Genera Concept Behind My "What's Wrong...." Question.

[David Sklansky]

Is that you must never make a play merely because it has a positive EV. You must always compare it with its alternatives. In the example given, we must stray from the usually right idea that if a pure bluff with no outs shows a profit, than if you add in outs a bet is the best play since it will show an even greater profit. In spite of the truth of that, it is important to realize that a non bet may still be the best play of all.

[Leavenfish]

[ QUOTE ]
Is that you must never make a play merely because it has a positive EV. You must always compare it with its alternatives.

[/ QUOTE ]

Makes sense.

I still think there is a lot to be said for taking that final $10, going to the bar for a nice drink and pondering what it was about the session that led to your being in such a dire situation to begin with. I do not know if it’s my insatiable need for self-reflection or my love of a good drink that makes me think that though…

---Leavenfish

[CrisBrown]

Hi David,

A simpler mathematical answer uses equity:risk ratio. The net equity for this bet is $0. That is, the steal equity is +$8 (20% chance at $40), while made and draw equities total -$8 (80% chance of losing $10). (I'm ignoring the negative equity -- that you may bluff off more money in future betting rounds -- because the problem implies that this is a "one-and-done" bluff.)

The bet, and thus the risk, is $10. The equity:risk ratio is 0:10. You are risking $10 on a bet that stands to make zero long-term profit. For that reason, you should pass.

Cris

[AJo Go All In]

[ QUOTE ]
Hi David,

A simpler mathematical answer uses equity:risk ratio. The net equity for this bet is $0. That is, the steal equity is +$8 (20% chance at $40), while made and draw equities total -$8 (80% chance of losing $10). (I'm ignoring the negative equity -- that you may bluff off more money in future betting rounds -- because the problem implies that this is a "one-and-done" bluff.)

The bet, and thus the risk, is $10. The equity:risk ratio is 0:10. You are risking $10 on a bet that stands to make zero long-term profit. For that reason, you should pass.

Cris

[/ QUOTE ]

for those keeping score at home, this is of course another way-offbase-but-almost-sounds-reasonable post by CB.

the original question said GREATER than 20%. so the bet is not 0 EV. it is in fact POSITIVE EV. and, given the choice between betting here, or not being in the hand at all, YOU SHOULD BET.

david's point is that it may be the case that checking has a GREATER EV than betting here. and, in fact, it does, assuming you will steal the pot (20+epsilon)% of the time, where epsilon is infinitesimally small. so, the EV of betting will be infinitesimally small, where clearly checking has a finite positive EV.

[CrisBrown]

Hi AJ,

Why not keep the personal insults in the WPT forum. [img]/images/graemlins/smile.gif[/img]

That having been said, I was responding to why 20% was not the "magic number." That is, at 20%, your steal equity is zero, so at some tiny amount above that "magic number," your steal equity doesn't justify the risk. Instead, as David said, your "magic number" must be a bit higher than 20% to make bluffing-with-a-marginally-greater-than-X% clearly better than checking.

Cris

[AJo Go All In]

[ QUOTE ]
so at some tiny amount above [20%] your steal equity doesn't justify the risk.

[/ QUOTE ]

read my post again. this is precisely where you are wrong. at some tiny amount above 20%, given that your only options are betting, or not being in this hand:

The bet is +EV.
It is a good bet.
You should bet.
Bet.
Kindly bet.

any of these will do.

[CrisBrown]

Hi A.J.

[ QUOTE ]
given that your only options are betting, or not being in this hand

[/ QUOTE ]

Yes, if it were a bet-or-fold situation, betting is the better play. But the point of the question was that there was another alternative (checking) that had better equity than the minimal-equity bet. In order for betting to be a better alternative to checking, the magic number has to be more than the zero-equity 20%.

The point of the question was that you need to look for the play which offers the best equity and/or equity:risk ratio. (There are situations when the better play is based on the equity:risk ratio, but those tend to be risk-of-ruin situations.)

Cris

[garyc8]

In fact, when the steal chance is exactly 20% there is a very small positive EV based on the chance that you are called by a better hand that gets counterfeited by the turn and river.
However, David's point is well taken. I whiffed on this one.

[donger]

So does this basically boil down to the concept that you should not bluff in close situations when checking still has a possibility to win you the pot? Am I understanding this correctly?

[CrisBrown]

Hi donger,

My answer (not David's) comes down to: don't risk money on break-even propositions.

Let's say I make a pitch to a CEO for a new business activity. The start-up cost is $100,000 and I show that, over the long term, this new business will simply break even, repaying the start-up cost but showing no profit at all. Do you think any rational CEO will back my idea? Of course not. If he's going to risk $100,000, he'd like to know there's some profit in it.

The same applies at the poker table. If you're going to put chips out there, you want a +EV situation, not a zero-EV situation. If the best you can hope for is a zero-EV situation, you shouldn't risk the chips.

Cris