Difference between BB Won and All-in Adj BB

BuzzKillington

BuzzKillington

Rock Star
Silver Level
Joined
Oct 20, 2017
Total posts
438
Awards
1
Chips
1
So, since around a week, I have been on the losing side. This is okay, but I am trying to understand what is going on, so I can fix it next week.

According to PokerTracker 4, the difference between my loss and my all-in adjusted EV would be around the same amount as my loss, i.e., my all-in adjusted EV is around 0.

Is this really a matter of variance (i.e., bad beats), or could it also be something else?

I don't understand why the difference is this huge.

If the attachment is not readable, then try: https://imgur.com/mhHVyUk
 

Attachments

  • Screenshot from 2017-10-21 04-02-25.jpg
    Screenshot from 2017-10-21 04-02-25.jpg
    15.9 KB · Views: 106
vinnie

vinnie

Legend
Silver Level
Joined
Apr 12, 2013
Total posts
1,208
Awards
1
US
Chips
50
4,000 hands is a really small sample size. But, this is just the way things are sometimes. You will find times when you run under expectation and others where you run above it. In a single session, it can be pretty crazy.

I have had sessions where I ran 9 buy-ins under expectation (that's 900 x bb), and under 2k hands.

This is really not something you fix. Obviously, if your line is positive, that means you are getting the money in good, but the line itself is misleading. Sometimes, you will get the money in at the right price but your EV line won't show that (say you are being offered 3-1 and have 35% equity). If the pot was heads up, it will show a negative line (-30xbb if the effective stacks were 100xbb at the start of the hand). But, if you put the rest of your stack in with the right price, it wasn't really a bad choice.

Anyway, the line itself is meaningless. If you're losing, and the line is going up, then you can be pretty sure you're running badly. But, aside from that, it doesn't really tell you what it "seems" to tell you, which is how good your all-in decision making is.
 
koadyawn

koadyawn

Rock Star
Silver Level
Joined
Apr 23, 2008
Total posts
198
Chips
0
in most simplest words, your all-in EV is just the amount of $$$ you would be at if you won all of your all-ins.
 
A

AlexTheOwl

Visionary
Silver Level
Joined
Aug 22, 2017
Total posts
860
Chips
0
in most simplest words, your all-in EV is just the amount of $$$ you would be at if you won all of your all-ins.

What? No.

PokerTracker itself agrees with Vinnie that this stat is too noisy to mean much of anything:
http://www.pokertracker.com/blog/2011/10/the-problem-with-all-in-ev-all-in-equity

Edit: That article addresses the "unadjusted" stat. PT4's adjusted stat only uses heads-up pots. But Vinnie's concerns about Pot odds still apply. And it's not even a measure of overall "luck", it's a measure of luck in heads-up pots that go to showdown.
 
Last edited:
B

braveslice

Pull-ups!
Silver Level
Joined
Dec 9, 2013
Total posts
1,988
Chips
0
Yes that is form of variance. I don’t really get why people are so against it, and very good players too. I understand that sometimes it says wrong, but any stat is wrong during one hand, it’s the large data sample that makes statistic to work.

250bb is not that big difference though if you think about it as in buy-ins 2.5BI

Essentially it says that you have lost 2.5 flips (something like AK vs Pair) more than won during 4100 hands.

Hand 1 Eq 50%, shove 100bb, result lost; Green line: -100bb, Yellow line 0
Hand 2 Eq 50%, shove 100bb, result lost; Green line: -200bb, Yellow line 0
Hand 2 Eq 50%, shove 50bb, result lost; Green line: -250bb, Yellow line 0
Difference 250bb, same than you
 
BuzzKillington

BuzzKillington

Rock Star
Silver Level
Joined
Oct 20, 2017
Total posts
438
Awards
1
Chips
1
Now that I think about it, you are right that the sample size isn't even that meaningful. Perhaps losing 2.5 times the buy-in isn't that much and it's possible that I'm too obsessed with the idea of having at least a net profit. The fact that it is negative is what bothers me. Not the idea of losing money per se, but the idea that I am still losing despite studying and practicing. Is it the ego? Maybe. Maybe not. I think it's also something else. I've invested so much time into studying this last month and I really want to see that reflected in my game. I feel that I'm improving (my decisions are a lot more sound than they used to be), but I want (more) confirmation from my actual performance. What makes studying meaningful is that you get feedback that your studying and practicing actually contributed to your goal(s). I feel that this is still lacking.

I've been reading about the all-in stat and I see how it is flawed, but like you said, in the end the actual performance should converge towards the all-in adjusted equity. It's just hard to see what the right sample size is. As a beginner, playing thousands of hands is a lot, despite the statistical significance being quite low. I'm still struggling with this idea. I used to play quite some chess. In chess, every game is meaningful in the sense that you can actually say something about a single game and be able to figure out a lot about a player's playstyle and mistakes. In poker there is so much variance that apparently even thousands of hands are not enough to say anything meaningful about performance. This is still hard to grasp.
 
Last edited:
BuzzKillington

BuzzKillington

Rock Star
Silver Level
Joined
Oct 20, 2017
Total posts
438
Awards
1
Chips
1
This is weird. I have been checking out the difference between BB won with and without showdown. The difference is really huge. Could this have anything to do with the difference between BB won and All-in Adj BB?

Larger resolution of the graph: https://imgur.com/a/Pne2a
 

Attachments

  • Screenshot from 2017-10-22 00-45-16.jpg
    Screenshot from 2017-10-22 00-45-16.jpg
    19.1 KB · Views: 66
vinnie

vinnie

Legend
Silver Level
Joined
Apr 12, 2013
Total posts
1,208
Awards
1
US
Chips
50
I will think about the last post in a bit, I am shopping. But -250 BB and break even is literally losing one flip (for 125xbb stacks). Two flips in 4k hands that went different and you could be up 250xbb.
 
R

Rational Madman

Legend
Platinum Level
Joined
Jan 11, 2017
Total posts
2,478
Chips
0
Graphs are to show off, they don't help at all. :)
 
vinnie

vinnie

Legend
Silver Level
Joined
Apr 12, 2013
Total posts
1,208
Awards
1
US
Chips
50
This is weird. I have been checking out the difference between BB won with and without showdown. The difference is really huge. Could this have anything to do with the difference between BB won and All-in Adj BB?

Larger resolution of the graph: https://imgur.com/a/Pne2a

I don't know your stakes, and the player pool at each stake determines how showdown and non-showdown winnings should interact. What I am going to point out is actually pretty obvious, so don't smack me upside the head thinking that I believe this is deep.

When I see a showdown line that low, and a non-showdown that high, it tells me that you're not folding enough once you put money in. You might be playing too many hands, and not giving up on marginal stuff. Or you might be trying to float and bluff too often for the stakes you are at. Or, at the lowest micros, you might be playing fit or fold, and the tables are just passive enough that you never need to fold because your opponents let it check down.

Anyway, you're almost certainly seeing too many showdowns. How you're getting there will determine how to fix the problem, but that is usually the cause of a graph that looks like yours.
 
BuzzKillington

BuzzKillington

Rock Star
Silver Level
Joined
Oct 20, 2017
Total posts
438
Awards
1
Chips
1
Graphs are to show off, they don't help at all. :)
I don't know. I feel that they are helping. Before I would kind of forget about my losses when I won some. Since I was playing at different stakes, the absolute numbers didn't always tell me much about my actual performance. Now with PokerTracker I can see how I actually perform in terms of BBs. Whatever the case, it makes me a lot more self-conscious about my play, which I believe is good.

I don't know your stakes, and the player pool at each stake determines how showdown and non-showdown winnings should interact. What I am going to point out is actually pretty obvious, so don't smack me upside the head thinking that I believe this is deep.

When I see a showdown line that low, and a non-showdown that high, it tells me that you're not folding enough once you put money in. You might be playing too many hands, and not giving up on marginal stuff. Or you might be trying to float and bluff too often for the stakes you are at. Or, at the lowest micros, you might be playing fit or fold, and the tables are just passive enough that you never need to fold because your opponents let it check down.

Anyway, you're almost certainly seeing too many showdowns. How you're getting there will determine how to fix the problem, but that is usually the cause of a graph that looks like yours.
I am currently playing at NL2 and NL5, mostly NL2 since I had experienced quite a drop after a while when playing at NL5.

I believe that you may be right. I have a difficult time folding marginally good hands after the flop, when I hit the board but not very hard, with at least one overcard or a possible flush draw for my opponent(s). And sometimes I miss, and I do a c-bet anyway mostly to test my opponent and sometimes fire another barrel on the turn in order to bluff them off the pot. Sometimes it works and sometimes it doesn't, but especially on the turn it makes me feel uncomfortable about what I'm doing. Other times I check on the turn and the opponent bets, which makes me feel like I had shown weakness, allowing my opponent to bluff me off the pot. But at that point my opponent could just as well have a good hand, or at least a hand that could beat mine with at least 50% pot equity. So I guess this is what forces me into bad decisions, but I'm still not sure when the initial bad decision was made. For example, a c-bet on the flop seems reasonable, even when you don't hit the board. But on the turn it feels like I can't really make a good decision, since checking and betting have both disadvantages (one is showing weakness and the other is putting too much mathematical pressure on myself).
 
R

Rational Madman

Legend
Platinum Level
Joined
Jan 11, 2017
Total posts
2,478
Chips
0
I don't know. I feel that they are helping. Before I would kind of forget about my losses when I won some. Since I was playing at different stakes, the absolute numbers didn't always tell me much about my actual performance. Now with PokerTracker I can see how I actually perform in terms of BBs. Whatever the case, it makes me a lot more self-conscious about my play, which I believe is good.
The HUD helps but the graphs are just there to show off to others or say 'help me' if it's unsteady.

In the end to become the best you can't be obsessed with steady graph you should want burst-like graph with more ups than downs. Lose here and there, get caught bluffing and lose some and then bam... cooler-hand them for big wins again and again until they weep before you.
 
BuzzKillington

BuzzKillington

Rock Star
Silver Level
Joined
Oct 20, 2017
Total posts
438
Awards
1
Chips
1
The HUD helps but the graphs are just there to show off to others or say 'help me' if it's unsteady.

In the end to become the best you can't be obsessed with steady graph you should want burst-like graph with more ups than downs. Lose here and there, get caught bluffing and lose some and then bam... cooler-hand them for big wins again and again until they weep before you.
I get that, but since you'll also be folding a lot of hands, you should encounter bumps just about 10 - 20% of the time? Maybe more in a table with fewer seats.
 
BuzzKillington

BuzzKillington

Rock Star
Silver Level
Joined
Oct 20, 2017
Total posts
438
Awards
1
Chips
1
So, I've been going with the advice to fold more marginal hands and bluff less. Playing tighter overall. I've also moved from 6-max tables to full rings. I've been multi-tabling with 4 tables (so I can see them all at once on my screen and make optimal use of my time and focus without getting bored). This also encourages tight play. My VPIP/PFR went from around 30/25 (6-max) to around 11/10 (full ring).

After around a thousand hands, I noticed a significant drop in my BB Won Without Showdown and a rise in my BB Won With Showdown and my All-in Adj BB. While I still made some silly mistakes, I generally felt a lot better about my performance, and more importantly, my decision-making.
 
Figaroo2

Figaroo2

Legend
Bronze Level
Joined
Sep 9, 2013
Total posts
7,363
Awards
16
Chips
13
Yeah your were playing the brute force game. Forcing out all their weak holdings which means by the river you are only facing strong holdings. It's actually a good sign for a beginner as it means you have the inate aggression that comes with being a great player. If you start off too passive its much harder to raise your aggression to what is required. All you need to learn is how to channel your aggression into the right spots.
Check out Alucard's cash game thread where the high red line low blue line dynamics have recently been thrashed out
Don't worry about your all in ev line you can do precisely nothing about it if you are playing decent poker. Ive mostly turned it off.
In an attempt to stop losing you have turned into a full ring nit, you won't lose as much in the short term but you wont win much either. Look up some opening range charts. Full ring you want to be aiming for 16 vpip 14 pfr for decent tight aggressive stats.
 
Last edited:
BuzzKillington

BuzzKillington

Rock Star
Silver Level
Joined
Oct 20, 2017
Total posts
438
Awards
1
Chips
1
Yeah your were playing the brute force game. Forcing out all their weak holdings which means by the river you are only facing strong holdings. It's actually a good sign for a beginner as it means you have the inate aggression that comes with being a great player. If you start off too passive its much harder to raise your aggression to what is required. All you need to learn is how to channel your aggression into the right spots.
Check out Alucard's cash game thread where the high red line low blue line dynamics have recently been thrashed out
Don't worry about your all in ev line you can do precisely nothing about it if you are playing decent poker. Ive mostly turned it off.
In an attempt to stop losing you have turned into a full ring nit, you won't lose as much in the short term but you wont win much either. Look up some opening range charts. Full ring you want to be aiming for 16 vpip 14 pfr for decent tight aggressive stats.
Thanks, man. I have tried to find a balance between being a nit and being overly LAG this time. My VPIP/PFR was around 19/17 on average this time (on a full ring table). I won some and I lost some, but most importantly to me, I have -- at least for the time being --overcome the net loss, even though my profit is very marginal at this point (below 1/4 of a buy-in at NL2). It seems that my intensive studying and practicing is paying off.

Larger resolution of graph: https://imgur.com/a/SMzd9
 

Attachments

  • Screenshot from 2017-10-24 00-40-43.jpg
    Screenshot from 2017-10-24 00-40-43.jpg
    17.8 KB · Views: 31
B

braveslice

Pull-ups!
Silver Level
Joined
Dec 9, 2013
Total posts
1,988
Chips
0
So observations,
In your graph you seen green line having different phases, those all called trends. For starting player learning is fast, and so are trends, if you look (first picture) hands from 2739 to the end 4102, you actually see that you made profit a bit so good job. It’s no way trustworthy result given sample size, but trends contain much more information than looking just last number net profit.

“ Could this have anything to do with the difference between BB won and All-in Adj BB?” Yes, only saw showdown hands are calculated to All-in Adj, because when you are AI vs AI you see the showdown 100%. But as you can see, it has no value to compare. You can’t change luck, so anything you do will not correct AI-adjusted line difference to actual winnings. Maybe you could pray the gods though, hmm... Gods are not famous about breaking universal laws just for anybody though, so if you manage to do that I would guess they write books about you.

Again, second Image if you study trends, your red line has been lightly going down last 3k hands witch is good, only first 750 hands or so you were over aggressive. Green line going up, your showdown value has improved a lot, see the trend it’s almost flat. Now actually you have been a bit unlucky (190bb) so actually, tadaa, your blue line should be that much more higher, making All-in adjusted blue line just fine.

I didn’t look your 3rd figure, but it can be already seen from first two that your adaptation is already made you profitable player during that last 1,4k session. Now just try to crap that flow and feel and continue it.

EDIT to add, even though it’s very very shitty feeling to be unlucky, the law of averages will finally rescue you, thus just play more hands. And try not to think law of averages when you are on heater :D
 
Last edited:
Top