Hi all,

I am wondering if I am just experiencing negative variance or having a serious leak in my game. I just moved from live to online poker and it is going terrible. Please help me and look at the two stats
a) Is this just variance? How good/bad is the winrate?


b) looking at bb/100 per position: where should I dig in? Are the blind win rates ok? This is the same sample as above, a bit over 100k hands. What would be good win rates per position (Blind, EP, MP, LP)?


Thank you very much. Any help is greatly appreciated.

Peter

I think that your stats show that you are too careful in striving for the min cash. 24.67% is way higher than it should be for a losing player.

With this sample size it is possible that you have run especially poorly after the bubble bursts, but the simple explanation is probably correct.

I guess it is also possible you are playing very large tournaments and you just haven't had the one big score to get you in the black.

From personal experience, I tend to tighten in the stage right before the bubble. This is the time you need to accumulate so that you can be a bubble abuser, and not be abused. I'd look there first.

This is a very low sample (some players play this number of games in a month, running like dog ass for a month obv hurts a lot less than for a year mentally). ITM looks fine if it's steep prize distributions. If these are big fields ala Pokerstars -200 avgBI nothing out of ordinary especially if you were firing some higher buyins (50s-215s) where you might have been running poor.
You can have a positive adjusted bb/100 from SB though, it's possible but I was making ~30% ROI with similar bb/100s as yours. You definitely need to work on your late game (pre-FT and FT) because that's where the money being won. Look into your FT finish distribution too.

Losing over a 1000 tournaments is nothing. With that BB/100 you could find yourself losing over 5000 tournaments.

I don't know if your BB/100 is good or not, since I don't know what good regs are doing. I think if it's an American site it probably should be higher, if it's PS, well... the games are pretty competitive, especially at those stakes.

Try to see what your BB/100 is according to stack size. Divide by deep play, resteal territory and push folding stack sizes. The deeper you play, the higher your BB/100 should be.

Thank you very much for your replies, 3for3poker, Hot*ShoT and leviathan74. I was feeling a bit lost&alone and your feedback is much appreciated. Putting things into perspective helps, especially when you are on a loosing streak. First note to my self is, that sample size is still small. Especially since at least half of the tournaments are huge PS fields. So nothing to worry extremely about.

Nevertheless I will do the following (not necessarily in this order):
a) Carefully look at my prebubble play.
@3for3poker: I really do not think that I adjust my play in a way to survive into the money. Actually I do not adjust much at all. If I have a larger stack at the bubble I am a bit more aggressive with my open ranges. Except that I play pretty much my standard game with standard ranges. At least this is what I think. I will see, what the analysis reveals, although it seems difficult to pull exactly this data set out of poker tracker

b) Look at my FT and Pre-FT game. Same challenge as above. Doesn't seem straightforward to extract the data.

c) Look at the data by stack size. This one seems the most straight forward with Pokertracker =>will do this first


Thank you again!
Peter

What's the average field size? 1k tournaments can be meaningful depending on field size/structures. Frankly the guys grinding with huge swings to make say 100k is not smart long term.

obligatory
http://www.nsdpoker.com/2011/01/mtt-pros/

also, i don't think viewing bb/100 is really helpful for mtts, so many of the times you're shipping it due to circumstances beyond your control as there are many situations where you should make a bad decision for the right reasons

when i was on the mtt grind i would look at the prize structure of the tournies i wanted to play and then using monte carlo would simulate a month of grinding it - i'd do this to try to figure out not how often i needed to min cash to be profitable in the long run, but how often i needed to go on a very deep run - it'd help me better reinforce to myself that min cashing is a losing strategy and for right or wrong, it made me happier to ship it lighter in more unclear situations simply because right or wrong, i needed to chip up rather than slowly bleed

the only way to ever be profitable in this format has nothing to do with min cashing or chip ev and everything to do with the frequency at which you make very deep runs

Thank you so much for the link. That is an awesome read. I think the message is clear, for large field MTT players, especially if they dont play super huge volumes: Shut up, and take it .

But this brings me to my next question. Ok, obviously variance is huge. But I am commited to improving my game. So I need to measure my performance:So how on earth can I measure my performance? Your link illustrates quite nicely that wins/losses/ROI doesn't say much, unless you have a super huge sample. And even then it might be off

Isn't bb/100 at least a small step into the right direction? How else would you objectify poker performance?

It did what you said, but in light of the previous discussion about sample size, the validity is unfortunately questionable. However, the pattern you described is observable. It seems that I should have a look at my 12bb-25bb game. Anyone knows what I should aim here for?



Now it really gets interesting, if I divde the short stack play even further. Seems obvious that my 12bb-25bb game needs improvment



Thank you all for your feedback so far. I think I am onto something. This stack size is quite prevalent after the bubble bursts. I think it is difficult to have good results with a leak here. Very worth investigating. I am quite happy now. Finally something I can look into

good question, i wish i had an answer other than you just gotta have faith

honestly, never quite knowing how good you are or aren't is common at all levels of mtt play, this is why the top guys all cross book each other because there is so little clarity on who is better than whom

you could spam out some sngs, which have much less variance but require similar skills to get some kind of sample but then you'd be on the sng grind instead of the mtt grind

That is exactly what I was thinking.... Its just that I like the MTT fell so much more....

1. your tournament sample size is dogshit [900 tournaments... you might be playing tourneys with 500-2k runners...]
2.Winrate is small but winning
3. your late game/icm play might be punting [ev/bb/100 doesn't really matter on final table etc as much as folding in correct tight spots]

http://www.nsdpoker.com/2011/01/mtt-pros/ put this into your head too... twitch.tv/wot214??? i think had a 200k 1.5 year downswing ... imagine

Part of your problem is looking at BBs/100, which is pretty irrelevant for tournament play. I agree it is obvious that you are playing too much for the mincash, and that is a big problem

I had +8bb/100 over 300k hands and lost money
Doesn't mean **** rly
Forget bb/100 if anything worrying about that is going to make it worse
In big field tourneys you can play like total crap 90% of the time and still win or play good 90% of the time and still lose because its that 10% of hands where it really matters

From your stats tho I would say sb and btn are the areas you need to work on

Hi guys,

thank you for your additional comments. I know now that I should not fixate on BB/100, but nevertheless I think one should use it for at least some conclusions. One would be, that I have to look at my small stack play and specifically ICM.

I have however one additional question:

@Frogman; How do you come to that conclusion? Isn't the SB on par with BB, and BTN looks very positive?

I think btn should be in the 30s and sb should be less than -10 but I'm not a hud guy
Try playing without a hud for a while because its mostly bs for mtts where you barely see the same players, especially if you are worrying about stats

I think you need to forget about BB/100 as a stat that applies to tournament poker.

Look at it this way: in the early stages everyone is 100 bb deep and a lot of guys are playing LAG poker, thus creating pots that are 20 or 30 bb's large several times during every orbit. But in the mid stages, when the average stack might be only 50 bb's, those opportunities dry up, and by the late stages, when the average stacks are under 30 bb's, you will see a pot that large much less frequently and only when shorter stacks are all in.

In order to use BB/100 as a stat in tournament poker, you will have to create some sort of formula or algorithm to adjust your expected return to the average stack size, which seems like a lot of trouble to go through to create a stat that would have limited value regardless.

Your BB/100 is always going to be a reflection of your stack size. Say you have doubled your stack after 100 hands in a tournament. Your BB/100 is going to be the same regardless whether you got there by winning 12 little pots or picked it all up by winning 1 big all-in.

BB/100 may be a useful measuring tool when playing a cash game where BB's are static, but in tournament play this stat is only a distraction.

As for how to find your leak, I don't need to do a deep dive into your stats at all. Your have an acceptable cash rate of about 25%, yet you are a losing player. This is is a clear indicator that you need to work on improving your short stack play.

How you operate when the average stack is under 30 BB's: in tournament poker, that is where the rubber meets the road.

It's not an either/or. BB/100 is instructive, especially when you re using it to track your progress and not against other players.

Yes, in theory, someone could be winning a massive amount of BBs when deep while being break even when the stacks are shallow. That man is going to go deep, but isn't going to win much since he isn't winning when most money is in play. But that's rare though, it's not always this clear cut in practice.

You could also have two players. One winning 5BB/100 with shallow stacks, the other winning 3BB/100. If the second player is far better at ICM however, he could be winning far more money while the former burns $EV for cEV.

But if you can keep these things in perspective, you can use BB/100 as a benchmark and as a way to see which areas to improve on.

It's not icm that's the issue usually . It's that late game you need to be much more looking at stacks and changing gears, putting pressure on people etc. Early game you play like cashgame but late game there's so many spots you want to avoid or take, depending on stacks and dynamic. A great player can take a stack and get results from it whereas a mediocre player just gets average results like his stack doesn't matter he just carries on playing the same. Or he over adjusts and thinks having a big stack means he gets to play every hand or he tightens up so much because he's deep
It's a mixture of being fearless and knowing when to get out early from situations that have more downside than upside

I'm running at 12.68bb/100 and at 13.58/100 ev this year (only 50k hands) and probably about break-even. Late game skill (ICM) is obviously crucial, but so is late game run-good. Remember there's a TON of luck involved over small samples and even large samples, and I can't tell you how many times I've been sucked out on in big spots or lost multiple flips in row at crucial points of an mtt.

Just keep trying to improve and enjoy the process regardless of results. If you have an edge in your games then you should be able to see it clearly in practise (imo)

I don't understand why being able to adjust better to the late game -changes which by default are driven by ICM pressure- won't have an effect on your BB/100.

I can guarantee you that if you are making 12BB/100 after one million hands, you will make serious money. But I doubt that's sustainable at midstakes.But even less than that, it's very good I think.