Once again adjusted values

    • getdotacom
      getdotacom
      Black
      Joined: 06.04.2008 Posts: 607
      I'm playing 9man SNG's on Pokerstars and I wanted to ask about luck adjusted values. Here is my graph
      And I want to know how can I trust that adjusted line, and if it's correct, I just have been unlucky ?
  • 6 replies
    • lessthanthreee
      lessthanthreee
      Bronze
      Joined: 30.06.2009 Posts: 16,300
      it happens.

      im also 60BI < EV on my current limit. stick with ur game it will turn around. you dont even have a 1k sample yet, still a long way to go :)

      edit: i would be worried if the two lines were reversed. its good news that ur red line is trending up nicely
    • getdotacom
      getdotacom
      Black
      Joined: 06.04.2008 Posts: 607
      Thx man. I have won some buyins back , but still looks like red line is going up and the green one down. Hope not for long here's my latest graph
    • lessthanthreee
      lessthanthreee
      Bronze
      Joined: 30.06.2009 Posts: 16,300
      are you sure these are 9 mans? this looks strangly like an multi-table sng effect. whereby the red line will trend away from the green to infinity.
    • getdotacom
      getdotacom
      Black
      Joined: 06.04.2008 Posts: 607
      these are 9 mans, maybe there are few HU games included, but not more than 5%. I haven't played other games since using HEM, so I start thinking are these lines correct, but I made profit before, now it looks strange. I have changed my strategy a little bit, but I can't understand now am I losing or am I very very unlucky...
    • getdotacom
      getdotacom
      Black
      Joined: 06.04.2008 Posts: 607
      I just played 12 games and get very nice stack on most of tables at the beginning, but lost 6.27 buyins at all, but red line says I should be winning 1.5. LOL looks I'm unlucky :(
    • pzhon
      pzhon
      Bronze
      Joined: 17.06.2010 Posts: 1,151
      The all-in luck adjustment is unbiased (when it isn't buggy--don't trust it for multitable SNGs). It gives you a more accurate estimate of your skill than your results. However, there is still some luck that it doesn't factor out. If you get dealt KK and get all-in, it's just luck whether you ran into QQ or AA, but this type of luck is ignored by the all-in luck adjustment.

      If you play turbo SNGs, the all-in luck adjustment might remove about 50-60% of the variance. That means your luck-adjusted results after 400-500 games are about as reliable as unadjusted results would be after 1000 tournaments (with a standard deviation of 5% ROI). After 1000 games, your adjusted results may be as reliable as unadjusted results would be after 2000-2500 tournaments, with a standard deviation of about 3.3% ROI.

      While the luck-adjusted results are unbiased by themselves, they are (conditionally) biased in the direction of your actual results. If your true ROI is 5%, then if you look at the days when your ROI was over 10%, your adjusted ROI on those days will tend to be over 5%. Similarly, don't be too concerned if you have a bad day and run at -10% ROI, and the luck-adjusted result is -5%. When you get lucky all-in early in the tournament, you have more opportunities to play skillfully later.