Starting hand evaluation

    • Loreit
      Loreit
      Bronze
      Joined: 20.11.2012 Posts: 31
      I'm a bit confused. If someone know a thread already discussing this please just link to it (I have tried some basic search)

      A simple compare would be AKs vs QQ.

      I mainly follow Sklanskys starting hands chart, and according to this QQ beats AKs. QQ beats AKs.

      However SngWiz claims that AKs is stronger than QQ if I understand it correct? Looking into the Hand Range tool players who plays 1.2% of their starting hands plays AKs, and players need to play 1.7% of their starting hands to play QQ.

      Above is the simple example.

      An example where they deviates a lot is KTo vs 98s. SngWiz Hand Range tool claims that players needs to play about 25% vs 46% of their starting hands to play either hand. In the same chart as above they have about the same winning rate.

      As I said first, I'm confused. Since SngWiz is all about pushfold and the chances of winning a showdown, the chart I already have linked twice should be a better reference than the inbuilt Hand Range tool?

      Or do I miss understand the Hand Range tool?

      I'm quite sure that I'm not the first to ask this question, however I don't find the answer anywhere?
  • 5 replies
    • YohanN7
      YohanN7
      Silver
      Joined: 15.06.2009 Posts: 4,711
      I don't even have SngWizard, but there is no way that AKs beats QQ in any way. AKs doesn't even beat 22 (unless both deuces are of a different suit than the AK, in which case it is really close) in an all in situation if they are put against each other.

      The situation is different if it is all in 3-handed or more, where the third hand is random. Then AKs is much better than (say) 22, but QQ is better still. If it is not all in, and the stacks are deep, then some players may think that AKs is simpler to play than QQ. This is clearer with 89s vs KTo (many players wouldn't touch KTo with a bargepole). Still, KTo is a pretty huge favorite over 89s when it is all in preflop.

      Then there is the issue of "live outs". Suppose you 3-bet all in with 89s (as a bluff) and get called. This is not a good situation, but is probably better than 3-betting with KTo and getting called ,since the hands that call often have a better K in them than KT.

      /Johan = :f_confused:
    • wiarygodny
      wiarygodny
      Bronze
      Joined: 16.07.2009 Posts: 1,395
      Don't know if thats gonna help a lot to clarify your doubts but imo the short answer is that you can never think abt a single, ultimate hand ranking that is gonna be accurate in every single situation. Depending on the opponent's range youre facing the EV (which would be used to construct a 'ranking') of a particular hand of yours changes. Just to make sure we're on the same page im talking here abt push/fold situations in an SnG. In different situations the 'rankings' you should use to determine the action with a particular hand you hold are gonna differ form each other. Some examples from the top of my head:
      - calling an ATC push in a situation when youre extremely risk averse AKs<88 (which is not necessarily true in many other situations)
      - calling a tight push getting great odds ATo<JTs

      i'd suggest you play with the Equilab a bit to get the feel of it yourself. Maybe even better for that would be playing around with Wiz: set a particular calling/pushing range for your opponent and then change the hand youre holding manually and compare different EV results vs a particular range. This way it should be pretty easy to see what kind of hands (e.g. small suited connectors) gain/lose value in particualr situations.

      I'm not familiar with the ranking youre linking to but if i understand correctly it's not exactly Sklanky's ranking,which you can find here, but some ranking that was developed through a study af a massive amount of hands and their results, but in cash games!. Tbh it looks a bit weird to me and the way is was developed doesn't seem to be very clear to me, but im no cash game player so what do i know :) When it comes to Wiz it uses the Sklansky-Chubukov (Sklansky-Karlson) ranking, more on that on their website.

      Hope that helps at least a bit and doesn't confuse you even more :)
    • Optroot
      Optroot
      Bronze
      Joined: 11.05.2008 Posts: 250
      AKs is a better hand than QQ simply because AKs dominates more than QQ, it has nothing to do with their equities compared to each other. People like to push/call with Ax, which makes AKs better than QQ according to them. There's also a little bit of card removal too.

      Personally I think it should be AA,KK,QQ,AKs,JJ,AKo. But it really doesn't matter. In fact, I can prove that there exists no correct ordering or hands, and any hand order you use will lead to an incorrect calculation (yes this means all ICM calculators are wrong to a small degree!). Interestingly enough the proof uses the position of AKs.

      I can provide that proof if you want

      Optroot
    • YohanN7
      YohanN7
      Silver
      Joined: 15.06.2009 Posts: 4,711
      Originally posted by Optroot
      AKs is a better hand than QQ simply because AKs dominates more than QQ, it has nothing to do with their equities compared to each other. People like to push/call with Ax, which makes AKs better than QQ according to them. There's also a little bit of card removal too.

      Optroot
      No.

      QQ is better than AKs versus any range in the world when it comes to push/fold poker period. It has all to do with equities versus ranges.

      It is another matter that AK is a hand that is suitable for a push with pretty big stacks, especially oop. (You would still prefer to have QQ in that spot, in which case not pushing, but raising is the better play.)

      /Johan = :f_confused:
    • YohanN7
      YohanN7
      Silver
      Joined: 15.06.2009 Posts: 4,711
      Ok, so I was wrong. There are freak ranges against which AKs is slightly better off than QQ. These ranges do not contain pairs below QQ. Anybody interested is best off playing with Equilab.

      /Johan = :f_confused: