Pattern Poker

[Tommy Angelo]

"I believe the frequency of specific betting patterns, viewed in isolation, has little to do with their profitability or lack thereof."

Agreed.

"For example, the EV of your "Pattern One" is probably very dependent on the relative frequencies with which you check-call or check-raise the flop heads-up after having open-raised."

Agreed.

"The frequencies with which you take various actions on the turn and river may also interact with your preflop and flop proclivities in complex ways that affect the EV of a particular betting pattern."

Agreed.

"Furthermore, betting pattern analysis which incorporates parameters of hand value would potentially be much more useful than analyses that ommit hand value considerations."

That would depend on the answer to "useful at what?" What if an analysis is defined (in part) by the omission of hand value considerations?

"Other factors, such as player characteristics, can also be thrown into the mix. Of course, this sort of analysis quickly becomes overwhelmingly complex!"

If I understand you right, you are saying that two of the variables that determine the +EV of a betting decision are the cards and the opponents. I agree. :-)


Tommy

[Tommy Angelo]

"I fail to see how raising with the intention of check-folding the flop when called can be profitable"

If it's folded to me in the cutoff, and the button and both blinds are all super duper tight and straighforward.

[ChrisW]

> I check, he bets, I raise. He doesn't have a set with any
> of the big pairs, nor KK, TT, AK. He didn't 3 bet did he.
> If he has AQ, oh well, I'm sure I'll find out before the
> river but, probably not AQ either (no 3 bet).

Most opponents will have two big cards to call preflop here. With a pair, they would either fold or 3bet to get heads-up and take the lead. Thus, IMO it is almost 100% that you are beat. The question is whether you can get him to lay down KJ or KQ.

With those holdings, your opponent will almost certainly call the flop checkraise. Thus, your decision not to fold the flop commits you to bet the turn for a total of two big bets. If your opponent folds the turn, you've won 3.75 big bets, so you'd need to win 35% of the time to break even. Do you really think he'll fold the turn 35% of the time? Remember, he doesn't have to have KQ or KJ; he could have two pair as well.

Assuming you're called on the turn, bluffing the river breaks even if you'll succeed 13% of the time. Of course, your opponent only needs to be right 12% of the time to show a profit by calling you on the river. In a moderate sized pot such as this one, most opponents with a questionable made hand and a weak draw would either fold the turn or call down. So, I don't think bluffing the river will add much (if any) equity.

[Pensive Gerbil]

You seem like a most agreeable fellow! [img]/images/graemlins/grin.gif[/img]

[ QUOTE ]
What if an analysis is defined (in part) by the omission of hand value considerations?

[/ QUOTE ]

It may be interesting to analyze betting patterns without reference to hand values. In practice, however, it is usually useful to consider the value of your hand when making betting decisions. [img]/images/graemlins/wink.gif[/img]

Regards,

PG

[Kevin J]

You'll have to provide some more examples. It's hard to imagine it could ever be wise to develop a patternistic approach to such a dynamic environment as poker.

In example #1, check/folding could be correct. Not because of any pattern, but because of what you know about that particular situation.

Same with example #2. It could be right to fold for one more bet, but I don't see what "pattern" has to do with it. You'd fold because of your assessment of how likely you are drawing dead compared to what this player would check/raise with.

In fact, I see these two examples as being extremely non-patternistic. Alter either situation even slightly, and you might arrive at completely opposite plays.

I realize that you're way over my head and I'm just not understanding what you're getting at.

[pfkaok]

[ QUOTE ]
In fact, I see these two examples as being extremely non-patternistic. Alter either situation even slightly, and you might arrive at completely opposite plays.

[/ QUOTE ]

I'm not sure if this is what he meant, but I got it to mean that the "pattern" here is the fact that very few players would ever even consider taking this line. So its a pattern among most players that if they openraise and get 1 caller, they will never consider the check/fold line, even if that would clearly be the most +EV play. Maybe since its not common for the check/fold to be the best play when HU, most people would NEVER do it, so you could use that pattern to narrow down what course of action they're planning on taking.

Very interesting idea though, I'd be curious to hear more lines that are extremely rare by most players.

[pfkaok]

I'm very curious about this also b/c as a multitabler I'm sure that I fall into certain patterns, and I'm sure that it costs me $$$ at times. Any player who is truly great can look at each situation independantly and figure out what play is most EV, but this is probably a big problem for almost all average to good players.

[Senor Choppy]

[ QUOTE ]
"I fail to see how raising with the intention of check-folding the flop when called can be profitable"

If it's folded to me in the cutoff, and the button and both blinds are all super duper tight and straighforward.

[/ QUOTE ]

Of course there are situations where any play is the correct play, but you didn't say they were super tight in the original post.

I have nothing against the concept or the example, only the lack of relevant details originally.

[Tommy Angelo]

(This post was accidentally prematurely posted so sorry if there is some huge mixup.)

(Thanks to the follow-up posts, I now I have a vague idea of what I was talking about in the first place, which is

"Of course there are situations where any play is the correct play ..."

Bingo.

I'm using "betting pattern" to mean the betting and positional information of one hand and nothing more.

Intead of thinking, okay, here are my cards, here is my position, here are my opponents, here is me. Instead of thinking of those four things are the main variables, and then solving for variables to find the answer to an EV equation, what if we looked at nothing but the bets and position, and solved for cards and opponents?

There can be many solutions. But hey.

For example, I openraise before the flop from the cutoff, one of the blinds calls. He bets out on the flop, and I fold.

Too many solutions. Pointless.

But what if you cook up a rare pattern, or one happens upon you. What I do is I think, "Could this pattern ever be correct for me?"

If the answer is "yes," then I want to be ready to use the pattern when I should.

And what if it's a common looking pattern, like the quick fold on the flop example above. I might think, okay, it is definitely right to use this pattern sometimes, but how often?

Then we're onto larger patterns of these smaller patterns, which isn't necessarily the same as being on to something.

Tommy

[Tommy Angelo]

"I realize that you're way over my head and I'm just not understanding what you're getting at."

I really messed up by not defining "pattern" in the first post.

A "betting pattern" is the betting and positional information of one hand.

Button openraises. Small blind calls and the big blind folds. Headsup. Flop: check, bet, call. Turn: check, bet, fold.

That's a pattern.

You could look any of the three players, if you wanted to, and come up with ranges of hands and opponents and heros for whom each of their bets was correct. Or incorrect. Or plug yourself in as any of the three players, thereby assigning a value (youself)to one of the four main EV-correctness-determining variables (hero, opponents, cards, position).

But what mainly got me thinking about this pattern poker thing is when I'm sitting there at the table after I'm involved in a rare (for me*) betting pattern, and I wonder, could that have been correct for me? and if the answer is yes, then maybe I should be doing it more often.

(Examples: 1) Checkraising the flop headsup. 2) Cold-calling an openraise from the cutoff.)

Tommy