16s: Surprised SNGPT says fold

[Shrooma]

Chart changing when simulating future games, showing J4 is a call when SB pushes 10%:

[Slim Pickens]

Can you elaborate a little on exactly what the future games simulation is doing? I don't really trust those until I see exactly what it's putting in.

[Shrooma]

The Simulate Future Games option has been beta for a long time but should be working properly according to SNGWizard support.

For an explanation, I'll quote the SNGWizard help file (reference SFG):

[ QUOTE ]


Future Game Simulations


Perhaps the single biggest criticism of the Independent Chip Model (ICM) is that it only takes into account the stack sizes of the players in the current hand. It ignores what may happen in future games.

This limitation is not significant for most situations, but it can be significant when you have a very short stack in the game. ICM does not account for the fact that the approaching blinds will soon eat up most or all of the short stack's chips. Therefore, ICM tends to overvalue short stacks.

The SitNGo Wizard attempts to overcome this limitation by introducing Future Game Simulation (FGS). For every possible outcome of the current game, FGS creates a simulation of what may happen in next few games. This provides a more accurate estimate of the possibility of the short stack busting out.

The actual simulation of each future game is fairly simplified. It only considers the possibility of each player pushing all-in and another player calling the all-in bet. It does not consider pre-flop limps or raises or any post-flop decisions. However, when you have a very short stack in the game often the only action that significantly affects equity is all-in showdowns with the short stack. The value of FGS is estimating the probably of the short stack busting out.

When is FGS enabled?


For most tournament games, FGS adds little value. The chances of a player busting out within the next few hands are too small to make it worth the effort. FGS is enabled when all if the following conditions are met.


The Future Game Simulation option is turned on, either by setting SimulateFutureGames to true in the Options dialog or by checking the Simulate Future Games game view option (the Options link in the upper right hand corner of the game view).

There are 5 or fewer players in the current hand.

At least one of the players has a stack size of 3 big blinds or less.



How does FGS work?


For every possible outcome of the current game, a future game simulation is created. The final stacks from the current outcome are used as the initial stacks in the future game. The player's positions are rotated so that the blinds are correct.

The simulation starts with the first player to act. Based on opponent modeling, the Wizard computes the probability that the first player will push all-in. Then the Wizard then computes the probability that each of the remaining players will call the first player's push. Each call creates a showdown and the equity from each showdown is computed. When nobody calls the first player's push, he picks up the blinds.

The simulation repeats the procedure for each player in the hand; each player pushes and each remaining player calls or folds. The equity from all of the outcomes is summed up to produce the total equity of the future game.

If the total probability an outcome of a future game is greater than about 1%, another future game simulation is created for that outcome. It is possible to have several levels of nested future games.




[/ QUOTE ]

[Slim Pickens]

Well, yeah. What I'm asking is what are the ranges it's using in the future games simulation and are those ranges reasonable? One thing worse than making a range error once is making it six more times.

[eurythmech]

I would love to see the exact mechanics behind FGS

[Shrooma]

You're right, I think the ranges used there are questionable but not completely retarded.

For this particular situation one future game is simulated. Villain is now the button and we are in the small blind. Starting stacks are now: button 1870/ hero 5560/ BB 6070.

Hand ranges:

Button push 25%
Hero call 20%
BB call 25%

Hero push 20%
BB call 10%

Now this future simulation lets the fold equity drop down in OP's hand.

List of equity in simulated future game:

Button wins blinds 35.51%
Button beats Hero: 31.37%
Hero beats button: 41.75%
Button beats BB: 35.23%
BB beats Button: 37.87%

Hero wins blinds: 37.34%
Hero beats BB: 46.36%
BB beats Hero: 20%

BB wins blinds: 35.89%

[SitNGo Wizard]

FGS uses the opponent model assigned to each opponent to determine player hand ranges. You can click on any outcome link in the analysis details to display the Future Game browser and see exactly what hand range is used for every future game showdown.

FGS is still considered experimental and somewhat limited in Release 1. Release 2 will permit more user control over FGS hand ranges.

[Shrooma]

[ QUOTE ]
FGS uses the opponent model assigned to each opponent to determine player hand ranges.

[/ QUOTE ]

However, we are not able to assign a hand range to the button (BB in future game) here as he was not involved in the first hand.

[SitNGo Wizard]

In that case, and in the case of the Hero, it uses the Average model.

[Fishing2do]

For me this situation all depends on the skill of my opponents, I guess this is sorta FGS principle.

If the Btn was a Good player, and villain a really Bad one... I would be more inclined to fold and start attacking the Btn. However in this situation you got an equal stack to the Btn, so attacking him would be harder/less profitable.
Therefor I'd call no matter what the skill of the opponents are. You got a decent $EV situation, so why not try to take him out now you have a legitimate chance.

But if it was like you have 8k chips, and the Good player on the btn has 4k and this Donkey in the SB pushes.. I would check the auto-check/fold btn. And continue to attack the Btn.

this is my way of dealing with the limitations of ICM.