Is this a useful stat for tracking winrate?

I'm trying to understand whether I'm a winning player at 10NL.

My All-in Adj BB/100 is 10.68 after 4000 hands, but my net won is -$26.33. I feel like I've been running dreadfully, but not sure whether I should move down or play through the downswing. Depends if the All-in Adj BB/100 is relevant or not.

If that is your 'net' from only the All-In hands, then yes, you are running poorly. But are you getting it in 80-20 or 55-45 'all' the time? You can break those hand down even more if you take the time to look at them.

Moving down is more of a bankroll and mind set decision IMO.

What is your net from all the other hands?

Certainly every stat has value, it's just what exactly are you looking for from the stat. You should have a 'general' win rate from your 10NL hands .. that's where you start. Then you go to break-down stats like this one .. and it appears you know how to get your chips in good when All-In. So you probably need to look at other areas of your game where you might be leaking chips .. blind play, too much Flop floating, flatting OOP when you should be 3Bet-folding PF.

Also, 4000 hands is too small for online play .. let's get to 10K first. GL

Cheers.

I've played up to about 10k hands now. This is everything that I've played at 10NL. Bankroll isn't an issue really, I can reload if I need to, but I'd probably be too annoyed to bother. I had to stop because I am mega tilted atm, as you can see from the last bunch of hands. I really don't know whether to continue. I want to continue if it's highly likely I'm a winning player at 10NL, but if I'm just losing I'd rather quit tbh.

I honestly feel like I should just cash out every time and I'd be way up even if I have to pay extra rake. I'd rather that than be constantly punished tbh. I certainly can't take another 20-30K hands of this downswing if that's what I'm going through.

I just need to know if I can be confident that I can carry on with my strategy or not but I don't know how to assess whether I'm a winning or losing player. Any advice?



https://ibb.co/Thg7gWP


I don't know how to make the image show.

all-in ev / 100 is only useful for bad players who want to cry that they're running below EV
if you want to improve, ignore it

you need to tag the link to the actual image, not the web page
in your case https://i.ibb.co/HgHbHhG/111.jpg

right click the img and copy image address

You are missing the point. I want to figure out if I'm winning so that I can improve my existing strategy bit by bit. But if I'm just a losing player then I have to rethink what I'm doing completely. I don't know what the correct thing to do is.

If I'm to ignore all-in ev /100 and bb /100, then how am I supposed to figure out my winrate? I need to estimate my winrate, it is important for me to figure out what to do next. Take it as a given that I'm trying to improve.

You've played 10k hands, there's no way to know if you're winning or not.
Just focus on the process, keep playing and keep studying.

If you're not sure, assume you're not a winning player until you know you are.

But how do I know when I am? What is the measure?

My opinion is ...... being down 2.6 buy-ins in 10K hands is nothing. I've been down that much in just 20 hands.

If you were really bad, I would think that you would be down substantially more than 2.6 buy-ins

I understand yes and I agree. But I also look at the orange line and if I cashed out I'd be up 10 buyins. Is that evidence that I'm winning, or is it meaningless?

But really what I'm trying to figure out is whether the strategy I'm using is good enough to win over the long run or not. Are there clues in my results that help me understand whether I'm a winning or a losing player? Do I really have to play 100K hands to find out the answer? Is 100K hands enough?

Should I keep playing 10NL or am I going to lose my bankroll? I got 12 buyins left.

I'm just speculating here, but that's what you seem to be looking for.

I think you're most likely around a break-even player, maybe slightly winning, maybe slightly losing.
If you keep playing seriously, even with very little studying, you'll probably become a winning player, since people learn from experience.
If you study, you're going to become a winning player faster, or win more faster.

What games are you playing?
The softer the games are the less of a sample you need. Well actually the higher your win rate is the smaller a sample you need.

If you were winning at 25bb/100 over 10k hands, it would be extremely likely you're a winning player.

You're losing -5bb/100 though, which could mean anything since it's only 10k hands.
It most likely means you're a slightly losing player.

Forget about all-in ev. All it does is make you think you deserve to win more than you did.

Just work on your game and you'll become a winning player.
You'll know when you get there.


Are you playing 6 max or full ring? Zoom or regular tables?
The differences in variance are huge.
12 BI's for 6-max zoom is a joke and even the best players would have a decent chance of going broke.
12 BI's for very soft full ring regular tables is much less risky, but still not great.

nice reply, yeodan.

Yeah I understand it's speculation, I was just trying to figure out whether it's 90% likely im a winning player or 10%, I realise it's difficult to see over 10K hands but was hoping a more trained eye could take an educated guess, and thanks, I appreciate it. I was playing 10NL 6max zoom on pokerstars for that sample. I'm currently 2 tabling 5NL zoom now just to be on the safe side. I only have one screen so i don't like having lots of regular tables open.

I see what you are saying about the winrates, it makes sense. To be specific, what I was wondering is whether the all-in EV line is the "expected winrate for my strategy", i.e. would the green line tend towards the orange line over a larger sample size - and is my green line currently deviated to the bottom of my "winrate range". Or, is the orange line completely misleading and I could well be at the top of my "winrate range" and am still not doing well. So I was trying to figure out whether the orange line could tell me anything about my "winrate range", but from what you say it sounds like it doesn't tell me anything, so I'll turn it off if that's the case.

I figure this would be known if we looked at a bunch of players' databases who do have a large hand sample, I should take a look and see if that holds true now I thought of it.

I doubt all-in ev means what you think it means, but I'm honestly not sure what it means, so can't say.

Zoom variance is much, much (really much) higher than reg table variance.
I've known a pro player who went on a 200k hand break-even swing.

Yeah I have a difficult time figuring it out myself because I do get that the all-in line doesn't tell you how you are running out of all-ins. 200K downswing must suck!

I thought about it some more and most of my wins are coming from my red line which seems pretty stable, and when I look at google image graphs, the red lines look pretty stable there too, possibly meaning the red line is not affected so much by variance. Then that means how I'm running at showdown and all-ins is the deciding factor to my true winrate. Now I'm running about 10 buyins below my all-in EV, which means for the green line to reflect my true winrate, my non-allin showdown winnings need to be running about 10 buyins above EV to cancel the all-in EV out. So maybe that means there is a good chance that I'm a winning player since it's unlikely my non-allin showdown winnings are running that good? It definitely doesn't feel like it's been running that good.

Hey man, I hope you're a winning player.
If you're not yet one, just keep working on your game and you will become one!
Don't worry about it, just keep working!

Don't worry about how good you are or who's better than you, just keep focusing on getting better every single day and you will eventually get where you want to be!

I'll keep at mate and thanks for your help. Poker is just so complicated that it's hard to know how you are doing. I still feel very much blind in the forest.

You'll be fine once you get in a million or so hands

It's more accurate than just looking at your overall up/down as it accounts for one source of variance, but it is not perfect. You'll need to play more hands to get a more accurate idea of what your real win rate might look like.

For a sample size of 50k hands and a std dev of 100bb/100, your "real" win rate is your observed win rate +/- nearly 9bb. This means to get a really good idea of what your true win rate is you have to play an absurd amount of hands. Just move up/down mostly based off your own feel and whether or not your bankroll supports it.

That gives me something to work with, thanks. I did move down just to be on the safe side, I'll grind 5NL for a week and see how I feel after that.

AIEV shows what your winnings would be if you ran exactly at EV during your allin hands, calculated knowing your and your opponent's cards. It is a more accurate representation of your true win-rate than results are but it only eliminates one aspect of luck when there are many others. It's very possible to be up AIEV but still be running poorly, or vice versa.

Even if your true win-rate were 10BB/100, which is generally considered solid for micros, this would be a very standard downswing, so you need to learn how to get used to it. The lower your win-rate the more severe your downswings/breakeven stretches will tend to be.

I've done a tiny bit of research and concur with browni above.
It does seem all-in adj bb / 100 is exactly what it seems to mean.
It just takes your normal wins/losses (where you're not all-in) but then uses your equity in an all-in spot instead of the actual result.

So I have to adjust my statement from earlier.
With a 8.6bb/100 (all-in adj) it is somewhat likely you are a winning player. Though not certain.

If we looked at 1000 players with your stats, the majority of them would be winning players.

So keep up the good job and try to win some more all-ins

This seems pretty hard to know, actually. It depends on the distribution of true winrates among the players you're selecting from. If you found a roulette player up 8.6 units/100 spins after 9k spins what's the probability that they're a +EV roulette player? Still 0 for a fair wheel. In poker most players are losing so you need significant evidence to show that someone is winning before you can separate a long term winner from a fish on a heater statistically. Is 9k hands enough evidence in this case? I have no idea.

Well I didn't say OP is a winning player, I just stated that based on the statistics it's likely, which should be true?

Out of 1000 players with these exact stats, maybe 700-800 will be winning long-term players. So 200-300 would be losing players (just making up these numbers)
That means a 70-80% chance (again made up) that OP is a winning player.
It's definitely somewhere above 50% probably around 70-80% imo.

All I'm saying is it's likely.

Are you just stating your opinion based off of your experience and intuition or asserting a fact? I agree with your opinion, at least to the point that I think OP is probably a > 0EV player. All I'm saying is that there's not enough information to strongly support that opinion statistically.

I'll refer to the roulette example because it works for me. If OP were a roulette player with the same statistics, how likely is it that he's +EV?

Wouldn't it be a fact that if:
You take 1000 random poker players who are all winning at 8.6ev/bb over 10.000 hands.
More than 50% of them would be winning players over large samples if the conditions remained the same?

All I'm saying is that if the above is a fact, which I think it is, then it's likely (as in more than 50% chance) that OP is a winning player.

Not sure what roulette has to do with this.