Posted by Dr Fro 10:18 PM

If you see two guys on TV with percentages next to their hole cards of 57% and 43%, do you think that is each of their percentage chances of winning the hand? Do you think they indicate that the guy with the 57% has a higher EV than the guy with the 43%?

Really?

Because I don't.

First of all, understand that ESPN shows you the percentage chance of a hand being the best hand in the event of a showdown.

There are two reasons why these percentages can be a misleading indicator of who the favorite is.

The first is because, as we know, the best hand does not always win. The uninformed fag at this point says, "yes, because of bluffing." That is why he is an uninformed fag. Yes, bluffing is one reason why the hand with the highest ESPN percentage might lose. So is semi-bluffing. And despite the fact that the two concepts seem to only be separated by a prefix and a hyphen, they have very little to do with each other in either purpose or expected result. Furthermore, sometimes the better hand has to fold to an inferior hand because, all else equal, the better hand has worse position. The ESPN percentages do not care what your position is. Your percentage chance of winning the hand does. Often late position is an advantage. Heads up in a race, it is good to act first. If you go all-in, you are now much more likely to win the hand because your opponent might fold.

So for starters, the percentages do not indicate the percentage chance of winning the pot.

Even if the percentages could possibly go through all permutations of possibilities and accurately assign probabilities to what each player might do in response to each other's actions, that would still only tell us the percentage chance of winning the pot - a more relevant measure than the percentage chance of having the highest hand in the event of a showdown - but still not a complete measure of the EV of a hand.

This is because, as we know, some people will do a better job of eking more out of winning hands and cutting losses on losing hands. So even if you have a percentage chance of winning a hand of, say, 55% against Daniel Negreanu, I would say that your EV is probably negative. He will take your shirt off your back when he wins and not pay you off win you win. So your EV is = +(55% * not much) - (45% * a lot). Sucks to be you.

Although everything I have written thus far is intuitively obvious, I don't think people really think through this properly in poker.

For instance, people will say, "I think he has XX, and that makes me a 3:2 favorite, so I called." What they should say is "I think he has XX, and that makes me a 3:2 favorite to win the hand in the event of a showdown. I think that if I call, he will shove all in and I won't be able to call and get to a showdown - bad for me. Furthermore, he is a more seasoned player and he won't pay me off if that third spade comes. So, I need to adjust my calculation to account for the fact that my EV is less than what I might suppose based solely on my chance to win the hand in the event of a showdown."

Of course, the opposite can happen. If a very good player is playing with a table full of numbnuts, he is right to call pre-flop with all sorts of junk because he knows that his post-flop playing abilities make him an EV favorite with just about any holding.

So it is important to not only put your opponent on a hand (or a range of hands), you have to think through how the hand might play out on subsequent streets to determine what your next action is. Whether you are "ahead" or "behind" is not as important as what will happen on subsequent streets in determining your EV. After all, if you wanted to maximize the number of showdowns you win, the optimal strategy is to check/call every street. If you want to maximize the number of pots you win, you should go all-in pre-flop every street. But these aren't your goals, making money is. Adjust your decisions accordingly.

1 Comment(s):

Posted by  Johnnymac, at 2:37 PM, August 05, 2007  

Excellent post. I think you may be reading too much into the percentages - it's simply just intended as a measure of relative hand strength, after all - but otherwise your point about anticipating your opponent - playing the player - is spot on. Good cards are always important, but the reason good players win money, just as you said about Daniel Negraneau, is because they win more on the pots they win than they lose on the pots they lose. Yes, you can learn good starting hands and betting strategy from a book, but as you like to point out, anyone can read books and most people at this point have. To transcend the "book smart" player you have to raise your game to another level and hone your people skills and the only way to do that is through actual hours spent sitting at the table. Good post.

__________________________________________________

Post a Comment