PDA

View Full Version : $EV calculated wrong



allwind
01-15-2009, 11:21 PM
Just had the following hand. HEM gets the $EV calculated wrong. Is this something that every time there is a river, or action on the river?

Grabbed by Holdem Manager (http://www.holdemmanager.net)
NL Holdem $4(BB) Absolute
Hero ($817)
BB ($537)
CO ($1,029)
BTN ($994)

Dealt to Hero Q:diamond:K:diamond:

CO raises to $12, fold, Hero calls $10, fold,

FLOP ($28) 3:diamond:T:club:9:spade:

Hero bets $20, CO raises to $60, Hero calls $40,

TURN ($148) 3:diamond:T:club:9:spade:8:diamond:

check, check,

RIVER ($148) 3:diamond:T:club:9:spade:8:diamond:6:diamond:

check, CO bets $120, Hero raises to $745, CO calls $625,

Hero shows Q:diamond:K:diamond:
(Flop 15.9%, Turn 22.7%)

CO shows 3:spade:3:heart:
(Flop 84.1%, Turn 77.3%)

Hero wins $1,635

morny
01-16-2009, 01:49 PM
I dont see whats wrong here?

btw you cant calculate EV on the river as there is no more cards to come so your either 100% or 0%

allwind
01-17-2009, 03:58 AM
Here is my calculation to $EV for this spot. HEM does not give percentages preflop, which btw, I should HEM should.

PF: (0.508-0.492)x12= +0.19
Flop: (0.159-0.841)x60= -40.92
Turn: 0
River: (1.00-0)x745= +745

So my $EV is
0.19-40.92+0+745= +704.27

My profit was 817. So the $EV diff should be.
704.27-817= -112.73

I had a luck of 112.73$ in this pot. HEM gets it to even. Obviously to me, HEM is wrong. Is there something I have gotten wrong?

morny
01-17-2009, 03:09 PM
There is no all in EV calculated in HM unless your all in before the river, since there is no cards to come on the river the equity is either 0% or 100% thus the EV diff is also not counted so its not working out that its breakeven it just didnt calculate anything at all for it

allwind
01-17-2009, 04:24 PM
Sorry, butt my calculation takes part in others than river. Seems like your calculation thinks river is the nut solution for everything. Sorry, but HEM $EV is wrong.

morny
01-17-2009, 04:33 PM
I see your calculation but HM dosent calculate the overall streets it only goes by street it got all in on, its already been suggested in the suggestions forum to have a system where it calculates like youve done so you can vote for it here if you like: http://holdemmanager.uservoice.com/

allwind
01-17-2009, 10:45 PM
Hello,

Regarding voting for change. I need to sign in for something, where it does regegnize my username etc. So I dont do that already, that uservoice. Instead I will post my thoughts here.

I understand $EV is only regardless the street where the money gets all in. Now all the numbers in HEM are correct. HEM calculates them all correct. Only the $EV should be calculated differently. This is what I think.

However, I do believe the $EV should be calculted in an other way. I do believe $EV should include all streets. I also do believe it is interesting to look at seperately how many "$folded" (to your opponents) and how many "$bet" where you bet your opponents out of the pot. And of course with the ability to be expressed in bb/100 as well. Preferable for each street. For the first part it is very important to be the $EV calculation will include all streets.

Also there is an other bug. Lets say there is a short stack in the game. The short stack gets all in preflop. When we get money in the pot preflop we have the best hand. On the flop we get the money in good against the third player in the pot. On the flop we have the better hand than the big stack, but the shortstack now beats us. HEM gets confused with this $EV calculation and gets it way wrong. It gives me a negative $EV. But the truth is, in this pot, I only had the money in with positive expected outcome.

Finally to sumarize.

* I want $EV calculation to include actions on all streets.
* I want a new stat called $bet.
* I want a new stat called $folded.

allwind
01-18-2009, 03:10 PM
Grabbed by Holdem Manager (http://www.holdemmanager.net)
NL Holdem $4(BB) Absolute
SB ($994)
Hero ($829)
UTG ($912)
CO ($545)
BTN ($813)

Dealt to Hero K:spade:J:spade:

fold, fold, BTN raises to $14, fold, Hero calls $10,

FLOP ($30) Q:diamond:3:diamond:Q:spade:

check, BTN bets $30, Hero raises to $92, BTN calls $62,

TURN ($214) Q:diamond:3:diamond:Q:spade:2:diamond:

check, check,

RIVER ($214) Q:diamond:3:diamond:Q:spade:2:diamond:K:club:

check, BTN bets $125, IGNORING MY RAISE OBTION, NOW 3 THINGS CAN HAPPEN.

1) I FOLD AND MY $EV WILL BECOME -106 FOR THIS HAND.
2) I CALL AND LOSE. MY $EV WILL BE -125$
3) I CALL AND WIN. MY $EV WILL BE +125$

CORRECT?

morny
01-18-2009, 03:35 PM
Im not sure what you mean? As i mentioned HM only calculates EV when your all in on the turn or before at the moment. Unfortunately leaving the suggestion here will mean it will disappear and be forgotton about, voting for it on Uservoice will increase the chances of it being implemented quicker if it gets more support.

Mike chops
01-19-2009, 01:25 PM
We are working on the all-in with multiple pots. There is a bug because we are currently using the win% for all the players and applying that to the most they can win. We should be calculating equity in each side pot.

The all-in EV stat is an attempt to measure your luck for pots when you arer all-in. What happens after then is completely luck - there's no way for anybody to outplay someone else. It isn't the only source of luck in poker by a long shot, but it is easy - well relatively easy - to calculate, because it doesn't involve making judgements on how poeple will play.

Calculating EV on the turn for someone with say a flush draw, does involve making a judgment about the implied odds.

peter
02-21-2010, 12:20 AM
This is quite annoying. Here is an example:

.........................................
Hero ($10 in chips)
P7 ($30 in chips)
P8 ($30 in chips) Dealer
Hero: posts small blind $0,25
P7: posts big blind $0,50
Hero [5d,Qd,Qh,4d]
P8: calls $0,50
Hero: calls $0,25
P7: checks
buko***9: calls $0,25
bald***1: checks
*** FLOP *** [Ks,8h,Qc]
P7: checks
Hero: checks
P8: checks
*** TURN *** [3h]
P7: checks
Hero: bets $1,50
P8: raises to $3
P7: calls $3
Hero: is all in $9.5
P7: calls $6.5
P7: calls $6.5
*** RIVER *** [9h]
P7: is all in $20
P8: calls $20
Hero: shows [5d,Qd,Qh,4d]
P7: shows [7h Jh 5s As]
P8: shows [Ah 2h Js Qs]
***SHOW DOWN***
P8: wins $70 with Flush, Ace high
.........................................

If this happens, HM should show me positive EV as I was favorite when I went all in and had much over 33% chance to win the pot. Problem is that two players didn't go all in on turn, so HM doesn't show anything in EV $ diff and my EV graph goes down by $10.

As I am often playing Omaha with 50BBs, this kind of hands happen very often and it messes up my EV calculations. One 3-way pot can send my EV graph completely in wrong direction. Hopefully it's fixed soon as it shouldn't be too complicated to have EV for side pots.

Patvs
02-21-2010, 01:10 AM
Oh.. it is complicated...
But a different approach to EV... using a "EV per street" calculation will be added in the near future... (will take months though)

allwind
02-21-2010, 02:44 AM
Oh.. it is complicated...
But a different approach to EV... using a "EV per street" calculation will be added in the near future... (will take months though)

Please let me know if I can be of any help.

sc78
08-10-2010, 07:35 AM
Seat 2 is the button
Seat 1: ( $1490.00 USD )
Seat 2: hero ( $1510.00 USD )
hero posts small blind [$50.00 USD].
seat 1 posts big blind [$100.00 USD].
** Dealing down cards **
Dealt to hero [ Td Kc ]
hero raises [$150.00 USD]
seat 1 calls [$100.00 USD]
** Dealing Flop ** [ 7c, Th, 5s ]
seat 1bets [$200.00 USD]
hero raises [$500.00 USD]
seat 1calls [$300.00 USD]
** Dealing Turn ** [ 6s ]
seat 1 bets [$200.00 USD]
hero raises [$810.00 USD]
seat 1calls [$590.00 USD]
hero wins $20.00 USD
** Dealing River ** [ Tc ]
seat 1 shows [5h, 2h ]
hero shows [Td, Kc ]
hero wins $2980.00 USD from main pot

So.. My question is that what is it that I am not getting here? HEM shows 11% for the 52 on turn on which we went all-in, which is about correct. So that equals 790 (what seat 1 has left on turn) times 2 for the total (all-in) pot of 1580 for the calculation of EV. Still the EV diff shows 339 chips of difference. that would be about 21% for seat 1 would it not? So the correct EVdiff would be 174 chips? Or am I just calculating it wrong? I have tons of examples in my hand histories of the same kind of strange discrepancy. Is my EV graphs totally wrong in my data because of this or I have not understood something? I d really appreciate if someone could help me out.

Mike chops
08-10-2010, 10:12 AM
The history you posted is missing the game number and buy-in information. Can you please post the full hand history so we can import it and see for ourselves.

sc78
08-10-2010, 01:49 PM
ok, so here is the full history for the hand.

PokerStars Game #47957407622: Tournament #299906381, $33.00+$1.50 USD Hold'em No Limit - Match Round I, Level IV (50/100) - 2010/08/10 13:51:44 EET [2010/08/10 6:51:44 ET]
Table '299906381 1' 2-max Seat #2 is the button
Seat 1: chuff (1490 in chips)
Seat 2: score (1510 in chips)
score: posts small blind 50
chuff: posts big blind 100
*** HOLE CARDS ***
Dealt to score [Td Kc]
score: raises 100 to 200
chuff: calls 100
*** FLOP *** [7c Th 5s]
chuff: bets 200
score: raises 300 to 500
chuff: calls 300
*** TURN *** [7c Th 5s] [6s]
chuff: bets 200
score: raises 610 to 810 and is all-in
chuff: calls 590 and is all-in
Uncalled bet (20) returned to score
*** RIVER *** [7c Th 5s 6s] [Tc]
*** SHOW DOWN ***
chuff: shows [5h 2h] (two pair, Tens and Fives)
score: shows [Td Kc] (three of a kind, Tens)
score collected 2980 from pot
chuff finished the tournament in 2nd place
score wins the tournament and receives $66.00 - congratulations!
*** SUMMARY ***
Total pot 2980 | Rake 0
Board [7c Th 5s 6s Tc]
Seat 1: chuff (big blind) showed [5h 2h] and lost with two pair, Tens and Fives
Seat 2: score (button) (small blind) showed [Td Kc] and won (2980) with three of a kind, Tens

Patvs
08-10-2010, 04:40 PM
1490 (effective stack)
Allin on the turn:

[Td Kc] (88.64%) vs [5h 2h] (11.36%)

board [7c Th 5s 6s] [Tc]

---> So your cEV is: (88.64% * 1490) - (11.36% * 1490) = +1151
You actually won 1490--> so the EV Diff is 1151 - 1490 = -339


All looks perfect.



Read: For a long time I thought it was unfair (and I wanted a "EV per street" calculation) until I read: http://forums.holdemmanager.com/3rd-party-programs-compatible-holdem-manager/16665-enter-sect-7.html


Summary:

The EV "problem" what some people don't understand:
A: If a player has 0 outs, or
B: the allin situation takes place on the river there is NO EV DIFFERENCE.
C: If a shortstack goes allin preflop, and is called by two bigstacks. And the two bigstacks continue to bet on the flop, turn or river, this situation is treated as situation B. (EV = 0)
Fozzy: "You can't calculate all-in equity if you don't know the hands you are up against."
D: If you commit 99% of your stack with the best hand, but your last 1% goes allin with the worst hand, the $EV Difference will be calculated by your entire stack.

Why "EV by street" (which people who often see situation D want) is a bad thing:

Example:
- you have AA, you raise to 80% of your stack, donkey calls, flop comes K83 rainbow.
- you then go all-in, no matter the flop, because you're committed.
- out of 100 times, 88 times donkey folds.
- 12 times donkeys calls with a set (33/88/KK).
What shall EV by street wrongly do? It shall do no computations for the 88 times where donkey folded--> "no more calculation".
What shall EV by street do the 12 times where donkey calls with a set? "Show that donkey sucked out and that you got unlucky".

So although you ran obviouslyreally good by having donkey folding 88 times out of 100. EV by street focuses on the 12 times where donkey hit his set and tells that you're running below EV.
This is a well-known gambler fallacy. And this is why "EV by street" is biased.

Note: Tristanblue writes "it's precisely because EV by street does nothing to your adjusted-graph on these cases where the opponent folded that it is wrong".

But what if your opponent never folds? Suppose there are two players A and B.
Player A has AA, B has KK. (both have $100 stacks). They commit half their stack preflop and the flop comes AK6. Player B (KK has 1 "out") to win the hand.
If I would play this hand I would always make sure I'm allin on the turn.
However Player X always commits the rest of his stack on the flop and turn *except for one dollar*. And he commits on the river.
Of course, 4% of the time, the rivercard is the case King. Player X's EV Diff is always 0. (because he goes allin on the river)
My EV Diff (I go allin on the turn) is -$4 (96 out of 100 times) and +$96 (4 out of 100 times)
So our EV graph actually looks exactly the same after 100 of these hands.
(-4 x 96 + 96 + 4 = also equals 0!) So the EV outcome is the same no matter how you play the hand.

sc78
08-10-2010, 05:07 PM
Thanks for the reply