Player Discussion

[El Diablo]

[ QUOTE ]
My answer was that if someone was all-in, the jacks would be better, but if they each had chips left, which was the case here, the ace-queen suited was better.

[/ QUOTE ]
[ QUOTE ]
The question is regardless as to whether it is right or wrong to call the raise, if you do go ahead and call it, which hand would you now rather have.

[/ QUOTE ]

[Edited to add: As I read through the thread, I notice that slong already did the second calc I present below]

Mason, given your assumptions about how the players play post-flop:

Numbers used below are:
[150 = money put in preflop.
165 = money opp put in preflop + 15 blinds.
225 = money put in on flop.
67/26/7 are from Mason's post re: how these players will play post-flop]

67% of the time, AQ loses 150
26% of the time, AQ wins 165+225 = 390
7% of the time, AQ loses 150+225 = 375
= -100 + 101 - 26 = -25

67% of the time, JJ wins 165
26% of the time, JJ loses 150+225 = 375
7% of the time, JJ wins 165+225 = 390
= 110 - 98 + 27 = +39

So, this seems to me like given the scenario (JJ reraises and AQ calls reraise), I would want JJ.

Now, as the question is framed in terms of all the pre-flop already having happened, we should really consider the 315 in the middle dead money - so "if you[AQ] do go ahead and call it, which hand would you now rather have?" - AQ or JJ, given the 315 dead in the middle.

So now:

67% of the time, AQ makes 0.
26% of the time, AQ wins 315+225 = 540
7% of the time, AQ loses 225.
= 0 + 140 - 16 = 124

67% of the time, JJ wins 315.
26% of the time, JJ loses 225
7% of the time, JJ wins 315+225 = 540
= 211 - 58 + 38 = 191

I did not check my math and I did taste 12 scotches earlier tonight, so perhaps I've done something wrong here, but it looks like whether I look at this in terms of the whole hand start-to-finish or in terms of just which hand is better to have post-flop with chips left given the pre-flop scenario that just occurred, Jacks seem better using the assumptions about how these players play that you stated.

Now, I think the situation is far worse than this, as AQ will often pay off far more on A or Q flops and will get stacked by JJ far more than JJ gets stacked by AQ. And other considerations like JJ blowing AQ off best hand more than AQ blows JJ off best hand.

But, ignoring the other considerations I feel are important, I still can't see why, using all of the assumptions you state about how the players will play, you prefer having AQs to JJ.

[excession]

'the jacks would be better, but if they each had chips left, which was the case here, the ace-queen suited was better...'

that seems to be going a lot further than just saying that AQs player 'has profitable strategies available' - in effect you're trying to prove mathematically that (a) AQs OOP in a re-raised pot when you aren't the re-raiser is better than (b) having JJ in position in a re-raised pot when you are the re-raiser. The trouble is empirically that this isn't the case, as I'm sure all our experience and PT stats would clearly demonstrate..so something is wrong..

Further you're not taking into account number of players, reads, betting history, actual position or stack sizes...so any NL player is immediately suspicious that you simply don't have enough information in the system to draw any relevant conclusions

[psuasskicker]

When I worked for Northrop, and we were interested in blowing up the Soviet Union, I used similar techniques many times, and we weren't concerned with no limit poker, but we were concerned with real no limit.

That just leaves me with a real warm and fuzzy feeling all over. [img]/images/graemlins/smirk.gif[/img]

This is a very good way to analyze many statistical situations.

In general I think this is a pretty good way to put analytical rigor behind lines of thinking. I don't disagree that one can try to do that with problems such as these.

The problem is that in NL I feel it's too short sighted. In limit, the analysis fits much better, where you will win or lose a set amount based on the number of bets that will go into the pot.

In NL, you're putting your entire stack at risk in every hand you're playing, which means those sort of numbers wind up going out the window in scenarios like this. The problem I have with it is that you don't know what the guy with JJ holds. So even if you flop a pair, after having faced a reraise, you're left in a situation where you're just not going to be all that comfortable.

I just think it's too limiting to say he's going to bet 2/3 pot all the time, and will never put in a second bet unless he's tough which will be offset by the times you can flop a draw to continue with. I think it makes sense to put analytical rigor into many things, but this problem is simply too complex to boil down to the rather straight-forward way it was done in this analysis.

Another issue I take with the analysis is that it seems to contradict your earlier thinking, where you said it would be better to fold before the flop; but then your analysis seemed to show that it was +EV to actually make the call. I know which one of those two I believe to be correct, but I'm not sure anymore which you're advocating.

And if you're advocating folding as the correct play, then I don't see how your analysis can be considered relavent for the initial question (since you're saying calling is better). The actual initial question you were trying to answer, which I thought was basically "which would you prefer (which hand is better) after taking the flop, with chips remaining in one's stack?" Maybe I'm confused on the question?

No. I'm not advocating any strategy. All I'm trying to do is to show that a strategy exists for Player A who holds the ace-queen suited that will produce a positive expectation.

That strategy seems to be check calling one bet and folding to any second betting. And it only considers one pair situations, and assumes villain will always act in the same way. I think it's far too simplistic for this situation, and advocates a less than optimal strategy after facing a reraise.

As I state in one of the posts above, I analyzed a more simple problem and then made inferences about the more complex problem.

This is basically the heart of why I think this is much more applicable an analysis in limit than in NL. In limit, there's a defined amount you can win or lose in a given hand. In NL, with your stack at risk, you're left in a situation where the betting on the next few streets can be much higher or lower.

So for example, take this problem in a 20/40 limit situation. You bet on the flop and get raised. The most you can lose by calling down that $20 is another $100 total (including the $20). So you can do an analysis based on your betting and getting raised some times and calling down from there much easier, cause you know the bet puts out $20, and the call down is another $120 in a defined size pot.

But in this scenario, you're left with some amount of money behind your bet sizes. You note the $225 bet, but with another ~$750 behind this, the analysis can go pretty haywire quickly. And it gets worse the deeper your stack.

That's why I think this is too complex to try to tackle by simply showing one level of analysis, then applying a few general principles/assumptions to the variables behind them. The variables make a huge difference in the outcome of the hand, and I don't think it's easy enough to simply say "it still seems to me that A's expectation remains positive" and "I'll contend, perhaps incorrectly, that [other scenarios] don't occur that often to impact the overall expectation significantly."

Just my .02

- C -

[kagame]

mason,

i love you so much. no seriously, thanks for replying to my posts buddy.

[nigelloring]

[ QUOTE ]
[ QUOTE ]
Mason,

Please explain your original clain that E[AQs] > E[JJ].

Matt

[/ QUOTE ]

I was just going to ask the same thing. You claimed to make a statement and said other players in the game were bad because they believed the opposite. What is your position and how are you backing it?

[/ QUOTE ]

Yes, please do this, Mason. If I were you, I would either:

(1) Call - demonstrate E[AQs] > E[JJ] in some reasonable fashion or
(2) Fold - admitting you are wrong and cutting your losses.

I'm not saying either of these actions is necessarily right for you.

To me it seems you are reraising with air - refusing to admit you're wrong and getting around demonstrating E[AQs] > E[JJ] by refusing to entertain arguments around it. It's very nice to see that you're responding to so many people, but I for one cannot see where you have addressed this point.

If you are attempting a proof by verbosity (http://en.wikipedia.org/wiki/Proof_by_verbosity), I don't accept it as valid. If you're not, please set me straight.

In my opinion, your reasoning as to E[AQs] > 0 may or may not be valid, but by itself does not show that E[AQs] > E[JJ].

In fact, as others have pointed out, since your reasoning shows E[AQs] < half-of-money-in-pot, I draw the conclusion that E[AQs] < E[JJ].

Thank you.

[Matt Flynn]

[ QUOTE ]
I believe he said the players were bad because they didn't understand the discussion, not because of their opinion on it.

[/ QUOTE ]

From Mason's original post:

[ QUOTE ]
Then someone said, "Let's ask Mason since he writes all the books." My answer was that if someone was all-in, the jacks would be better, but if they each had chips left, which was the case here, the ace-queen suited was better.

Well, no one understood what I was talking about. No wonder the games are good.

[/ QUOTE ]

[SunOfBeach]

Just when i thought that this thread was all but over...

[ QUOTE ]
If you are attempting a proof by verbosity (http://en.wikipedia.org/wiki/Proof_by_verbosity), I don't accept it as valid. If you're not, please set me straight.


[/ QUOTE ]

now THIS is funny...

[SunOfBeach]

[ QUOTE ]
all I did was attempt to show that there is a strategy available to Player A who holds the ace-queen suited that can show a profit in this situation.


[/ QUOTE ]

which begs the question, "why on earth are you doing this"? surely that does nothing towards defending your original hypothesis.

[SunOfBeach]

[ QUOTE ]
When I worked for Northrop, and we were interested in blowing up the Soviet Union, I used similar techniques many times, and we weren't concerned with no limit poker, but we were concerned with real no limit.


[/ QUOTE ]

now this is just annoying... kind of like a pompous version of the "i'm smarter than you, and this is old hat compared to what i've done in my life, so you just should accept that i'm right" defense.

as an MIT-trained mathematician with vast experience in game theoretic application to national defense, i can assure you and everyone else that:

1. ability to peform gt analysis of natdef issues is in no way relevant to the discussion at hand.

2. these issues involve far more complex mathematics than the high school-level algebra being used in this discussion.

3. no limit war strategies are far different (and must be analyzed in a different way) than no limit poker strategies.

and while we're at it:

4. showing that EV(AQs)>0 in some situation surely does not show that EV(AQs)>EV(JJ) in that same situation.

5. anyone who once advised on or analyzed anything dealing with nuclear proliferation is quite clear on #4, and is hence ignoring the original question at hand because he now knows that he is wrong.

[mikech]

i figured it out! i did an ev calc and finally solved this problem! first of all, we have to adhere to the scenario mason described. forget about the preflop call AQs has already made. given that there's $315 of dead money by default in the pot, AQs has HUGE ev from the flop onward:

AQs will put money into the pot when behind only 7% of the time! and only $225 when he does so. HOWEVER, he'll be ahead 26% of the time, in which cases he'll win the $225 AND the $315 in the pot! the other 2/3rds of the time he'll simply fold and not lose another cent!

[ QUOTE ]
When I worked for Northrop, and we were interested in blowing up the Soviet Union, I used similar techniques many times, and we weren't concerned with no limit poker, but we were concerned with real no limit.

[/ QUOTE ]
i'm amazed any of us are even alive at the moment.