Why Go With A Low Variance Style?
I was hoping that I would get more comments on my Low Variance manifesto, but oh well. I figured rather than pick a part an aspect or two of it, the high variance players would call me out for a general aversion to gambooling. Why play poker, If I seem to not want to gamble? Wes pretty much nailed what I expected on the head with this post below.
The following was going to be a comment on Blinders blog, but since I never blog here anyway I figure I’ll throw it up here too since I think it conveys an important point.
I would never want to be one to care about if I play a high variance or low variance game. I always play in a game that I am VASTLY overrolled for and thus I am never affected by the individual results of each session. If I go up or down 10 buy-ins, it is all relative and I should not worry about that individual session as long as I was playing my A game. If there is a situation that yields a +EV spot, why should I take that spot up? Simply put, I shouldn't’t.
Don’t be results oriented. Don’t think about lowering your variance. Think about how to improve your game, about how to always play your A game, and in turn because of this your win rate.
I totally agree with what Wes is saying here, but I am going to qualify it a little bit. In the manifesto I alluded to playing for less profit potential by lower variance could actually be a good thing. I will now say why, and also argue that it can ultimately be more profitable to take the less profitable line long term. Did you just read that correctly? Yes!
So Wes is overrolled for the level he plays at, plays his A game, and does not care if he goes up or down 10 buy-ins in any given session. He plays his A game regardless of what variance throws his way. I don't think this is the case for most poker players. Variance tends to throw people off their "A-Game". It can throw you off your A-game when variance is overly positive and when it is overly negative. Since the variance signal in your bankroll is so large, it is constantly trying to throw you off your game.
So lets look at it mathematically. I don't know what Wes's long term win rate, but lets just say for argument it is 10-12 BB/100h (BB = 2x the big blind). Let's say for him it is always this rate because variance never throws him off his A-game. Now lets say my win rate is 6 BB/100h playing my lower variance style. My game like most people is effected by variance so my win rate is a combination of my A-Game, B-Game, C-Game, F-Game... win rates, and the time that I spend playing each version of my game. So to simplify things lets say that I play my A-Game 50% of the time for 8 BB/100h, and my C-game for 4 BB /100h for 50% of the time. Overall, I win 6 BB/100h. So if I could completely shake off the effect of variance on my game, I could win 8 BB/100h, but if I want to win like Wes does, I need to add more variance to my game that will also bring the profit rate up to 10-12 BB/100h.
But, I like most people am effected by variance, so this gets a little dangerous to my win rate potential. At the start of the $300 bonus challenge, I was running on the hot side of variance. I was getting sloppy with my table selection, and number of tables open as a result of this lowering my long term win rate (not on my A-game). So running hot put me off my A-Game. Then when things went south and got on the wrong side of variance, I started overplaying AA post flop, and pressing too hard to get even or up. So the downturn also threw me off my A-Game. Because variance was randomly aligned (towards up then down) large during the challenge, I never was able to play my A-Game. So I spent a lot of time playing the 4 BB /100h C-Game, and with the extra bad negative variance ended up at about 1 BB /100h for the challenge.
So for me to add profit rate, I need to add variance. But I am already playing a pretty profitable style, so there is not a lot of "low hanging fruit" out there. The tweaks that I will need to make to my game will add a small amount of profit rate, but add a huge amount of variance. The extra variance added as a result is more likely to throw me off my A-game then if I stick to a lower variance style. This can decrease my overall win rate, even though the adjustment in theory is positive to win rate. Let me give a quick example.
Sample tweak to game:
Raise all pairs 55-99 from EP and MP vs. the lower variance limp/call line.
Profit rate increase expected .3 BB/100 h for all game versions (A-game, B-Game...)
8 BB /100h 50% of the time
4 BB /100h 50% of the time
6 BB /100h Total
8.3 BB / 100h 40% of the time
4.3 BB / 100h 60% of the time
5.9 BB / 100h Total
So my tweak in theory increased win rate by .3 BB / 100h, but in practice lowers the win rate by .1 BB / 100h even though all versions of my game are more profitable after the change. The added variance just tends to throw me off my game more than what would normally happen which is expected. So if you really believe variance can't throw you off your game, then play the most profitable style you can (play like Wes), but if variance can effect your play, or very much effects your play, the low variance style might be the most profitable style you can personally play. Hope this makes sense to everyone. Also, this really stresses the importance of separating results from decisions. As soon as you can focus on the actions (decisions) you made and if they were right or wrong, and not on the results (if you won or lost), you will start to play your A-game much more often and your win rate will go up. Even if you have done nothing else to improve your game.