Vaguely interesting SnG calculation

[eastbay]

Kind of a silly model, but the numbers that come out of it are maybe vaguely interesting:

10 player SnG. No blinds or antes. On every "hand" 2 players chosen at random go all-in (with the usual rules about being covered). One of the players has a constant edge on the others, say, 2, 5, or 10%.

For each of those edges for the better player, what would you guess for his:

ROI
ITM
distribution of 1sts, 2nds, 3rds?

eastbay

[Lori]

How do you give the 'expert' his edge in the model?

Lori

[eastbay]

[ QUOTE ]
How do you give the 'expert' his edge in the model?

Lori

[/ QUOTE ]

For each all-in confrontation, a random number between 0 and 1 is chosen. For no edge, if the random number is greater than 0.5, player 1 wins, less than 0.5, player 2 wins. The edge is created by offsetting the 0.5 criterion to favor one player.

So by "2% edge" I mean that the expert wins 52% of the time. You might also call that 4% edge depending on how you look at it, but that's what I meant.

I picked 2% as the good side of a "coin flip" situation. I was kind of curious if simply taking the best side of a coin flip in this kind of scenario would win any money.

Anyway, here's some results, based on a 10% vig and 20/30/50 payout.

2%: ~0% ROI (just making up the vig), 32% ITM
5%: 15% ROI, 36% ITM, 15/11/10 for 1st/2nd/3rd
10%: 42% ROI, 42% ITM, 21/11/9 for 1st/2nd/3rd

Taking the bad side of every coinflip:

-2%: -17.6% ROI (ouch!)

eastbay

[Lori]

Interesting, thanks /images/graemlins/smile.gif

Not sure what we can make of it, but I suspect the model might be of use in the future.

Lori

[PrayingMantis]

Great stuff. Very interesting.

I remember something a wise poster once said on the MTT forum (I don't remember who). It was in regard to "playing like Hansen". He said something like: The difficult part is not simply pushing small edges. The difficult part is recognizing where and when you have these edges, and that what makes a great player.

Another thing is, that your model, however artificial, simulated, unrealistic, etc, etc, beatifuly illustrates the constant mistake in seeing PPs vs. Over-cards showdowns as "coin-flips". According to your model, the guy who usually has the PP will be, in the long run, and in terms of ROI, WAY ahead of the guy with the over-cards. Actually, the one with the PP will be a very significant winner, while the other will be a long term *big* loser. Something to consider.

[eastbay]

[ QUOTE ]

Another thing is, that your model, however artificial, simulated, unrealistic, etc, etc, beatifuly illustrates the constant mistake in seeing PPs vs. Over-cards showdowns as "coin-flips".

[/ QUOTE ]

Yep. And that was really the main point of the exercise.

eastbay

[chill888]

Praying Mantis said:
Another thing is, that your model, however artificial, simulated, unrealistic, etc, etc, beatifuly illustrates the constant mistake in seeing PPs vs. Over-cards showdowns as "coin-flips". According to your model, the guy who usually has the PP will be, in the long run, and in terms of ROI, WAY ahead of the guy with the over-cards. Actually, the one with the PP will be a very significant winner, while the other will be a long term *big* loser. Something to consider. </b>


chill888 replied:
Careful. That advice runs the risk of making people even more likely to play small and medium pairs far too fast - on the belief that they are on the right side of the coin flip.

Let's take my recent favorite 99. Sure it is a slight favoright versus 2 overcards (AK, AQ, AJ KQ, etc.) but it is a 4-1 dog versus overpairs. Like the KK I held 10 minutes ago in the 2+2 tourney where a guy raised my bet and pushed 99 early in tourney.

When your pushing these medium pairs or calling big bets, who knows if your facing 2 overcards or a big pair?

Regards

[eastbay]

[ QUOTE ]


chill888 replied:
Careful. That advice runs the risk of making people even more likely to play small and medium pairs far too fast - on the belief that they are on the right side of the coin flip.


[/ QUOTE ]

If you had to refrain from saying anything that might be misinterpreted or misused on this board, you'd never post.

Mantis' statement is correct.

eastbay

[chill888]

lol,

No insult was intended. What I said was also correct.

[eastbay]

[ QUOTE ]
lol,

No insult was intended. What I said was also correct.

[/ QUOTE ]

No insult taken, and true enough.

eastbay