a simple question

    • w4terman
      Joined: 04.04.2011 Posts: 757
      how a good graph should look for a NLH FR game ?

      showdown winning
      non showdown winnings
      all in ev

      if the stakes matter that i am sure there are i play on low limits (for now !)
      thank you :)
  • 3 replies
    • purplefizz
      Joined: 12.03.2008 Posts: 4,508
      hi w4terman,

      i will leave this question to our good members, but i personally think there is no standard "good graph". i would say personally it's good as long as it is trending upward over a large sample :f_biggrin:

    • pzhon
      Joined: 17.06.2010 Posts: 1,151
      The breakdown of hands into showdown and nonshowdown categories is not particularly useful. It tells you a few things about they games you play in and your playing style, but not whether you are playing well. The same player should have different breakdowns into showdown and nonshowdown results when playing more passive opponents (typical in low stakes games) versus more aggressive opponents.

      If you were losing money from the button, this would be a huge problem, and one you could fix by working on your button play. Regardless of what happens in other positions, you should not be losing money from the button when you could break even by folding every hand. Every hand you play from the button should be winning money on average and many hands are very profitable there, so if you are losing money then something is wrong. By contrast, it is possible to lose money in the nonshowdown category when nothing is wrong. This can happen when you correctly do not defend unprofitable hands in the blinds, while your opponents do. This causes small losses in the nonshowdown category (mainly when you post the blinds) while it gives you larger profits in the showdown category. These categories are linked, and you should concentrate on plays which increase your total profit not on one which might improve your nonshowdown winnings at the expense of your showdown winnings.

      As you start out, just about any level of profit is good. Luck dominates in the very short run, so don't get too excited over early wins or upset over early losses. A rough 95% confidence interval for some NL cash game players is the observed result/100 hands +- 1700/sqrt(#hands). So, if your win rate is 8 big blinds/100 hands after 10k hands, so that you are up 8 buy-ins, you only have strong statistical evidence that your win rate is somewhere between -9 bb/100 hands and +25 bb/100 hands.

      If you play with a deep stack and rarely get all-in preflop or on the flop, the all-in luck adjustment doesn't do much. It should converge to your true win rate slightly faster than the unadjusted results, but it doesn't filter out much of the luck. It is much more helpful for tournament players who frequently get all-in preflop. Many cash game players can ignore this statistic.

      So, 3/4 of the statistics you mention aren't particularly helpful, and the overall win rate helps only after a huge sample. What should you watch instead? Look for your VPIP/PFR stats by position, steal rates, blind defense rates, continuation bet rates in heads-up pots, 3-bet percentage, etc. Make sure that you are positionally aware, that you are not overdefending the blinds, and that you are properly aggressive when your opponents have shown no real strength.
    • w4terman
      Joined: 04.04.2011 Posts: 757
      thank you that was very helpfull :)