Poker Snowie Experiments (part 3)
Posted by Stelios Serefidis
Video Player is loading.
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Resume Video
Start from Beginning
Watch Video
10
Loading 9 Comments...
The early 7h5h hand was a 2 pot ✔️⬆️ not a 1/2 or pot size according to Snowie. It is important to pay attention to snows 2 pot plays they reveal a lot about its strategy. It is especially interesting when we can note what the balancing plays are -- they are available in Snowie's range information.
it s kind of strange in this hand. It mentions that 2pot is the best sizing but still it shows we should flat rather than raise 2pots. This might mean we should develop a 2pot raising strategy in this flop and thus flat 57hh and raise higher flushes!
Its suggested sizing was 2 pots.
Watch carefully when I click evaluate on the flop, it pops up in the 2pots tab and shows call as the highest ev option despite the fact that 2pots is the recommended sizing
yep, it want's that strategy not that hand. Sorry for the mistake its hard to see sometimes on the video. I suppose it 2 pot ✔️⬆️ slightly stronger flushes since that is likely the cusp hand.
Also, I really enjoyed these videos.
One thing to mention is that you should clarify your purpose in making and explaining your plays. Exploring the optimal strategy with Snowie's allowed bet sizes is a great reason. But note that if you believe Snowie has converged its strategy and has found a global minimum (or local minimum close to the optimal ) then you literally cannot beat Snowie unless you employ a strategy with different bet sizes. It has the optimal strategy in the space of 0.5, pot and 2pot allowed sizing strategies. So in a game with only those allowed sizings this is optimal -- nothing more can be done non-collusively.
Whether the algorithm has found that strategy is a distinct question. You will even find spots where Snowie shows a 2 pot bet as higher EV and chooses to bet pot. I presume this is because the bigger bet is not able to be balanced in the solution space, i.e. it doesn't maximize value for the range which is the optimization function.
Thanks again. Clever series.
I am not trying to beat snowie, my purpose is to see if I do any plays that snowie considers as big mistakes and try to find out why I m wrong, that's why I stick to the given sizings.
Be the first to add a comment
You must upgrade your account to leave a comment.