2

I'm trying to find a way to assess how good my postflop hands are in my open source pokerbot. Normal equity calculation certainly gives an indication how good the hands are if we play until the river but I'm looking for a better alternative.

One particular approach I've heard about is to put that equity in relation to the equity that I would have with my postflop range vs. the opponent's postflop range.

It would mean we have the normal equity (let's call it absolute equity) that gives me the equity of my cards vs. the opponent ranges

Then we have the equity of my postflop range vs. my opponent postflop range (let's call this range equity).

Then as a finaly step we would take the relation between the absolute equity vs the range equity (we can call this relative equity).

What are your views on that metric? Would there be any merit in using it? If so, under what circumstances?

  • I'm intrigued. Could you include an example of a hand to make it more clear what you're getting at? – Dr.DrfbagIII Nov 17 '16 at 13:59
  • happy to check in more detail: best join me on google hangouts with dickreuter@gmail.com – Nickpick Nov 17 '16 at 14:09
  • Not clear to me. Don't follow and what I follow I don't see value. Why use the term post-flop range? They play the hand or not before they see a flop. – paparazzo Nov 17 '16 at 15:21
  • Post flop because I assess their (and my) range from the preflop play. Once the flop comes I then assign a range to each player and run a montecarlo simulation to calculate the equity based on the ranges of each player, including myself. This is then compared with my actual cards vs. the ranges of the other players at the flop. – Nickpick Nov 17 '16 at 15:23
  • That is one approach. If you know your hand then not seeing how a bot considering your range would add any value. – paparazzo Nov 17 '16 at 15:55
1

The metric is a ratio of what I think my equity is vs what the foe think my equity is. The higher the disparity the more volatile or different the perception of the outcome is to each ones perspective. It can be used, but it is definitely not even remotely worth it.

My opinion is because:

  • The ratio does not add anything above existing metrics. Why use 2 when you have 1 and 1. What will you do with a high ratio where someone (could be you) is off by a bit with their expected equity? This can probably help you determine doing a check/raise at most.
  • you have your absolute equity, which you use for your EV/strategy. This adds most value.
  • you could not be profiling/modelling your opponent correctly as it is a range; and a way more valuable improvement would be to correctly set that range of your foe (for your absolute equity calc).
  • you assume your opponent measures/profile you exactly the same way. He does not. And acting on thinking what you think your opponent thinks when he does not think what you think he thinks is worthless, e.g. you cannot really act on the range equity.

PS paparazzi is correct in that you can already set the range when game starts. Even from then (meaning even before the flop) you can start refining his hand from betting behaviour. You don't change the range, you only order it by equity (and aggregate the equity with weights for final ev).

-1

Ok Tjorriemorrie. So lets say we raise from BTN and BB calls.

Flop comes AK4 .

Which range has the advantage here ? Which betsize should we use on that flop ? Which hands can our opponent have and which hands can we have on that board ? Do we act upon the actual hand we have or the actual range we can have ? Is the range

  • But you do you know your hand - there's no advantage to be ignorant thereof. Did you finish this answer though? – Tjorriemorrie Dec 19 '16 at 10:28
  • Is this meant to be a comment on @Tjorriemorrie post? – Toby Booth Dec 19 '16 at 23:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.