Friday, December 16, 2005

Bring on 2006, Baby

I basically finished up any and all work I have remaining for 2005 yesterday, so we're looking at four straight work days at the monkey factory with absolutely nothing to do, whatsoever, other than to occasionally bang on these keys in a pitiful attempt to simulate actual work.

Abandon hope, all ye who enter here.

The lemurs bested me last night, but man, there are some soft games at Doyles/Tribeca. It can get low traffic at times, though, especially at higher limits, so it's pretty hit or miss, as far as the quality of games, without a ton of alternatives if you can't find a good table.

I've been experimenting with a much more aggressive style the last few days, which has had mixed results. If you wade through all of the number-crunchin' exercises, the numbers are usually a bit surprising, as far as exactly how often a blind steal needs to be successful for it to be profitable, or how often a turn/river bluff needs to be successful, etc. It basically all boils down to the fact that it's good to be the one firing, especially if it's heads up, as most opponents will fold a high enough percentage of the time for it to be profitable, regardless of the actual cards that you both hold.

But we all know that aggressive poker is good, so that's not news, nor particularly helpful. So the real sticking point is correctly picking the times to apply pressure, when to give an opponent who plays back credit for a hand, and myriad other fine details that make poker the wonderful, complicated beast that it is.

Or, you know, you can just ignore all of that and just fire away, at juicy shorthanded tables, with loose passive opponents that like to cold-call too much and call to see the river.

Which isn't a natural style for me, at all, but it's definitely been eye-opening, and a good learning experience, as far as exposing leaks. Far too often in the past I'd fire, miss the flop, fire, miss the turn, fire, then close up shop on the river blank, either check-calling/folding or just checking behind. I tend to selectively remember all the times I couldn't push a calling station off third (or fourth) pair, and let that affect my river play too much, becoming way too passive, especially if I don't have at least A high.

Which is pretty horrible play, given the size of the pot at that point, and just how few pots I need to steal with one more bet, when an opponent folds a winning Q high hand. Especially when playing with loose passive opponents, who'll call down to the river because, hey, who knows, maybe they'll pair that J4s.

You're playing $10/20 6 max, and everyone folds to you on the button. You raise with J4o. SB folds, BB calls.

Flop is Q, 2, 7 rainbow.

BB checks, you bet, BB calls.

Turn is 10.

BB checks, you bet, BB calls.

River is 5.

BB checks. There's $105 in the pot. What do you do?

Or, more accurately, how often does the BB need to fold for a bet to be profitable? Guess what percentage that'd be, then we'll look at some maths.

A few quick ground rules, as far as number crunching. Since it was a rainbow flop with no obvious potential straights, we're going to ignore the possibility that the BB was on a straight/flush draw, and that our chance of winning with J high is 0%, so we lose every time the BB calls and we only win if the BB folds to a final river bet. If the BB check-raises, we fold 100% of the time.

If we bet on the river, it's a $125 pot, in which we've invested $70. Every time the BB folds to a river bet, we net $55 in profit. Every time the BB calls or raises, we lose $70.

If the BB folds 80% of the time, net profit is $30/hand.
If the BB folds 70% of the time, net profit is $17.5/hand
If the BB folds 60% of the time, net profit is $5/hand
If the BB folds 50% of the time, net profit is -$7.5/hand

Which is interesting, really, and not quite as bad as it seems at first glance, if you pull back a bit. One thing to keep in mind is that is pretty much the worst case scenario for your steal attempt from the button, since you didn't take it down uncontested pre-flop, on the flop, or the turn. Even then, all you need is for the BB to fold 60% of the time for an additional river bet with just J high to be correct.

One final thing to ponder is that at the point the BB checks to us on the river, we've already invested $50 in the hand. We can only win by betting. If we check or fold, our net profit is -$50/hand. If we bet an additional $20 on the river, we only lose more than $50/hand if the BB calls ~85% of the time. So we only dig a deeper hole for ourselves by investing the extra $20 on a river bet if the BB calls ~85% or more of the time when faced with a final river bet.

And it's also worth pointing out that the board actually isn't the best for you, either, as far as inducing a better hand to fold. If the river card puts up four hearts on the board and all you have is middle pair, no hearts, can you make that call on the river, faced with a bet? Can you make the call with A high, no hearts? Can you make the call the 50-55% of the time that you need to do to correctly punish the button in that situation?

Stepping back from the raw numbers, put yourself in the BB shoes. Even if you know the button is aggressive with steal attempts and call with a hand like K8o, how often are you really able to call that last river bet? Can you really consistently call it 50-55% of the time, which is the frequency you need to call it with to punish the button for his aggressive play? Can you even call it to the river, faced with a turn bet?

Which is my typically long-winded way of pointing out a very simple fact: I've been abandoning an aggressive stance too often on the river, when faced with an opponent that simply calls me down to that point. Remember, if it's heads-up, most flops (and boards, to some extent) don't hit either player hard. While you may think that the necessary 60% success rate of a river bet causing a fold is too high, the simple truth is that the majority of players won't call you down often enough on the river to punish you. And it's a truth you don't realize (or at least I didn't) until I willfully forced myself out of my comfort zone and started flinging chips.

That said, there are obviously situations where you need to deviate from that. If an opponent has proven that they'll never, ever fold to a river bet, only a very dumb monkey would continue to fire at them. You're not operating in a vacuum.

While it seems a little counterintuitive at first, I would wager that most successful ultra aggressive players actually, on average, pay much closer attention to action at the table, and are much better readers of players than their more traditional tight, selectively aggressive counterparts. Yeah, on the surface it merely looks like they're on autopilot, bombing pots at every opportunity and simply running over opponents, if you dig a little deeper you'll start to recognize that it's only against certain opponents that they behave that way.

So yeah, interesting stuff, despite spewing a few too many chips the last few days. It's nice to engage the brain again, and I'm liking the fact that I'm forcing myself to pay more attention to the actual tables, and think things through and really question some of my ingrained behaviors, instead of just grinding out hands on auto-pilot like I've been doing of late.


cc said...

Very interested in this scurvy. Lots of words and good analytical stuff, but what I want to master (and you may be moving toward) is to have the ability and lack of fear to flip the aggressive switch rather than to either revert to one of two behaviors: pretend poker (not able to read the strength of the guy or lady in position who has you beat and won't go away) or fraidy-scared poker (as you've described). Your analytical work, in so many words, gives some good evidence for a nice way to play occasionally, aggressiveness will win a good chunk of change immediately, staying aggressive will get some more chips, getting away from strength loses less than what is won from the earlier times, then the occasional time when you hit the runner-runner off your aggressive play drives the rest of the table absolutely nuts. Keep me posted.

Meek said...

Scurvy, how about a different way of examining this mathematically. Only look at the river bet. The bets that have already been called are considered 'sunk.' Then it's a $20 bet with a chance to win the $105 pot. At that point, you are getting over 5-1 on your money. If he calls less than 80% of the time, That bet is profitable.

I've seen the return/probability looked at this way on each street instead of as a total. What are your feelings on that perspective?