Monday, January 25, 2010

Luckbox...

I did not play a ton of poker this weekend. No action at all on Thursday, as I instead watched the Isles defeat the Panthers in a shoot-out, and then went out for a few brews with a good friend I hadn’t seen since before the holidays. Normal session on Friday, but Saturday was curtailed in light of a continuing education commitment I had to fulfill (for my securities licenses) in the early afternoon at a local Prometric testing center. (I somewhat regret not having attempted to procure the number of the surprisingly attractive, engaging and ringless proctor, who I pegged at mid thirties to early forties, but who wore her tight black jeans, even tighter white shirt and rocker-chick boots extremely well. Perhaps I need to work on improving more than just my poker game lol...)

Didn’t play at all on Sunday in light of, well, a fairly significant hangover. But I did get in a couple of Stoxpoker training videos; one of which was an enlightening piece by Ed Miller describing the 25 stages of a developing TAG. I was pleased to learn that most of the stages involve concepts that I’m currently incorporating into my game. Perhaps I’m further along in my development than I’ve realized…

I was a complete luckbox during my Friday session – considering that of my top 10 most lucrative hands, I was a significant underdog 4 times when all of the chips went in. Three of those cases were fairly academic in terms of the action (in one case, vs. a short stack, I had trip aces against a boat, but fortunately rivered quads; the other two were preflop scenarios - AKs vs. KK, and KK vs. AA). However, the other was an interesting hand. The history follows:

MP: $160.50 (160.5 bb)
CO: $101 (101 bb)
BTN: $38.35 (38.4 bb)
SB: $194.30 (194.3 bb)
Hero (BB): $213.50 (213.5 bb)
UTG: $105.55 (105.6 bb)

Pre-Flop: Hero is BB with Kh Jh
UTG folds, MP raises to $3, 3 folds, Hero raises to $9, MP calls $6

Flop: ($18.50) Ts 9h 4d (2 players)
Hero bets $12, MP raises to $36.50, Hero raises to $204.50 and is all-in, MP calls $115 and is all-in

Turn: ($321.50) 5h (2 players, 2 are all-in)
River: ($321.50) Qh (2 players, 2 are all-in)

Results: $321.50 pot ($3 rake)
Final Board: Ts 9h 4d 5h Qh

MP mucked Td 9d (two pairs, Tens and Nines) and lost (-$160.50 net)
Hero showed Kh Jh (a flush, King high) and won $318.50 ($158 net)

This would probably look like an incredibly spewy play to a lot of players. Interestingly enough, one of the later stages of development referred to by Ed Miller in the video mentioned above covers exactly this type of play – using aggression to fight for relatively large pots in instances when you have some equity (even to the limited extent of a gutshot draw), but suspect that you’re an underdog. I have specific rationale for why I played the hand this way, which involves both a player read and some poker math, which I’ll get into in a later entry (after posting the hand for feedback on the forums).

Speaking of being a luckbox, I mentioned in a prior posting that I feel it is inappropriate to use aggregated EV stats as a means of determining whether you’ve been running lucky or unlucky over a period of time. Here is a reprinting of a thread I started and continued on this topic (and which appears to be dying a slow death from lack of interest, but whatever):

It seems to me that these stats don't necessarily tell the entire story as to whether you've been running lucky or unlucky, particularly over an extended period of time. I think factors in a player's style could affect the extent to which luck will be reflected as an EV differential.

For example... if a player will generally try to maximize his value on all three streets as a post-flop favorite, he's generally not going to be looking to shove his stack before the river, causing his opponent to fold. Yet the same player, as an underdog but with decent equity in the pot, would be more likely to shove before the river in an attempt to take advantage of whatever perceived fold equity he may have.

If his opponent catches up in the first scenario (and the final bet doesn't go in until the river), his EV is going to be 0 (notwithstanding the fact that he ran unlucky for the hand). But if he's called in the second scenario and catches up, his EV is going to be significantly negative (reflecting his positive luck).

It further seems to me that this effect will be more pronounced for more aggressive players, particularly when playing in more passive games (where they're facing more calling stations, and are not going to be facing as many speculative plays post-flop).

I thought about this a bit more, and then adding the following:

I think a more accurate description of the concept is that EV stats only tell you whether you've been running well or poor specifically in all-in situations prior to the river. But I see a lot of discussion where EV relative to bottom line is used to determine whether a player has been running better or worse than expected, overall.

When I review my sessions after the fact, on average, I may be all-in only 5 times or so before the river over the course of a couple of thousand hands. Sometimes it's slightly more, sometimes it's slightly less. But the EV calc only reflects the differential between expected and actual value over the course of that comparatively limited hand sample. The reality is, even with the relatively large pot sizes associated with all-in scenarios, the aggregate amount of money risked in non-all-in situations over the course of a typical session is far greater than the aggregate amounts risked when all-in.

You can have an EV that's lower than your bottom line, yet in reality, you have no way of knowing whether you've been running hot or cold overall, because there's no way to track it with respect to the vast majority of the hands you play.

Given that, I'm not sure I see the value of the aggregate EV stat. I suspect that it's highly overrated.

So there!

No poker tonight, since I need a pre-game nap before my team’s late game...

Current Bankroll: $14,269.84
Friday PM: 2601 hands; up $194.85; EV: -$216.78
Saturday PM (1): 654 hands; up $96.05; EV: $176.57
Saturday PM (2): 610 hands; up $165.30; EV: -$16.69
Month To Date: up $1,583; 5.61 bb/100

Music by the ELS Experiment