what about the calls you save by not being bluffed at? what about POTS you save by not folding to bluffs because you get checked to w/ hands you might fold in marginal heads up river confrontations?

-Barron

Notice you don't support the notion that the EV is -1 BB. The pot is laying you such a price that if you lose 94% of the time or less it's +EV. You claim that the overcall will lose at best 98% of the time. How accurate are your estimates and how do you know how accurate they are? Personally I don't think mine are accurate enough to be precise in distinguishing between 94% of the time and 98% with any degree of certainty. But I'm a pedestrian player.

It was three way. There was an over call. They will never know what you had.
There is no reason to call for meta-game reasons here.

That's right. I think hero will lose 98, 99 or 100% of the time. He's not winning 90% or 95% of the time here.

EDIT: you asked how confident.
I'm 95% sure he will be beat 98-100%.
5% of the time will be beat 90-95%.

You didn't answer my question though. How accurate are your estimates and how do you know how accurate they are? I've asked this question for years when people make these claims that they can be highly accurate in their estimates of these situations where the pot is laying them a big price. I've never gotten straight answers to these seemingly simple questions. The truth is you probably can't distinguish to an degree of certaintly between 95% of the time and 98% of the time.

not when James coldcaps from the SB and bets every street.

Adios, I have a strong suspicion that if you check James's PT database and filter for those times that he cold capped out of the blinds with >3 people in the hand, you will find exactly ZERO times that he will lose to AJo on this board.

He will NEVER thats right never have KQs or ATs, and if he had those hands he would also very rarely bet the river against this lineup.


Overcalling the river here has EXACTLY -1 BB ev. AJo will never ever win with this specific action.

I suggest that there always is at least some uncertainty even though if it's 1,000,000 - 1 against. It's not a cinch that the Sun will rise in the east tomorrow morning either .

Let me say that I respect your point, I respect the way you presented it and FWIW I think it's a valid one. I will say though that I'll bet that the number of times this situation arose is small i.e. the sample size is small.

well thanks, and i suppose there is always some possibility of a misclick

Uh huh. AJo is a loser 100% of the time against James and a caller here with this action. If you're talking about metagame, then you're very very wrong.

Hey Andrew - you are saying that when I cap out of the small blind and lead every street into multiple opponents, even regardless of their action, there is a greater than 0% chance that AJ unimproved is good on the river? Nope, their isn't. AJo is good exactly 0% of the time here. Therefore, the expectation of this call is -1 BB.

Oh yeah, let me know when you scrape together your evidence of people with 30% VPIP in very tight and aggressive games who are big winners, I'm excited to see the data.
-James

There is not a 1 in 50 chance that he is good here. I would never cap worse than AJ here on purpose. If you forced me to go out and figure exactly how many times he was good I would say roughly 1/100,000 because I probably misclick once every 75,000 clicks and then throw in that someone has already called and we start to get closer to the probability. So maybe he only loses 59 dollars and 96 cents on this call at the end and we were incorrect.

Also, Barron, you know better than to site metagame conditions here. It's not like he flashed his hand at the end and said "woops" or something. I had to check the HH. And regardless, he has called down in a number of situations where one more time making a stupid call down is not significantly contributing to his future EV.
-James

With the hypothesis that the topic of this thread is in fact a big winnier with 30% VPIP, have you worked out the P value that the hypothesis is false? You know, you can use Baysean Statistical Inference to blend expert oppinion and empiracal data (though as the amount of data rises, the influence of the expert oppinion on the interpretation shrinks).

We are not dealing with pure math here. Since humans are involved, this is a question of science. To explain that: In math you assume certain truths and then derive implications. You can show expicitly that certain things are true given the assumptions. In science, you formulate threories to explain observations, then make novel predictions based on the theories and test those predictions experimentally. Then you statistiacally evaluate the results of those experiments to determine the level of confidence you have in the theories.

In science, there are no absolute laws, there are only theories which you have confidence in. (For those who say "Well what about the Law of Gravity??" It is the theory of gravity and the "Law of Gravity" (Fg = G*m1m2/r^2) does NOT reflect our current theory how gravity works, though it is a good enough approximation in many cases.)

The whole point of me (futilely) chiming in here is that there is a lot of oppinion and intuition being bandied about with bravado. We have numbers, we have data, why are people arguing oppinion instead of doing math?

Hey fnord - your point is well taken, but I think we all agree that this would be(if he were in fact a loser), statistically speaking, a very very rare anamoly. There is no arguing with that. Math tells us that with those stats he should almost certainly be a winner, but we are trying(probably unsuccessfully) to have a discussion about whether a player like him can win under the assumption that he is the proverbial "lottery winner."
-James

yea i forgot a) it was multiway, and b) he just does this all the time and doesn't seem to care.

-Barron

Hopefully Barron will be able to do the statistical analysis he suggested soon. I don't know how unlikely the stats are; if it turns out a losing player with his observed mean and SD has 5% chance of hitting these numbers over the sample size I would be surprised (it was that high) but not shocked.

If it turns out that he is likely a winning player despite his unorthodox style, there is just a ton of analysis that would be interesting. One thing I would like to see in that case is how he does against specific profiles HU (e.g. TAGs, LAGs, etc.), how players numbers change when he is in a pot (that is, how does the mere fact that he is in the pot change the dynamic of the game), how he does in multi-way pots, etc.

Here's one non math thought I had about some of the apparently poor post flop play on some hands: He may make some plays purely heuristically. For instance, maybe he makes a river overcall that is clearly -EV given the opponents and the fact that he knows the opponents because he is involved in hands on two other tables at the time, it's his action, and he sees a large pot and a hand with some apparent show down value. That is, maybe some of the blatently bad plays are due to the multitabling effect.

If the chances were anywhere near 5% we would have come across a zillion of these guys by now.

Finally a voice of reason in this argument. Overcalling this river vs. James he is going to be losing exactly 100% of the time. If you disagree, tell me one hand, just one, that James would ever play this way that loses to AJ. Not to mention the person in between. I would say AJ is behind to James' and James' hand only 100% of the time here.

Although I'm sure you realize this, it's probably worth emphasizing that the first and third possibilities are not mutually exclusive. From my reading of your statement, you feel 80% confident that not only is his win rate unsustainable, but that he can't beat the game in the long run.

An interesting thing is that, even if he can't beat the game, the expected time it would take for his winrate to reach 0 is probably so long that he could go a very long time believing he is a winner. Of course, by the description of this opponent, it's very possible he will start tilting as soon as things go against him.

I also thought that I would point out that there's a lot of good that can come out of investigating why this particular play has such a large winrate.

An interesting project would be to try to find an extremely weak-tight player whom people have a large number of hands on, and see how this player fares against the person under discussion.

The more that this player is able to run over weak-tight players, the longer variance can sustain an abnormally high win rate.

It will be fun to publicly embarrass you when this guy loses 60k back in a month. He's probably half way there. These quotes will be fun to dig up.

I think he has embarrassed himself enough already

BicycleKick hacked my password.

And what about the possibility that poker results do not follow a normal distribution, but rather something with fatter tails. It would make sense that both strongly positive and negative results can feed off themselves, causing tilt in the player or in his opponents.

But he is on Sklansky's top ten smartest list. I wonder how that physics P.H.D. is going?

I like how Andrew points out the he plays "this level regularly and higher."

The obvious translation is that he plays 30/60 and takes shots at the 80/160 game. Which is great and all, but the main posters in this thread, play 1-2 and 150-300 and take shots at 300-600 and higher.

I don't believe he ever made a statement one way or another about whether the guy was a winner or not, just that some people were speculating rather than running the numbers, and that there were holes in the logic underlying the speculations.

If someone presents an invalid argument that something is true, and that something turns out to be true, it does not suddenly make the argument valid.