This site uses cookies to improve your browsing experience. By continuing to browse the website, you accept such cookies. For more details and to change your settings, see our Cookie Policy and Privacy Policy. Close

[NL2] (FR) AA - cbet size in dry board vs donkey (?)

    • pguest
      pguest
      Silver
      Joined: 01.04.2009 Posts: 1,899
      $0.02 NL - Holdem - 7 players

      BB: 105 BB (VPIP: 18.30, PFR: 2.56, 3Bet Preflop: 2.60, Hands: 237)
      UTG: 61 BB (VPIP: 36.65, PFR: 12.99, 3Bet Preflop: 0.53, Hands: 550)
      UTG+1: 101.5 BB (VPIP: 15.38, PFR: 10.99, 3Bet Preflop: 0.00, Hands: 93)
      Hero (MP): 115 BB
      CO: 140.5 BB (VPIP: 20.34, PFR: 18.64, 3Bet Preflop: 12.50, Hands: 62)
      BTN: 117 BB (VPIP: 16.97, PFR: 9.87, 3Bet Preflop: 4.89, Hands: 553)
      SB: 66.5 BB (VPIP: 50.85, PFR: 22.03, 3Bet Preflop: 4.17, Hands: 59)

      SB posts SB 0.5 BB, BB posts BB 1 BB

      Pre Flop: (pot: 1.5 BB) Hero has A:heart: A:club:

      fold, fold, Hero raises to 3 BB, fold, fold, SB calls 2.5 BB, fold

      Flop: (7 BB, 2 players) Q:diamond: 3:club: 6:spade:
      My thoughts:
      The sample size was very small, but based on these stats, I assumed the vilan was something between a maniac and an aggro-donkey.
      In such dry board, should've I c-bet'ed less? Maybe half pot, so I could've been payed by a tp with an ok kicker or by a middle pair?

      SB checks, Hero bets 4 BB, fold
      Hero wins 7 BB
  • 11 replies
    • la55i
      la55i
      Moderator
      Moderator
      Joined: 27.01.2013 Posts: 8,230
      I think half pot or even smaller could be just fine.

      Interpreting the stats on this sample size is a bit problematic. I would categorize villain like that if the VPIP was 80 but now it is just 50. This can very well be variance. Of course there is a big chance that villain is on the loose side but then again on the long rung his stats could settle somewhere around 30/25 too, who knows.
      Ofc in this hand we pretty much didn't need the stats, but in the future you should be careful :) You can look at the stats but don't make too big adjustments too early.
    • pguest
      pguest
      Silver
      Joined: 01.04.2009 Posts: 1,899
      Originally posted by la55i
      Ofc in this hand we pretty much didn't need the stats, but in the future you should be careful :) You can look at the stats but don't make too big adjustments too early.
      What would be a good sample size for those PF stats? Should I completely ignore them until I have 1k+ hands of samples of that vilan?
    • la55i
      la55i
      Moderator
      Moderator
      Joined: 27.01.2013 Posts: 8,230
      Noo you shouldn't totally ignore those. You can start making some guesses after 100 hands. VPIP/PFR should start to settle somewhere between 100-200 even if the guy did run bad or good for a while. But 3bet stats and postflop stats need hundreds of hands.
    • RubbyDubby
      RubbyDubby
      Bronze
      Joined: 15.02.2018 Posts: 193
      Originally posted by pguest
      What would be a good sample size for those PF stats? Should I completely ignore them until I have 1k+ hands of samples of that vilan?
      Nooo of course not! :)

      It's a bell curve..

      The more hands you have on villain the less likely the chances are his stats are off but most of the time (depends a bit on his stats, 25 hands with a 100/100 is different than 25 hands with a 35/25) for VPIP and PFR you start to develop a feeling after 25 hands, you can have a feeling at 50 and have even more of a feeling at 100. You would throw away info if you would not use those stats at all before a certain threshold.

      But not all villains that look like a fish are a totally fish.

      edit-1: Ow.. I was typing a looong time and Ia55i beat me (and I have a little different opinion about when stats start to say some).
      edit-2: My biggest mistake in poker was that I kind of 'forgot' that 'recreationals' can have monsters too!
    • la55i
      la55i
      Moderator
      Moderator
      Joined: 27.01.2013 Posts: 8,230
      Yea we can start getting some sort of a feeling on a very small sample also but I usually just prefer to be safe :) We can consider the stats but we shouldn't do too big adjustments based on that.

      I have played some sample sessions where I follow my own stats and see how long it takes before they get somewhat accurate. And in some cases it has required a lot of hands. For example in one case I played 13/11 after 50 hands and 16/13 after 100 hands. It really took 200 hands before it got up to 26/20 which is somewhat close to reality. And 3bet stats needed about 500.

      But that was a bad run. Of course sometimes we get reliable numbers faster but I think it is quite important not to adjust too much. Some villains might have thought that they are playing against a nit who folds 100% vs 3bets after they played 150 hands against me but that is very far from the truth :f_biggrin:
    • RubbyDubby
      RubbyDubby
      Bronze
      Joined: 15.02.2018 Posts: 193
      Originally posted by la55i
      Yea we can start getting some sort of a feeling on a very small sample ..... We can consider the stats but we shouldn't do too big adjustments based on that.
      I 100% agree!

      But (wild guess) I don't think you have ever looked down at your stats after 25 hands and saw something likee 85/15 :D
    • pguest
      pguest
      Silver
      Joined: 01.04.2009 Posts: 1,899
      Somewhere, here on PS, do we have info on "the generic player type", for the specific limit that we play?

      For example, based on data gathered from X millions of PS players, for NL2 FR, the generic player type or the stats of the most common player at this limit would be something like 30/5, so we could categorize him/her as Y and, in the case of a small sample size for a random vilan, just assume that he/she fits in the "generic category" and play him/her accordingly.

      That story of the most common player type being a calling station is a kind of a myth nowadays...
      Most players that I've seen, that would look loose preflop, play (as la55i has said before) fit or fold post flop :f_confused:
    • la55i
      la55i
      Moderator
      Moderator
      Joined: 27.01.2013 Posts: 8,230
      Yea those extreme stats like 85/15 are a different thing :D

      Somewhere, here on PS, do we have info on "the generic player type", for the specific limit that we play?
      What you are looking for is some kind of a population/player pool analysis. I don't think we have anything very accurate here at pokerstrategy. But I can tell you that at stars micro limits (both zoom and regular, NL2-NL25) an average guy plays something like 26/16 with quite decent stats but on the passive side.
    • RubbyDubby
      RubbyDubby
      Bronze
      Joined: 15.02.2018 Posts: 193
      We're getting totally off topic but hey...

      I do and I do not understand your question totally but some of my type of players I can give you.

      • 50/4/0 or whatever... Those guys are doing 1 and only 1 thing and that is trying to hit sets or suck out with straits. If they wake up !!!FOLD!!! but normally value bet the hell out of these guys.
      • 24/19/8 ... Good regs. Play only 'in position' and do not get fancy (but you can bluff them).
      • 24/19/1 ... Bad regs. Those guys will pay you off with AA when you hit a monster. Also good to bluff against.
      • 35+/25/0 ... Many times these guys raise preflop and then fold later streets when they face resistance. Good for floating.
      • 35+/25/15 .. Crazy maniacs. Tighten up. They'll hang themselves.


      PS. As you can see I think 3 bet stats are important here. You need some sample for them to get a bit more accurate.
    • pguest
      pguest
      Silver
      Joined: 01.04.2009 Posts: 1,899
      Thanks, RubbyDubby! Thanks, la55i!

      Out of curiosity: are those average values for FR or SH?

      PT4 "Leak Tracker", tells me that healthy values for VPIP are:

        - 11%-23% for fullring
        - 17%-30% for shorthanded


      and, for PFR:

        - 8%-14% for fullring
        - 13%-20% for shorthanded


      But I don't know how much of truth are on those PT4 values...
    • la55i
      la55i
      Moderator
      Moderator
      Joined: 27.01.2013 Posts: 8,230
      The stats RubbyDubby posted look like they were just meant to give you some kind of an idea what we could interpret from those stats.
      First there is a guy with huge VPIP but very small PFR which means he is a passive fish.
      Second we have quite standard stats, third we have the same stats with 1% 3bet which tells us he might be on the passive side. And so on..

      There are different type of fish with different leaks playing with different hands, so we can't always just say that villains VPIP/PFR must be this and then we can put him in that category. He could have preflop stats of a good reg but then his postflop AF might be very low and he is folding 80% vs cbets.

      About those leak tracker stats, not sure if I would really trust those. I don't know how it works, how it has figured out optimal values for different stats and how often it is being updated.. But a healthy PFR value of 13-20% at 6max seems off. I would say 13 is too tight and over 20 would be just fine.