ICM Training results

    • Eclipt
      Joined: 11.05.2007 Posts: 755
      Hi. Always before sitting to SnGs I launch ICM Trainer and go through 100-200 exercises. Most of the times I have arround 85%-95% accuracy. Is that good result or should I spend (much) more time training?

      What is your accuracy?
  • 5 replies
    • helemaalnicks
      Joined: 21.09.2007 Posts: 7,195
      95, but icm is not the perfect path imo. Its good to get that feel, and nice to learn the concept, But other play is often better.
    • ManniXXX
      Joined: 05.09.2007 Posts: 707
      I've been thinking about this recently actually. If your on a table where no one knows about ICM, then it works beautifully but what if the whole table were playing ICM? Surely at that point it becomes more of a crapshoot. Not a complete one but more so..?

      I guess at that point the identification of player ranges becomes more important, and also at the higher limits I should imagine. Would love some discussion :)
    • erob60
      Joined: 08.03.2007 Posts: 165
      The ICM trainer calculates push ranges using based on Nash equilibrium. This means that if you're at a table of people who all play perfectly according these push ranges, then so must you, for any deviation from this will then be none optimal.

      However, if none of the people at the table have a clue what they're doing, then playing with this strategy is not optimal at all. If, for instance, you are on the bubble, with even stacks, you would normally assume your pushing range can can quite loose, but your calling range should be extremely tight. If however there is someone behind you who will call with any two cards greater than a nine (these clowns exist), then its very dangerous for you to be pushing with less than premium cards.

      So yeah, if the opponents don't know ICM, ICM is not optimal. If they all play it perfectly, you can't possibly expect to win more money than your chips are currently worth..
    • chenny8888
      Joined: 03.10.2007 Posts: 19,324
      yeah the beauty of poker is adapting your push range to suit your opponents tendencies.

      i've noticed, for example, on nick's limit that a lot of the casual players think they are very good at poker, and will hence not take that chance that the other player has AQ or AK when they are holding AJ, for example. yet they will liberally call small stealraises and happily play postflop.

      against such players i would make any-two pushes from late position, basically whenever there is a good chance they do not have a hand that they will call me with.

      on my limit i have villains that call very liberally, so i will tighten up a slight bit, without making too many shoves.

      then again there are villains that are just plain tight. you can make stealraises against these villains followed by a very profitable c-bet if they do call you.

      usually speaking though the opponents have some idea of what the bubble is, and will therefore call your shoves at the bubble a lot tighter if you have that opponent covered
    • carusel
      Joined: 28.10.2007 Posts: 202
      if you play ICM and no one at the table does so, you have the edge. if everyone at the table knows about ICM your chance to win is proportional to your stack, but as soon as someone deviates from ICM or makes a mistake, the edge is back to your side

      Lee Jones introduced this concept at wppc 2004 i think