Announcement

Collapse
No announcement yet.

Standard Deviation of Live $1/$2 Winrate?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • reeeeeeper
    replied
    If you want a guesstimate of what bankroll to start with read jonathanlittlepoker.com/bankroll

    If you want to know your standard deviation, record your session data variance and then sqrt the avg. Variance of a given session would be:

    (session_winrate - average_winrate)^2

    Then get the average of all that session data and take the sqrt. That's your standard deviation. With that data and using an acceptable risk of ruin percentage to your situation you can calculate your minimum bankroll requirements:

    -(stddev^2 / winrate) * ln(RoR%)

    So like if your stddev is 120 and your hourly is $6 live (about 10BB / 100 hands) and your acceptable RoR is 3% then your min BR is $8416, or about $8.5k.

    But guess what, you don't have to do all this crap. Go to the bankroll link I provided in the first paragraph, look up 10bb/100 and you'll find $8k which is pretty much the same. So I'm guessing he used around a 100 stddev to calculate those. If you really need to you can calculate your own, but the ones JL provided should be enough.

    Notice also I didn't use the max winrate, I used a pretty conservative one. If you were that confident you could win 25BB/100 you wouldn't really need to do these calcs for low stakes or be worried about BR. You wouldn't be playing at $1/2 for very long.
    Last edited by reeeeeeper; 11-06-2019, 10:18 PM.

    Leave a comment:


  • zer0sum
    started a topic Standard Deviation of Live $1/$2 Winrate?

    Standard Deviation of Live $1/$2 Winrate?

    I just finished Jonathan Little's $1/$2 hand review in PokerCoaching.com. It was excellent.

    In this video JL says that a $20/hr win-rate is possible if you are good enough (possibly as high as $30/hr but I'm sticking with $20/hr to be conservative).

    My question is: playing in the style that JL advocates, what sort of standard deviation of win-rate can you expect? Assume that your competition is absolutely terrible (lots of maniacs, weak-passive, and loose passive opponents). Is $150/hr standard deviation too high of a guess?

    With this piece of information I can run some simulations to see frequency of going bust with different starting bankrolls (also to get a feel for how bad the down swings can get).
Working...
X