Overpair on Paired board - RANT

[Heisenb3rg]

I am sorry, I really dont understand why calling down is correct here.
I also raise the flop.

I haven't heard a single argument as to why calling down is correct either other than "it just is."

Clearly you're getting 3-bet when your behind, so you lose an extra SB. But if he has the 9 infrequently, this isnt a concern.

I can't see any metagame reasons for not wanting to raise either. Are you trying to encourage him that "passive play is good, look im passive too, continue to be passive"?
While every other player is raising the [censored] out of him?

Im extremly curious as to hear the reasoning behind ILP's call down here.

[ILOVEPOKER929]

[ QUOTE ]
WOW. I read the first two pages and had to stop.

The argument is [censored] ridiculous. I'm sick of hearing in the long run and always do this.

It's bullsh i t. Play the [censored] player.

[/ QUOTE ]

Kwaz, We cant play the player, all we have is a stat read, we have no history with the villain, we have no idea how he plays. In situations where you have no read playing your hand in such a way that makes the most money possible in the long run will be the optimal line.

For all those people out there who are sick of my esoteric ranting, I promise you and miles I will explain how calling down can make more money than raising the flop even if we assume that the villain will have a hand like 5x the majority of the time. I cant explain it now cuz I havent slept all night and im very tired, I will try to write my explanation some time later today.

Screw it, I'll just explain it now. To best explain my point I will construct a simple model where the hero has QQ and the flop is 995r and the villain donks. First I will list a few assumptions.

1) The villain has 3 types of hands, he either has a hand like 5x, (JJ or 22 is also lumped in this 5x category to keep things simple), Or the villain has a better hand than QQ, basically KK's or better, Or the villain has nothing, basically a no pair no draw type hand.

2) If the villain has a hand like 5x, and the hero raises the flop, the villain will always call down. Under this scenario the hero will make 3 big bets.

3) If the villain has a better hand than QQ, the hero will not be able to fold no matter what line the villain takes since the hero has no definitive read. To make this model as simple as possible we will assume that if the hero raises the flop, the villain will always 3bet and bet every subsequent street. So the hero will always lose 3.5 big bets.

4) If the villain has nothing, and the hero raises the flop, the villain will fold the flop 50% of the time, and the other 50% he will call the flop and fold the turn. The hero makes an average of .75 big bets

5) If the villain has 5x and the hero plays passively and calls all streets or bets when checked to, so the hero will always make 2.5 big bets.

6) If the villain has a better hand than the hero and the hero plays passively by calling down, the hero will always lose 2.5 big bets.

7) If the villain has nothing, and the hero plays passively , the villain will always bet the flop and the turn hoping the hero will fold, but the villain will only bluff the river 50% of the time. Under this scenario the hero makes an average of 2 big bets by playing meekly.

8) To keep this model as simple as possible all events when the villain improves to a better hand or the hero improves, are purposely ignored. Assume both events cancel each other out for the sake of simplicity.

Now heres my very simple model. When the villain donks the flop assume he will have a 5x type hand 60% of the time, a better hand than the hero's 20% of the time, and nothing 20% of the time.

Doing some quick math, the expectation of raising the flop is (.2 x -3.5BB) + (.6 x 3BB) + (.2 x .75) = 1.25BB's

The expectation of calling down = (.2 x -2.5BB) + (.6 x 2.5BB) + (.2 x 2BB) = 1.4BB's

So based on this model, playing passively will make the hero
.15BB's more in the long run than raising the flop. Also keep in mind other benifits to a taking a passive line: 1) The hero will now always be in a position to raise the turn or river should he improve to a full house.
2) If the villain some how improves to a better hand than the hero on the turn or river, the hero can never get checkraised since the villain is the one with the initiative.

How much value we should assign to these two benifits in terms of big bets is hard to say, but the value is certainly greater than zero, and therefore is another positive aspect of taking a passive line.

So now that this model has shown that a passive line can make more money than raising the flop even when the villain has a 5x hand the majority of the time, does this mean Im right and other 2+2ers like Miles are wrong? Not at all, this model doesnt prove anything. In fact all one has to do is change the parameters around and raising the flop will clearly be the best line. I'm sure if we assume the villain will have a 5x type of hand 90% of the time, a better hand 5% of the time and nothing 5% of the time, raising the flop will be the clear money making play.

I created the 60/20/20 model just to show people that raising the flop can still be wrong even if we assume the villain will have a 5x type hand the majority of the time. However, this was not done haphazardly. For this model is roughly equivalent to the mental model I use when I actually play poker and am confronted with this type of situation. For those who care I often use a 40/30/30 mental model online, but thats really not important, for the sake of this discussion assume I use the 60/20/20 one. People may disagree with the assumptions of my model, but atleast they will no longer wonder why I do what I do. I call down in this spot becuz my mental model says that calling down will make more money than raising the flop. Also, if people are wondering where I came up with the idea that taking a seemingly unorthodox passive line can possibly make more more money than an aggressive line. I learned this concept from David Sklansky and Mason Malmuth. Reading their ideas on handling flop donks changed my game dramatically, and im happy with the results.

Also remember, My model assumes no reads. A real read can change everything. For example if we knew the villain would 3bet the flop or checkraise the turn with many lesser hands, then raising the flop will likely be the best line.

[Heisenb3rg]

Hey, great post ILP!
Definitly gave me some new ideas, thanks for taking the time. I defintily like the idea about initiative if you improve vs he improving , never really thought about that.

I think that against a semi-agressive player my reflex would be to raise the turn, but your post made me reconsider it to actually be a call down. If he's VERY agressive, it's a wait to river to raise too.

I think your assumptions apply much more accuratly to a semi-agressive player than a passive player though.

I think it's very unlikely a passive player has air here, (much less than 20%)
I also think he's not too likely to check-raise if he has a 5x hand that improves, because he's shown he's inclined to donking rather than check-raising.

The very key part to making call down correct is the amount the opponents bluffing. Reminds me of the advice sklansky gives in HPFAP when you hold AK and the opponent flop donks you on an a-x-x rainbow flop.

Great post.

[milesdyson]

keep in mind i already did this math for a more realistic set of numbers. saying that this guy has air on the flop 20% of the time, bets the turn with it 100% of the time, and 3-barrels the river 50% of the time after that is pretty funny.

also, according to this 60/20/20, he is BET/FOLDING the flop a whopping 10% of the time. honestly, he is not taking this action 10% of the time. he will call the flop. and of course, 60% is just self-servingly low.

...and this is what i was pissed about. "solve the puzzle," instead of meaning, "no, there really is something interesting going on here," actually means, "miles, take that math you already did but plug in completely unrealistic numbers to make the math come out in my favor. once you have done this, you will solve the puzzle."

i wonder what numbers you started with before you realized how much you needed to change them.

[Hobbs.]

ILP,

nice post. Like I pointed out earlier in this thread the main disagreement is due to assumptions we are making about weighting the relative likelyhood of villains holdings on the flop. Anyway, my experience says an unknown has a 9 < 20% of the time, but whatever, to each his own.

[ QUOTE ]
Also keep in mind other benifits to a taking a passive line: 1) The hero will now always be in a position to raise the turn or river should he improve to a full house.
2) If the villain some how improves to a better hand than the hero on the turn or river, the hero can never get checkraised since the villain is the one with the initiative.

How much value we should assign to these two benifits in terms of big bets is hard to say, but the value is certainly greater than zero, and therefore is another positive aspect of taking a passive line.

[/ QUOTE ]

I disagree with this as we still have a right to raise later in the hand if we improve with the aggresive line AND were 3-bet on the flop. It is also a lot less likely that villain calls down a turn or river raise the times we have spike a Q for example. Anyway, it's not clear if a passive line really does imply that we stand to make more the times we improve.

Also, our equity when against a 9 is small, but not insignificant and will improve our EV for raising slightly.

[MATT111]

ILP: You see that if your numbers were accurate (which I doubt) raising the turn is equal in EV to calling down? (If we always call down a 3-bet and are never ahead)

[ILOVEPOKER929]

[ QUOTE ]
ILP: You see that if your numbers were accurate (which I doubt) raising the turn is equal in EV to calling down? (If we always call down a 3-bet and are never ahead)

[/ QUOTE ]

Matt, my numbers are 100% accurate, I promise you that. Ive done these calculations many times over my poker career assuming differernt opponent types. You can question the many assumptions of my model, or my perception of this situation, but my numbers are fine. About raising the turn, according to the actual model I use online (40/30/30), calling down makes me .35BBs more in the long run vs raising the turn, but even if raising the turn has the same EV as calling down, calling down would obviously be the superior line since it has a much lower variance than the turn raise strategy. The point of my 60/20/20 model was to show that it is possible that calling down can +EV compared to raising the flop even if the hero has a 5x hand the majority of the time. I'm not really discussing a turn raise strategy here.

Hobbs, the two benefits I listed will always be true when we have position and take a passive line. There's nothing to disagree with here. Those two benefits are an inherent advantage of having postion and are some of the many reasons why having position is so valuable. If we raise the flop, most of the time the villain will go in to call down mode if he doesnt have the hero beat. Therefore by raising the flop we will often lose out on the chance to raise the turn or river should we improve and we do expose ourselves to a checkraise on a latter street. If you improve to Queens full and you think the villain may not call your turn raise, then wait til the river. [img]/images/graemlins/smile.gif[/img]

Heisenb3rg, You are correct that a different opponent type would require a different model. You are also right that a very key part to making a call down correct is the amount the opponent bluffs. I believe online opponents bluff alot more than live opponents. This belief influences my model greatly. About youre statement "I think its very unlikely a passive player has air here" Becuz I think you may be interested, when Im playing poker in a real life low limit game where players tend to be passive and timid, I use a whole different model against an unknown. I usually assume they have a better hand than me 50% of the time, a 5x type hand 40%, and nothing 10%. Using these numbers calling down has a positive expectation of .425BB's over raising the flop assuming we dont have a good enough read to fold on any street. What this implies is that if you are playing against an opponent online that you think is passive, calling down has even more value. Raising any street in this circumstance would be borderline foolish. Notice how passive the villain was in Mossbergs thread. Perhaps raising the flop was an even worse idea than I thought.

Miles, If you dont think my numbers are realistic that's fine, as long as you know where im coming from then I have accomplished something. On this specific flop and this specific betting sequence I do believe the villain will be bluffing very often. I also would say that I actually think the villain having 5x 60% of the time is too high. In my experience online, betting this flop with 5x is actually a very unnatural play for most players. It seems our perceptions are different on this idea and there's no need to discuss this idea further. Also, if the villain has nothing, I would expect him to fold to a flop raise a significant percentage of time. I dont really see how this is a controversial idea. Once again though our perceptions differ and theres no use banging our heads against the wall any longer. Miles, the puzzle was figuring out how it could be possible that calling down will be +EV over raising the flop even if the villain has a 5x hand the majority of time. As youve seen in my post this morning, I have already solved the puzzle for you. You dont need to agree with my assumptions, all you need to understand is that it is theoretically possible that calling down will make more money than raising the flop in that scenario. I already know you understand so im not gonna talk about that any more. About your statement "I wonder what numbers you started with before you realized how much you needed to change them." I change my numbers all the time depending on whom I'm up against.

If you would like to read more on handling flop donks in these kind of situations. David Sklansky and Mason Malmuth discuss this situation in HPFAP and Mason talks bout it more than once in his "Poker Essays 3" book.

[philnewall]

Raising is obviously better than calling down..assuming it can be done so "safely". That is..his reaction to the raise musn't cause us to make an error.

I prefer waiting for the river to raise..inducing bluff bets on turn and river. It is also an easy fold to a river 3-bet which is probably not the case on other streets. End of.

[ILOVEPUZZLES995]

The jig is up!

[Guruman]

guru was here. [img]/images/graemlins/smile.gif[/img]