your help. Thanks for the help though. It's a shame that PT4 won't let you see other players' stacks. Is this feature going to be added because there's tons of amazing stats you could make using the stack sizes of other players. Darn, I'm sad now thinking about it.

[WhiteRider]

When you're build a stat it is based on information about that player - stats are based on all of a player's previous hands and there isn't a way to access the stack size (or other information) about other players in each hand. This would add huge layer of complexity to the custom stats mechanism and isn't really practical.

[Zangeeph]

So then, I suppose that Net Adjusted for tournament something we can never use with stats. Since Net Adjusted (in the tournament graphing section) uses ICM calculations, I previously thought that I would also be able to use ICM calculations with my custom stats. Perhaps in the distant future PT4 would allow players to do proper ICM calculations so they can create more interesting statistics.