Micro Variance.

[Downrock]

Just a simple question.. are 100+ BB downswings at 25/50c or 50c/1 a regular occurance or is this rare at micros and more likely due to some poor play? I here about big downswings on this site a lot but they almost always are at higher limits. Thanks.

[jaxUp]

100BB is very reasonable and should be expected IMO. Even 200BB should be expected if you play enough. This also depends on how bad you play.

[Niediam]

The likelyhood of having a big downswing is inversely coorelated with your win rate.

And of course higher winrates are easier to achieve at the lower limits...

[nofutures]

The stakes are irrelevant to the likelyhood of variance. What matters is a) your winrate (BB/100h) and b) your standard deviation (SD/100h). If you use PT and have played 30k+ hands then you'll know both and can plug it into some formula which calculates the likelyhood of downswings. If you are not willing to move down in stakes you need much more than 300BB in your roll. The thing is that moving down in micros doesn't make much sense if you know you are a winner, because you can get a few hundred bucks easier elsewhere.

The higher the winrate the more unlikely are downswings.
The smaller the SD the more unlikely are downswings. You can reduce your SD by playing tighter and passing on marginal +EV hands.

The size of the roll is a tradeoff between the rate of growth and risk of ruin.

[detruncate]

[ QUOTE ]
You can reduce your SD by playing tighter and passing on marginal +EV hands.

[/ QUOTE ]

People who are skillful enough to accurately assess this probably don't need to bother. But if you simply mean that micros should be cautious about the massive VPIPs that seem to be de reguire these days even for people who don't play particularly well post flop I agree.

Variance sucks. It's not worth worrying about. Trying to reduce variance isn't a productive struggle. Concentrate instead on learning how to deal with the swings. It'll serve you much better in the long run.

[nofutures]

Yes, right. Although I believe that spending some thoughts on how ones preference regarding risk of ruin, growth of bankroll and willingness of moving down in limits.

For instance I found I hate moving down limits and going broke. That's why I will use a 600BB rule from now on, after going broke 5 times or so. I also will move up only if I made 300-500BB at the current level.

[detruncate]

[ QUOTE ]
Yes, right. Although I believe that spending some thoughts on how ones preference regarding risk of ruin, growth of bankroll and willingness of moving down in limits.

For instance I found I hate moving down limits and going broke. That's why I will use a 600BB rule from now on, after going broke 5 times or so. I also will move up only if I made 300-500BB at the current level.

[/ QUOTE ]

Bankroll management is an important skill, and easily mastered with a little discipline. Bankrolls are finite. Of course you have to be increasingly cautious as your intolerance for moving down or busting out increases. And some people are just generally more risk averse than others.

Some people are perfectly comfortable with 300 BB. You need 600. In my situation I prefer 1000+. To each their own. But -150 BB stretches aren't uncommon, and some of the best brown trouts have suffered through much larger downswings. A necessity to keep earing and/or an unwillingness to move down necessarily requires a bigger bankroll to minimize your risk or ruin (let alone your discomfort during a stressful time).

The only issue I took with your post was your suggestion that taking steps to lower your standard deviation could be a productive means of reducing your risk of ruin. Again, it's very difficult for novice players to accurately assess the profitability of a given situation let alone figure out where one might productively lower one's variance without significantly affecting one's winrate. How's that for stodgy prose?

[DavidC]

[ QUOTE ]
Just a simple question.. are 100+ BB downswings at 25/50c or 50c/1 a regular occurance

[/ QUOTE ]

Yes. Way way way bigger than that, too. Like 300-500 bb. Furthermore, this doesn't include your natural winrate into the swing either (which for fairness you should). This is just regular BB's below a given starting point.

[ QUOTE ]
or is this rare at micros and more likely due to some poor play?

[/ QUOTE ]

Yes. If you're playing at the 0.5/1 level you probably aren't as good as a good winning 5/10 player. Therefore your edge would be smaller than if you were a good winning 5/10 player playing at the 0.5/1 game, and thus your swings will be further into the negative than someone with a higher natural winrate.

[Yerma]

I wrote a little something for questions like these. You can get your bankroll requirements for various win rates and standard deviations experimentally. I put it at http://www.pokervillain.com/HoldemSimSetup.zip

If anyone wants the source code for this (it's all C#), send me a PM.

Anyways, I'm getting that someone who plays 100,000 brick and mortar hours (about 300K hands) while winning at 1BB/hr (about 3BB/100 hands) with a standard deviation of 10BB/hr (I'm not sure what that is in BB/100) can have downswings of 620BB, although this is very rare! Usually when I run the simulation for 100,000 hours, it suggests that the worst downswing the player experiences is between 280-550 BB. Anyways, prepare for a 600BB downswing if you're winning at 3BB/100 is the lesson here?

In a bizarre twist, I just ran it and got that the player had a 595 BB downswing, but still ended up posting a 102,618 BB win in 100,000 hours--so he ran hot?! Go figure.

[kerowo]

After 100+k hands at .5/1 my worst single session down swing is about 120BB, my worst week is probably close to 300BB. My win rate at the time was around 2BB/100. At the end of the month I cash out everthing over 400BB. My bank roll has never been under 300BB.