Thursday, October 20, 2005
Oi, Sklansky ! No !!
I have great respect for the uber-nerd but something occurred to me out of the blue the other day, and I fear I must take issue.
In Tournament Poker For Advanced Players, page, err, no hang on, I sold it. Well, in there somewhere is an analogy along the following lines :
You can bet $100 today on an 11/10 shot that has a 50% chance of winning. But if you lose the bet, you will miss out on an opportunity tomorrow to bet $100 on a 5/4 shot that's also 50%. So, if you take the first bet your total EV is $10 + (0.5 x $25) = $22.50. This is less than the $25 EV you achieve by passing up today's bet and just betting the 5/4 shot tomorrow. The parallel is drawn with poker tournaments and why you should sometimes pass up a good spot because you figure you'll find a better one.
All well and good until I posit the following. What if winning the first bet empowers you to make TWO bets tomorrow at 5/4 (or to stake $200) ? I think that's more analagous to the situation in a poker tournament. It seems only fit to leave the resulting calculation, and the conclusions to be drawn, as an exercise for the reader !
In Tournament Poker For Advanced Players, page, err, no hang on, I sold it. Well, in there somewhere is an analogy along the following lines :
You can bet $100 today on an 11/10 shot that has a 50% chance of winning. But if you lose the bet, you will miss out on an opportunity tomorrow to bet $100 on a 5/4 shot that's also 50%. So, if you take the first bet your total EV is $10 + (0.5 x $25) = $22.50. This is less than the $25 EV you achieve by passing up today's bet and just betting the 5/4 shot tomorrow. The parallel is drawn with poker tournaments and why you should sometimes pass up a good spot because you figure you'll find a better one.
All well and good until I posit the following. What if winning the first bet empowers you to make TWO bets tomorrow at 5/4 (or to stake $200) ? I think that's more analagous to the situation in a poker tournament. It seems only fit to leave the resulting calculation, and the conclusions to be drawn, as an exercise for the reader !
# posted by Andy_Ward @ 7:10 PM 
Comments:
<< Home
Alternatively, one could argue that if you take the 11/10 shot today and lose, that does not mean that you miss out on the 5/4 bet tomorrow - just that the 5/4 bet occurs in a different tournament.
Hell, we've been through all this before. Let the fools think what they want to think. If people want to turn down the chance of a positive EV bet because they think they will get another, better, positive EV bet in 10 minutes, let them do it. In fact, I WANT them to do it.
"My calling days are just beginning".
Pete
Hell, we've been through all this before. Let the fools think what they want to think. If people want to turn down the chance of a positive EV bet because they think they will get another, better, positive EV bet in 10 minutes, let them do it. In fact, I WANT them to do it.
"My calling days are just beginning".
Pete
# posted by 
: 10:17 PM
: 10:17 PM
I know, I'm certainly not singing that song any more on forums. Although Fox does advance the notion that if someone has a wrong idea, ridiculing them will make them cling on to it all the more.
Andy.
Andy.
Actually I definitely want good players taking on small EVs that risk their entire stack: my equity goes up as a consequence. (as was semi-proved by a lad on BDD's forum last year, in a lengthy debate) I don't, however, want my opponents making lots of small +EV decisions, that risk small amounts of their stack.
chaos
chaos
# posted by : 11:57 AM
: 11:57 AM
" Actually I definitely want good players taking on small EVs that risk their entire stack: "
That doesn't mean it's wrong for them to do it. You have to be very careful with stuff like this. Malmuth constructed a whole chapter of very dodgy tournament strategy in one of his early books by extrapolating small factors like the different value of rebuys and constructing an entire strategy around them.
You can prove that a factor exists but it can still be so small that it's hardly ever worth considering. It is swamped by the uncertainty as to your opponent's exact holding.
Andy.
That doesn't mean it's wrong for them to do it. You have to be very careful with stuff like this. Malmuth constructed a whole chapter of very dodgy tournament strategy in one of his early books by extrapolating small factors like the different value of rebuys and constructing an entire strategy around them.
You can prove that a factor exists but it can still be so small that it's hardly ever worth considering. It is swamped by the uncertainty as to your opponent's exact holding.
Andy.
"That doesn't mean it's wrong for them to do it."
Well that's true; you can gain and they can gain from the decision w.r.t. a pure equity argument. However, if the margin is small, as discussed, then this is particularly unlikely.
It is also true that just because I say I want it, doesn't make it good for me: it may still be harmful - a bit like taking insurance in BJ: most people want it. Nevertheless, I stand buy it. There is little doubt in my mind that I would gain, on balance, if an unskilled and a skilled player go all-in on a race: even with the odds marginally in the skilled players favour.
Re the sklansky thing. No fan of Sklansky, but the argument is worth bearing in mind. There maybe lots of (subsequent) pots that a player might expect to pickup that do not require doubling up. So by taking the gamble, you've gone from having them picked up (with a declination), to having a 50ish% chance of having them by undertaking a gamble. Effectively you're losing equity on future opportunites by punting.
But then of course, doubling up will create it's own opportunites.
I'm not saying I wouldn't lump it anyway...
chaos
Post a Comment
Well that's true; you can gain and they can gain from the decision w.r.t. a pure equity argument. However, if the margin is small, as discussed, then this is particularly unlikely.
It is also true that just because I say I want it, doesn't make it good for me: it may still be harmful - a bit like taking insurance in BJ: most people want it. Nevertheless, I stand buy it. There is little doubt in my mind that I would gain, on balance, if an unskilled and a skilled player go all-in on a race: even with the odds marginally in the skilled players favour.
Re the sklansky thing. No fan of Sklansky, but the argument is worth bearing in mind. There maybe lots of (subsequent) pots that a player might expect to pickup that do not require doubling up. So by taking the gamble, you've gone from having them picked up (with a declination), to having a 50ish% chance of having them by undertaking a gamble. Effectively you're losing equity on future opportunites by punting.
But then of course, doubling up will create it's own opportunites.
I'm not saying I wouldn't lump it anyway...
chaos
# posted by : 5:29 PM
: 5:29 PM << Home
