Weak Tight Dreams

[borges]

I've been playing poker for about 2 months. I'm going through my first
horrendous downswing (100BB in 600 hands).....I had just switched limits
a couple of weeks ago (from .25/.5 to .5/1).....so I've been having a lot of
doubts about what I'm doing. Really even about fundamental philosophy of
how do play this game.

I love reading this board. The collective poker wisdon is really intimidating for me....I read the hands......try to come up with what I would do......(usually
CALL).....read what the maestros say.....hit my forehead....say..."Of course!
RAISE THAT **$>??* "

Any statistician will tell you that reporting a mean value without a standard
deviation (or a variance) is meaningless. I wonder if sometimes our
poker philosophy suffers from a not considering the variance of our actions.
It is to say, that we have set up a value structure completely based on maximation of expected value. But, this is clearly inadequate, right? For
the sake of argument, suppose we have two courses of action. Action
A gives us an EV of 1 (with a std. dev. of 3) and action B gives us
an EV of 1.01 (with a std. dev. of 20).....Clearly (except for the insane), the
better course of action is A even though it has a smaller expected value.

Of course we don't have such clear cut decisions all the time but I would think
that when EV decisions are close (say between folding vs. calling or
folding vs. raising or calling vs. raising), the weak tit decision will lead us to
making lower variance plays.

I realize that poker awards aggession. I'm not arguing against that. I'm just
saying that after all evaluations of various courses of action are taken (say
the aggressive play nets us an EV of .02 and folding nets us 0.....I would think
that the EV is so close that we should be guided by the lower variance option).

I realize that this is heresy.......May I ask to be shot instead of burned at
the stake? [img]/images/graemlins/smile.gif[/img]

This must have been discussed before. I tried using the search function but
didn't come up with anything.

[trdi]

You have to stop worrying, that's all. I was thinking your way, too, but with 250BB swings over 10k hands. Playing a lot makes you see the whole picture - 100BB is nothing, that's one day swing (and coming back). 600 hands? I don't know if I'm insane, but I play 2500 hands per night. 100BB downswing is not the time to even start thinking you are down.
About your "low variance play" - I was there. And now I see. There are no plays like you describe. Push your edge. Why don't you post a hand you think would be a good example for "low variance-almost same EV" game?

[dwedeking]

If you are more aggressive then your swings are going to be larger. When playing passive this means that your losses are going to be less. Your wins are also going to be less. You only pay rake on hands that you win. So the house is going to take a larger percentage of the pot on the hands you win (making it harder to overcome).

Also you are utilizing less of your skill set by being passive. When Microsoft runs across a competitor they do not play it safe knowing the little guy can't hurt the large company... yet. Instead they leverage their huge financial edge and put them away fast. If your AA you do not sit back and wait to see if the old lady at the end of the table can suck out on you, you pressure her hard from the start to make mistakes so she either folds (letting you win) or competes on an uneven playing field. Because of our study of pot odds, implied odds etc we can expand that leverage past just the few hands (AA, KK, QQ etc - the true monsters) that a tight weak would play.

[TheNomad]

I'm also going through a downswing. I'm aggressive just like I've been taught to be. But the opponents are passive. On Party 1/2 there are loose and tight opponents, but they are almost always passive. So, I've found I'm continually betting into someone who isn't showing any hand strength only to find two pair or some unlikely straight on the showdown.

I'm wondering whether it might be better 1) to tighten up considerably in ep. 2) the blinds are always a problem because you are out of position. If you are in BB and flop TP with a weak kicker, I'm wondering whether it might be better not to bet?
3)Even if you are passive and see whether they bet first, then all that could mean is they have TP with a weak kicker.

I've just dropped 100BB in one session mostly thanks to people who play 84s. (Not very happy just now).

[tiltaholic]

hello, relatively new poster.

- 100 bb doesn't make a horrendous downswing.

- you probably aren't properly evaluating the decisions that you are facing since the vast vast majority of them will have a CLEARLY positive expected value, while a negligble effect on your variance.

- your example about ev+/-stand. dev. is false. here's the situation as it applies to poker:

ev +$1 +/- $3,
ex +$1.01 +/- $20.
(this artificial variance is too high by the way. at .5/1 thats a 17 bb difference on each hand which is impposible)

this situation happens on each hand, and you will play one million hands. what is your choice and what is your expected final bankroll after one million hands?

don't think about variance. don't think about chance. learn what action/play gives you the highest expectation and let the variance even itself out on its own. it has to, because it's variance. the only thing you can change or control is the expected value of your action (and your emotional state). this is why we must be properly bankrolled at all times.

good luck. keep going.

[borges]

Thanks guys....I'm trying to keep this in perspective.....I think I'm going to
drop down in limits again and try to keep on firing instead of staying where
I am and giving in to those (for me) almost overpowering weak-tight tendencies

[sarsen]

Pick up Mason Malmuth's book 'Gambling Theory and Other Topics' or Poker Essays I. He talks about win rate and standard deviation in terms of rate per hour. His example is for a $20/$40 holdem player with a expectation of 1BB/hour and a standard deviation of 12.5BB per hour. This means that there's a 66% chance of between -$500 and +$500 and 98% chance of between -$1500 and +$1500 per hour.

[Buzz-cp]

If we have a bankroll that is sufficient to deal with B, then it is the more logical choice. However, I think your examples are exaggerrated. Pumping flush draws and value-betting marginal hands into LPPs on the river are some of the most +EV plays I have learned to use at the micros. Learning to deal with downswings is a separate issue entirely, but more than makes up for the aforementioned.

[ESKiMO-SiCKNE5S]

[ QUOTE ]
I wonder if sometimes our
poker philosophy suffers from a not considering the variance of our actions.

[/ QUOTE ]

no it doesnt.

[ QUOTE ]
It is to say, that we have set up a value structure completely based on maximation of expected value. But, this is clearly inadequate, right?

[/ QUOTE ]

wrong

[ QUOTE ]
For the sake of argument, suppose we have two courses of action. Action
A gives us an EV of 1 (with a std. dev. of 3) and action B gives us
an EV of 1.01 (with a std. dev. of 20).....Clearly (except for the insane), the
better course of action is A even though it has a smaller expected value.

[/ QUOTE ]

the better course of action is 100% B. pushing these .01 advantages is HOW you are successful at limit poker. variance is completely meaningless in the long term. expected value is EVERYTHING. (note: this is expected value in the broader sense; inclusive of things such as metagame considerations and tournament bust probabilities)

if variance is at all an issue in our thinking then we are utilising poor bankroll management skills. the idea behind the 300BB rule and others are that we are only ever risking enough money on any given hand for the variance factor to not be an issue. we deliberately play at these low (relative to bankroll) limits to purely maximise our expected value with these often high risk aggressive plays and ignore the impact of variance over time.

[Dunkman]

A couple things...first I felt the same way when I started (I'm an econometrician, so I tend to look at things in a similar fashion.) For instance, the aggressive play netting .02 EV and folding netting 0. From what I understand, once you get up to higher limits, a 2BB/100 is a very respectable winrate. That reduces to .02BB/hand. That tiny little edge is really all you have over the long run, so you can't afford to give it up.

As far as the EV of 1 vs. 1.01 with a higher variance, over the course of hundreds of thousands of hands the variance will take care of itself, and if you have a sufficient bankroll then you can afford to squeeze every small bit of +EV out of a hand. I think somewhere in SSHE (or maybe another book I've read) it does briefly mention the fact that if you are playing above you're roll that some plays could be played differently, as while the aggresive style of play does substantially increase variance, it also substantially increases winrate. Look, you can beat a .25/.50 game playing weak/tight, because you'll still take money from the loose/weak and maniac players. Like SSHE says, we don't want you to just beat small stakes games, we want you to crush them. Playing aggresive, lets say your upswing would be 400 BB, whereas with TPP play it may be 200 BB. AT the same point, lets say your downswing would be 200 BB (of course downswing is smaller, these games are very beatable) whereas playing TPP it would be 100BB. Those are rough, meaningless numbers, but I think they fairly represent what you are giving up over the long run.

Finally, concerning bankroll, my understand is that the 300BB bankroll is what most people use for live play. The bankroll size I hear most commonly for online play is 1000BB, due to multitabling and the vast number of hands that get played (if you're playing bad a day/week your gonna play 10x as many hands as you would live, and thus stand to lose much more.) That may not be totally necessary for these small stakes games, but to counter the variance at higher levels that the number I usually hear.