godfjigs0dfgi5gqklgcvb0df9jdfopkgdg40 plo can be such a nightmare............afgq34it3tqjrogkafgafda9rejiq o34tkq3l5gg

^ maybe it's time to stop 50 tabling like a bot and work on defying gravity instead?
<3 deldar

It is (a lot) more likely that he overestimated his edge than that he lost 5000 bb with 30 bb/100 edge .

Not trolling... plug the numbers into the variance simulator. Using 15 BB/100 winrate and 100 BB/100 standard deviation, the max downswing was ~-3000 BB (-6000 bb) over a 30 million hand sample. If you have a 50 buyin downswing in a small sample, it's very unlikely that you have a 15 BB/100 edge.

St1ck is a scumbag, I wish Stars would ban him!

dave, you have some points that are correct, but if your overall point is 'luck doesn't matter in poker' you're possibly the most deluded poker player i know. i never got the guys that ignored the maths of this game. the maths is central to everything. poker is 95% maths with a bit of psychology and flair mixed in.

I assume deldar means that the "inevitability of going on a crippling downswing" is a myth, as opposed to flat out denying variance exists, which would be mathematically obscene. Variance is simply the spread of data points (the variation in your "actual" winrate over a given amount of hands) around a mean (your "true" winrate, which is a hypothetical number that can never actually be determined) - thus a person who is winning at 6bb/100 over 100k hands might very well be running worse than somebody losing at 6bb/100 - it all depends on the "true" winrate.

So I can only assume that deldar means that if you play "well enough" you are elevating your "true" winrate to the point that in all but the most severe of cases you won't actually lose money. Now this is actually true for the most part - running a sim over 100k hands 100 times with a 200bb/100 standard deviation and a "true" winrate of 20bb/100 gives the worst run as being around 50 buy ins (expectation is 200), i.e. the worst 1% run will still win 50 buy ins over 100k hands. Somewhere about 0.3% is about breakeven.

However that falls into the common fallacy where the breakeven point is seen as the tipping point of variance/running bad (ie. it doesn't matter how bad you are running if you aren't losing money you aren't *really* running bad). If you are actually a 20bb/100 winner and you only win 50 buy ins over 100k hands you are running very bad, end of story. If you play really really really well you might avoid crippling extended losing periods (largest downswing giving the 100 trials and 20bb/100 winrate is still 72 buy ins - but that is mitigated by the fact that you are winning at a very high rate most of the time) but this has absolutely nothing whatsoever to do with "variance doesn't exist". The breakeven point is a completely arbitrary point that we give credence because it represents a real world concept (winning money or losing money) even though it means absolutely nothing in mathematical variance.

I also find it hard to believe that anybody at nosebleed stakes PLO actually has a "true" winrate that is anywhere near high enough to avoid losing a ton of $$ if they run bad enough.

And as somebody else mentioned earlier one of the largest sources of variance for a high stakes player is which stakes he runs good at/bad at not the amount of bbs running below or above expectation (e.g. chicagojoey has figured out how to do it correctly).

Cliffs: Deldar's version of variance has nothing to do with the mathematical concept of variance (but is probably held to in some form or other by quite a few people in the poker world). This is probably due to the fact that when most people talk about "variance" they just mean "boo hoo i lost money" rather than any attempt to actually quantify their winrate (if you are a breakeven player then a 20 buy in downswing is far more standard than if you are a 10bb/100 winner), or examine how much of the "variance" was due to bad luck or bad play (even 20bb/100 winners play like breakeven players sometimes). In that sense, deldar actually made some good points about ignoring variance and concentrating on the things over which you have some control.

Offcourse 15BB/100 EV winrate is hugely overestimated. But those games where 25-50 ante games and he bought in with 200bb and never left tables if he sat deep. Your stdev assumpion is way way to low.

where's abstract reasoning, pattern recognition, and statistics analysis in this? i don't think poker is anywhere new 95% math, unless you play 20bb or something.

150bb hu is 20% math at most, if i had to guesstimate.

in fact, i would say that 'poker 95% math' is a more ridiculous, and untrue, statement than 'there is no variance in poker if you don't suck'.

As I see it, most of the decisions done in poker can be put on a mathematical footing - e.g. handreading is an application of bayesian inference (whether you do it consciously or not) - and looking at it that way, I think wazz is correct.

A great post.

when i say maths i mean logic, because all logic is mathematical to me. and when i say '95%' i pretty clearly mean 'the vast majority' rather than 95%. the shorter your stack, the more maths is involved, yes, but to say that maths is less important of a skill when you're even 1000bb deep would be a bit ridiculous. even if you're not consciously doing anything mathematical, whether that's because you've done the work beforehand or you just intuitively understand it, you're still doing maths. when you mix up your play in a spot, game theory might state that you should call 80% and raise 20%, and you mostly call but sometimes raise, you're doing maths. the issue should be about what skills it's more worthwhile (+ev) to develop - your mathematical skills or your pscyhological/finesse skills. and it clearly depends on who you are. i would benefit greatly from understanding and applying psychological and finesse skills, while you guys would definitely benefit a sh**-ton from taking a learnedfromtv approach to the maths of poker. if you don't understand that you're doing maths, you're underestimating the role of maths in poker.

Yeah, +1. That was very well put.

poker as a job is 20% math, poker as a game is 95% math

I agree with the general point that its fruitless looking at EV graphs and whining - being a good poker player is all about just constantly trying to improve and not tilt and look at things objectively.

I have to say though that I always think people are deluded if they think they are beating semi tough line-ups for like 30bbs/100. We are talking about a game where pretty bad players go on epic heaters and the high stakes games run for a while around them. Those heaters can last hundreds of thousands of hands. All the regs play a ton of spots really similarly in 6max and edges are pretty slim between them in my opinion. Even the fish I don't think are really losing at an absurd bb/100 rate.

People who are really world-class heads-up I imagine could get a fairly big edge on other regs - like jungleman/ Rui/ Isildur.

We are all human though and the sample of hands that we have played in our careers shapes the way we think about things. 8 months ago I would have had a different view at the top of my heater and seemingly could do no wrong. Maybe if I go on a heater in the next few months, I'll be back and making outlandish claims of huge edges to be had.

Deldar, how do you live with yourself on downswings with that kind of mentality?

Endless self-doubt and questioning plays seem to be dominant and lengthy downswings

hey deldar...

2+3 = ?

While obviously some of his blanket statements were ridiculous and many of the critical responses to him are correct, I appreciate the point he was making. I have been kind of a bitch in the last few months due to downswings, and I could benefit from trying to replicate his attitude some.

While PLO has numbed me some to big losses, it has also hurt my attitude some towards poker. I have much more of a gambler's attitude towards poker than I used when I played mostly NL. Between the games getting harder/more aggro and playing more PLO than NL, my standard deviation has to be at least 2x what it used to be. It is also very likely that my winrate is smaller too in any game I play. As result, my daily/weekly/monthly results are far more dependent on how I run than they used to be. I think if this is left unchecked it can have a very negative effect on how you approach the game.

Deldar's posts were a good reminder of that, and I plan to actively work on my attitude towards variance and how I approach my game.

Very well said.

Deldars got the right mindset imo, and is probably chuckling over all the lame variance talk that has been sparked. Do you think Phil Ivey plays a session and then runs his hands through an EV calculator? He just knows how to dominate people, plain and simple.

Good post by gavz as well indeed. And I agree that the mindset Deldar portrayed is a lot better than the mindset of someone wallowing in self pity and blaming everything on variance. But it's a bit silly to assume that in order to beat variance you have no option but to deny it.

By the time I was senior in high school I got so sick of math that I actually asked one of my teachers, "what is the point of the this, I'm not going need any of this complicated **** later in life, when the f-- am I ever going to need graph a linear function" ect.

My teacher responded by saying that math isn't really about numbers and equations as much as it is about the art of problem solving. You are given a certain amount of information, and you have to use this information to solve a problem. This is really the only way I think that you could say poker is all math, in that it is also the art of problem solving.

Poker is about making a correct decision based on the information that is presented to you. The better/ faster you are at making this decision, the better at poker you will be.

What is the definition of math? kinda relevant when making assumptions on how much math is involved in poker

+1

^ but there is quite a lot of numerical maths involved, not just in game strategy, but maximising profits long term w/ bankroll and game selection efficiency