Search This Blog

Monday, March 22, 2010

Grade 2 goals

Now that I've finished analyzing my play over my first 10k hands at 10NL Rush, I'm going to set a few goals for grade 2, the next stage in my poker education. Although my bonus period is about to end, and that was the main reason for focusing on Rush over the last little while, I'm going to continue playing it for another 10k hands because it is a good training ground. It allows me to put up good volume and develop my basic skills before working on the more advanced skills of learning to read players and apply my skills to specific player tendencies.

Goal 1: Improve my stealing, especially on the Button and in the SB
I plan to achieve this goal by continuing to improve my stealing hand selection (not just stealing with ATC when the opportunity presents itself) and by working on my postflop play. More specifically, I need to pull back when appropriate and not just keep firing bets because I raised preflop. This is particularly important in the SB where I'll be playing OOP when my steal attempt gets called.

Goal 2: Improve my SB loss rate
I plan to achieve this goal by tightening up in the SB and improving my SB stealing.

Goal 3: Increase my postflop aggression
I plan to achieve this goal by firing more 2nd bullets in the right situations and to keep value betting in the right situations. I need to cut down on how often I go into check-fold mode. I'm going to work on closing the gap between my flop and turn CB rates.

Sunday, March 21, 2010

Progress report: Part 2

So I've spent some time studying my stats through my first ~10k hands of 10NL Rush and I think I've identified some definite leaks.

1) My loss rate in the SB is too high
I've read that one's loss rate in the SB should be better than -25 BB/100 and mine came out to a hare over -30 BB/100. When I dig deeper into the stats I find 2 problems. First, I've been playing too loosely in the SB in general. My VPIP in the SB during this stretch was 14.95, and when I filter by hands that I didn't raise to identify when I limped or called raises in the SB, my winrate is -8.72 BB/100. The biggest problem, however, seems to be failed steals in the SB. When I filter by steal or raise first-in, I actually have a positive winrate of 9.18 BB/100. But when I filter by steals that saw a flop to identify what happens when my steals get called, I find a winrate of -47.06 BB/100. One caveat here is that we're talking about a fairly small sample once you start filtering this deeply, but I did find one example of a problem hand in which I attempted a steal with 87s and overplayed it when it missed postflop, firing 2 bullets on a very wet board. I clearly need to be more careful about how I play failed SB steals postflop, for example giving up early if my speculative hand steals miss completely on wet flops.

2) Low W$WSF
My W$WSF stat over this stretch was only 36.67, which is definitely on the low side. It should be in the 'hood of 40. A low stat here suggests weakness postflop, usually a result of getting pushed off hands. This could be caused by playing fit-or-fold too much, which actually isn't necessarily an inappropriate strategy in Rush when you don't have reads on players to help identify floating opportunities. This is confirmed by my combined low WTSD (24.19) and high W$SD (57.60), which taken together indicate that I'm only going to showdown with strong hands. All these stats give a good picture of someone who plays weak-tight postflop poker.

However, it may also indicate that I'm not firing enough 2nd bullets and/or going into check-fold mode too often. Some other stats confirm this concern. My overall AF is on the low side at 2.48 (should be closer to 3), and my street-by-street AF drops from 3.25 on the flop to 1.81 on the turn. Another very telling stat is the difference between my flop CB rate of 73.72 and turn CB rate of 38.30. Thus, it would seem that I pull back on the turn too often and don't fire enough 2nd bullets.

3) Running into trouble with failed steals
When I first started looking at my steal stats I noticed a big problem with my Button steals. Over this stretch raising first-in from the Button resulted in a winrate of -16.12 BB/100. However, when I looked at the actual hands I discovered that this was due almost entirely to one hand w/ AKo, in which I stupidly overplayed TPTK by shove check-raising on the turn and ran into a big hand. That said, I still should have a stronger winrate on the Button even without that one hand, so there might still be a problem there. No doubt this has something to do with my aggression problems noted above, and my hand selection (which I noted in a previous post). Also relevant here is my discussion of failed SB steals in section 1 above.


Thursday, March 11, 2010

Progress report: Part 1

Now that I've played my first 10,000 hands at 10NL Rush, I figure this would be a good time for a progress report. I've been focused solely on Rush while I clear my new Full Tilt bonus, so I'm putting aside my 45-turbo goals for the time being. This progress report will occur in two parts: in part 1 I'm going to review the goals I set in my earlier post and give myself a report card based on my performance. In step 2 I'm going to examine my Poker Tracker stats more closely to identify leaks and areas for improvement. I'll use this information to set my goals for the next stage of my studies.

Goal 1: 3-bet more from the Button and Blinds
When I made the goals post on Feb. 21 I was a very nitty 3-bettor. My 3-bet stats preflop were 1.28 overall, 2.63 from the Button, 1.26 from the BB and 1.72 from the SB (through 3600+ hands). In the period since that post (~6700 hands) I've increased my 3-betting frequency quite a bit, to 3.04 overall, 4.17 from the Button, 4.55 from the BB and 2.89 from the SB. Of course an important question is whether that actually makes a positive difference in my results. If I filter my stats by preflop 3-betting, I earned 227 BB/100 in a very small sample of only 15 hands in the earlier period, while I earned 282.5 BB/100 in 64 hands during the later period. The sample size is obviously way too small to draw any real conclusions about the success of my 3-betting, but I have at least accomplished my goal of increasing my 3-betting frequency (overall and on the Button and in the Blinds) without hurting my results. Since this is in Rush where I'm 3-betting without reads, I'll have to learn to become more selective about my 3-betting when I move back to regular cash NLHE (which doesn't mean 3-bet less, only tailor my 3-betting based on my reads).
GRADE: A+

Goal 2: Stop getting involved in big pots with marginal holdings
This is a tough goal to evaluate, both because it would require some deep digging into my stats to see what kinds of hands I've been playing in big pots but also because the more I learn from my fellow NLHE players in discussion fora the more I have come to appreciate that the important skill to develop is not how to avoid big pots with marginal holdings, but how to effectively and successfully play marginal spots. It is, after all, impossible to completely avoid big pots with marginal hands. That being said, I do feel that I am becoming a bit more comfortable in marginal spots, learning when to check-call, check-fold, or bet, and when to make big calls. My postflop play with marginal holdings is probably the one area of my game I need the most work on going forward.
GRADE: B

Goal 3: Tighten up considerably in early and middle positions
Here is a comparison of my VPIP stats from various early/middle positions, pre- and post-Feb. 21:
SB - 17.12/13.79
UTG - 12.24/7.22
UTG+1 - 9.4/6.78
MP1 - 13.77/9.72
MP2 - 18.14/14.23
The change is pretty significant, having reduced my VPIP by 3-5% in all positions. I think I still have some work to do in the SB, however.
GRADE: A

Goal 4: Stop raising in late position purely to steal or set up a continuation bet - focus more on raising with hands that have good potential to hit big and are easy to play postflop
This is another goal that might require a bit of digging in my PT stats but I do know that I have made a concerted effort to be a bit more selective about my steals. I might be able to tell something from the changes in PFR stats and BB/hand winnings. Below is a comparison of my PFR and BB/hand stats, with the pre-Feb.21 stats and analysis posted first followed by the post-Feb.21 period.

Button: 9.42/0.15
CO: 10.30/-0.02
HJ: 11.64/0.01
The PFR stats are a bit odd as they indicate a wider stealing range from HJ, getting a bit tighter toward the Button. It should really be the other way around. This may be a result of compensating for the fact that most players identify Button steals as raises more than raises from other positions, so I was already being a bit more selective about my Button steals. The BB/hand numbers indicate that my raises from Button were profitable but I was running into trouble in CO and HJ. Now for the post-Feb.21 stats:

Button: 12.75/-0.01
CO: 12.31/0.26
HJ: 16.20/0.31
Well this is interesting. My LP raising game has opened up considerably, most notably on the Button and in the HJ. This isn't exactly consistent with the goal of being more selective with my steals. The $ results show that my performance on the Button has diminished considerably, so there is clearly a problem there. However, I've become quite profitable on my LP raises in the CO and HJ positions. Without looking deeper into the stats it's hard to tell whether the profitability is due to taking down more pots preflop with the higher steal rate, or improved postflop play. Clearly there is some more analysis needed with respect to my LP raises, especially the Button issue.
GRADE: C+ (though this one is hard to judge)

In Part 2 of my progress report I'm going to delve deeper into some of my stats to further understand some of the above issues and to identify other leaks as well.