PDA

View Full Version : EV Report (Long)



Makonnen
08-29-2008, 05:25 PM
I am looking to create a better EV report. Actually, this may be available already--if so, I can't find it. But, here is what I am looking for:

A report that will factor EV per street for all opponents whose holecards are known at the end of the hand. $ put in by opponents whose holecards are unknown remains unaccounted for in the report. The motivation behind this is that, especially at tables > 100 BB deep, the key bet is not the one that puts both players all in. As such, one's All In EV is often distorted. Here is how it would work, taking first a fairly simple example, putting the blinds at $1/$2 just to make the math easier (please, no comments on the hand, it is not meant to be either real or well-played).

Villain is the BB with 8h9h, Hero is the SB with Ah9s. Say there is no action preflop. There is now $3 in the pot. Hero has $EV of $2.02, Villain has $EV of $0.98 (all calcs from PokerStove).

Flop comes 9d5hTh. Hero bets $2, Villain raises to $8, Hero calls. So $16 was put in on the flop, with equities of 57.67% and 42.32%. Hero gets an additional $9.23, Villain an additional $6.77. Turn is 4h, another $50 goes into the pot, giving Hero another $7.95 in EV, Villain another $42.05.

River is the 7h, Hero goes all in for $54 more, Villain calls. Hero gets all $108of the $EV there, of course, but his $EV for the hand is only $127.19, compared to $49.81 for the Villain.

This more accurately reflects the flow of money in the hand, including the part of the pot Hero put in while well behind.

Rather than going through more examples, which I have and can do, I'll stop here. Is this possible in HEM? Does it exist now? If it is possible, but does not exist, I'd be happy to contribute some programming/database time towards making it happen, as I think this would be an invaluable tool. At least, for me it would ...

Rvg72
08-31-2008, 02:01 PM
Hi, would you want this just for all-in hands? For non all-in hands it could become very inaccurate due to the biases.

For all-in hands I can see it having some value. Anyone else want to comment?

Makonnen
08-31-2008, 02:10 PM
I'd actually like it possible for non all-in hands (perhaps a radio button choice). I think that, over a large sample of hands, the bias would be overcome, or, at worst, would be no more than the current report.

The real issue I am trying to get at is that 200BB tables really change where the tipping point of a hand is. At 100BB, it is usually when all the money goes in, so the all in report works fine. At 200BB, it often isn't--the key bet is on the turn, and the all in move on the river is really somewhat unavoidable, as the pot is too bloated to fold at this point.

Actually, what biases are you referring to? Hands have to go to showdown to qualify for the report at all (we need to know their holecards, too), and I'm not sure what bias exists in a hand where a small pot goes c/c on the river?

Rvg72
09-01-2008, 10:19 PM
The bias is that depending on your style the report may identify you as being unlucky or lucky even if that was not the case. There were a bunch of posts about this in the PEV thread and Phil admitted at the end that this method can be very flawed for non all-in hands. For all-in hands I can definitely see a use for this so I'll add it to the todo list

Roy