Wrong EV calculation on 3-way PF AI

iPoker - 1,400/2,800 NL - Holdem - 7 players
Hand converted by Holdem Manager 3

MP: 32,894 (11.7 bb)
CO: 127,742 (45.6 bb)
BTN: 1,623 (0.6 bb)
Hero (SB): 52,708 (18.8 bb)
BB: 2,997 (1.1 bb)
UTG: 65,377 (23.3 bb)
UTG+1: 167,864 (60 bb)

7 players post ante of 220, Hero posts SB 1,400, BB posts 2,777

Pre Flop: (pot: 5,717) Hero has Khttps://pokertools.holdemmanager.com...ons/heart4.gif Khttps://pokertools.holdemmanager.com...ons/spade4.gif
UTG calls 2,800, UTG+1 calls 2,800, fold, CO calls 2,800, fold, Hero raises to 52,488, UTG raises to 65,157, 2 folds

Flop: (114,893, 3 players) 4https://pokertools.holdemmanager.com...ons/spade4.gif 3https://pokertools.holdemmanager.com...s/diamond4.gif 5https://pokertools.holdemmanager.com...cons/club4.gif

Turn: (114,893, 3 players) 4https://pokertools.holdemmanager.com...cons/club4.gif

River: (114,893, 3 players) 2https://pokertools.holdemmanager.com...cons/club4.gif

Results: 114,893 pot (0 rake)
Final Board: 4https://pokertools.holdemmanager.com...ons/spade4.gif 3https://pokertools.holdemmanager.com...s/diamond4.gif 5https://pokertools.holdemmanager.com...cons/club4.gif 4https://pokertools.holdemmanager.com...cons/club4.gif 2https://pokertools.holdemmanager.com...cons/club4.gif

BB shows 4https://pokertools.holdemmanager.com...s/diamond4.gif Ahttps://pokertools.holdemmanager.com...ons/spade4.gif: (Straight, Five High)
(Pre 11%, Flop 23%, Turn 75%)

UTG shows Ahttps://pokertools.holdemmanager.com...cons/club4.gif Khttps://pokertools.holdemmanager.com...cons/club4.gif: (Flush, Ace High)
(Pre 23%, Flop 14%, Turn 23%)

Hero mucks Khttps://pokertools.holdemmanager.com...ons/heart4.gif Khttps://pokertools.holdemmanager.com...ons/spade4.gif: (Two Pair, Kings and Fours)
(Pre 65%, Flop 63%, Turn 2%)

UTG wins 114,893

On this hand the All-In adjusted difference shows 0.

We need the original hand history file sent to us for testing, with screenshots/details of the report stat errors, as instructed at the end of this FAQ.

edit: Was the Big Blinds post an all-in? That could be why.

Please zip and email the original problem hand histories and tournament summaries to us as instructed at the bottom of this FAQ with a link to this thread and your forum name. Make sure you also include screenshots of your reports/filters and a detailed description by tournament number, session, hole cards, filter, stat, etc. of some of the errors we need to test.

Quote:

---

Originally Posted by fozzy71

We need the original hand history file sent to us for testing, with screenshots/details of the report stat errors, as instructed at the end of this FAQ.

edit: Was the Big Blinds post an all-in? That could be why.

Please zip and email the original problem hand histories and tournament summaries to us as instructed at the bottom of this FAQ with a link to this thread and your forum name. Make sure you also include screenshots of your reports/filters and a detailed description by tournament number, session, hole cards, filter, stat, etc. of some of the errors we need to test.

---

The big blinds post was indeed an all-in.

Quote:

---

Originally Posted by SmallFry

The big blinds post was indeed an all-in.

---

I do not see any related open issues in our system so please send the original hand history file, screenshots, details of the error and a link to this thread as instructed in the post/link above.

Quote:

---

Originally Posted by fozzy71

I do not see any related open issues in our system so please send the original hand history file, screenshots, details of the error and a link to this thread as instructed in the post/link above.

---

I made a ticket about the issue a few days back and attached the HH there.

Here is another hand from today with the same issue. 3-way All-In preflop. I had around 50% and my opponents had around 20% and 30% respectively, yet the All-In Adj Diff shows 0 and my All-In adjusted graph, the orange one goes down around 5bb.
iPoker - 150/300 NL - Holdem - 8 players
Hand converted by Holdem Manager 3

MP+1: 1,417 (4.7 bb)
CO: 524 (1.7 bb)
BTN: 8,973 (29.9 bb)
SB: 32,789 (109.3 bb)
Hero (BB): 11,993 (40 bb)
UTG: 4,386 (14.6 bb)
UTG+1: 3,616 (12.1 bb)
MP: 9,684 (32.3 bb)

8 players post ante of 24, SB posts 150, Hero posts BB 300

Pre Flop: (pot: 642) Hero has Khttps://pokertools.holdemmanager.com...cons/club4.gif Ahttps://pokertools.holdemmanager.com...s/diamond4.gif
UTG calls 300, fold, MP calls 300, MP+1 raises to 1,393, fold, BTN calls 1,393, SB calls 1,243, Hero raises to 11,969, 3 folds, SB calls 10,576

Flop: (27,516, 3 players) 4https://pokertools.holdemmanager.com...ons/heart4.gif 5https://pokertools.holdemmanager.com...ons/heart4.gif 2https://pokertools.holdemmanager.com...ons/heart4.gif

Turn: (27,516, 3 players) 5https://pokertools.holdemmanager.com...s/diamond4.gif

River: (27,516, 3 players) 4https://pokertools.holdemmanager.com...s/diamond4.gif

Results: 27,516 pot (0 rake)
Final Board: 4https://pokertools.holdemmanager.com...ons/heart4.gif 5https://pokertools.holdemmanager.com...ons/heart4.gif 2https://pokertools.holdemmanager.com...ons/heart4.gif 5https://pokertools.holdemmanager.com...s/diamond4.gif 4https://pokertools.holdemmanager.com...s/diamond4.gif

MP+1 shows Qhttps://pokertools.holdemmanager.com...ons/spade4.gif 4https://pokertools.holdemmanager.com...cons/club4.gif: (Full House, Fours full of Fives)
(Pre 30%, Flop 35%, Turn 57%)

SB shows Ahttps://pokertools.holdemmanager.com...cons/club4.gif 7https://pokertools.holdemmanager.com...ons/heart4.gif: (Two Pair, Fives and Fours)
(Pre 20%, Flop 52%, Turn 32%)

Hero shows Khttps://pokertools.holdemmanager.com...cons/club4.gif Ahttps://pokertools.holdemmanager.com...s/diamond4.gif: (Two Pair, Fives and Fours)
(Pre 50%, Flop 13%, Turn 11%)

MP+1 wins 6,364
SB wins 10,576
Hero wins 10,576

What is your 6 digit support ticket number? Here is a direct link to your My Tickets page.

Please update your ticket and attach the original HH file that contains that new hand, and screenshots/details of the hand and issue. All original hands that are auto-imported get moved to an archive for performance reasons. Your \HM3Archive is in C:\HM3Archive by default, but you may have put it anywhere. The archive should be organized by \Network\Month\DayOfMonth (example: C:\HM3Archive\SiteName\2016\07\31).* It is configured in the Tools - Settings - Import menu.

When sending any files please zip them (select all files with CTRL+A > right-click > send to > compressed folder) before attaching them to an email. If the attachment is larger than 10MB please upload them to a file hosting site wetransfer or filedropper and copy/paste the download link directly into the support ticket email.

Quote:

---

Originally Posted by fozzy71

What is your 6 digit support ticket number? Here is a direct link to your My Tickets page.

Please update your ticket and attach the original HH file that contains that new hand, and screenshots/details of the hand and issue. All original hands that are auto-imported get moved to an archive for performance reasons. Your \HM3Archive is in C:\HM3Archive by default, but you may have put it anywhere. The archive should be organized by \Network\Month\DayOfMonth (example: C:\HM3Archive\SiteName\2016\07\31).* It is configured in the Tools - Settings - Import menu.

When sending any files please zip them (select all files with CTRL+A > right-click > send to > compressed folder) before attaching them to an email. If the attachment is larger than 10MB please upload them to a file hosting site wetransfer or filedropper and copy/paste the download link directly into the support ticket email.

---

The ticket number is 687398 and the status of the ticket is Waiting on Support.

Quote:

---

Originally Posted by SmallFry

The ticket number is 687398 and the status of the ticket is Waiting on Support.

---

I found the ticket but am not able to import the hands right now. I will try and make sure I get those hands imported and reviewed this weekend and discuss them with our QA/Testing Manager next week.

The first KK hand you posted might be a bug, but I am not sure. I think that is currently caused by the short-stack Big Blind being all-in, then 2 of the players that called his blind 'all-in' by limping the full big blind amount yet later fold to the all-in shove and re-shove from players behind them. We generally do not calculate All-In Adj stats on a hand if someone calls a short-stack all-in but then folds* to a raise/shove behind them (even if on the same street) but in this hand the 2 limpers called the full big blind amount which was more than the big blind's all-in so I am not sure if we should or shouldn't calculate All-In Adj values for this hand.

In the AKo hand MP+1 shoves all-in as a short-stack and the BTN and SB calls the short-stack all-in but then Hero shoves all-in and the BTN and SB fold.

HM3 uses the same equity code as PT4 now which is more accurate and less biased than the code we used for equity in HM2. In a hand where a player is all-in but other players are not, net adjusted results are not calculated. When one player is all-in preflop and other players are not, then they usually have the option to act postflop and this means that sometimes one of them will fold. Having the option to fold* after another street is dealt means that in the long run there will be more strong hands which get to showdown and less weak hands. If we were to calculate net adjusted results in hands like this it would introduce a systematic bias (because there will be more known strong hands) so we exclude all hands like this from net adjusted calculations. You can read more about this here.

*Note: Attempting to calculate adjusted results when unknown but folded cards already called the all-in also introduce a bias into the results so HM3 does not make these calculations in any situation where a player calls the all-in and later folds (even if it's on the same street).