Thanks for the comments DEKE. I think that you reinforced what I was trying to say very well. Which is, that losing years are rare but possible even for winning players. Secondly, that when players truly run into the bottom of the variance curve... they quit. Most people who continue to play poker for the long term appear to be those running in the top portions of the curve. Its just too miserable to be in the bottom.

I wasn't trying to say that 50K dowsing at 1/2 is going to happen. But it can at 2-5 sometime in a few million hands for sure. You have a very impressive session win %, mine is not nearly that high. Do you tend to play longer sessions? I find that when I am playing 2-3 hour sessions I only win 60% of the time. It goes up slightly for longer for longer sessions. My personal biggest drawdown was around 6k at 1/2 (definitely playing bad for a good portion). And now around 20k at 2-5 and 5-T, much of which was in a few large suckouts/cooler spots (AA against KK pre twice for 500bb and losing both).

Yes, this happens. Bad play gets excused as "variance"... I've done it too.

The bolded is very much NOT supported by statistical measurement.

Super sick long-term results! Congrats!

You don't multiply your SD/hr by 3 to get your SD/100.

It's something like SD/hr*square root of how many hours it takes to get 100 hands, IIRC. So if your are estimating 30 hands per hour, that's 3.333 hours. Square root is 1.826. So if I am remembering this correctly (not guaranteed), your 100bb/hr SDEV is approx 183bb/100.

You're using incorrect inputs. Standard deviation doesn't scale linearly, but with the square root of time, so your standard deviation in BB/100 would be 400/sqrt(3) ~= 230. This obviously makes a huge difference and it's no wonder your beliefs about variance are so skewed. Probability of loss over 30k hands with corrected input is 1.2%. Still uncomfortably high for a lot of people, but much different than 9.7% chance suggested by the incorrect numbers. You also lost accuracy to rounding. Your calculations imply you're getting 100/3 hands per hour, which would be 33333 hands in 1k hours.

I don't know why we're converting to BB/100 anyway. All your stats are in hours, right? Use hours and add two zeros to the last field.



I'd also note that your implied standard deviation of 133BB/h is very high. Someone with a 100BB/h standard deviation and otherwise the same stats has only a .08% chance of losing over 1k hours.

edit: p much exactly what Browni said lmao

Anyway, I'm using a winrate of 12.99bb/hr and a standard deviation of 90.89bb/hr. I did not do any unit conversions before entering my inputs and just translated the results (bb/100 becomes bb/hr; 10,000 hands becomes 100 hours; etc.) as both of these variables scale with time which is what the calculator is simulating.

Also, I would be sure I knew what I was doing with the simulator before claiming someone else is talking BS.

nice graph, but i can't possibly imagine winning a quarter million dollars in profit and remaining at 1/3 and 2/5

Thanks - when I first started I ran pretty bad so it was pretty bleak. I was making very little profit - and even messaged a guy on here that I was struggling - lost with sets and all sorts of hands, but than it just turned around - and than I had about 3 years of run good.

Yeah, I've had similar downswings. But I know several people with over 10k live hours that haven't come close. It's possible to run that good.

Well the 2/5 was $1000 cap and for me I didn't like dealing with the variance of losing more than $3000 in a session, so I never really moved up. I also always invested the money in the markets, which turned out to be a pretty good move so I never really carried a bankroll over 30k. Also I never thought I was all that great because I struggled beating 2/5NL on poker stars, although I hate online poker. It's just so boring. Also don't forget about bills and taxes. It was my only source of income - and making 80k+ a year isn't that much. Health insurance, investments, car, rent/mortgage, food. I'm glad I invested a lot of the money because its something to fall back on, but if this economy goes to **** meh - who knows.

I got into day trading which has been really tough - and lost some money but no big deal. I didn't take it as serious and the stock market is the ultimate casino. They give you access to a lot of money all at once so if you tilt it can be catastrophic. I kept my day trading bankroll in a separate account and my buy and hold in another one. If I could do it all over again, I would of just bought O stock. O realty. They pay .25 a share every month and are extremely healthy company. I buy 20 shares a month.


This march I took my whole poker bankroll and invested it because the market was so dirt cheap - it just seemed like a better risk/reward than to play poker with it. I've already doubled it but have not taken profits yet. I personally think there is more room for this market to run.


I'm starting over again with $2500 and I built that up to $5700 playing 1/3 in a pretty short span of about 50 hours. Running clean but also the play is just so bad.

Tempted to just sell some equities and move up to higher stakes, but no rush. I am inclined to just get it to $15000 and move to 2/5. It probably will only take 400-500 hours of play which is nothing.

Yeah and no offense DreamCrusher but some of your hand history comments aren't exactly world beater. For example, you said Rampage's 54s punt vs. Garrett was bad. It actually wasn't. And was a pretty good line overall. And this has been confirmed by much better players than me. His mistake was actually adding on and playing deep with Garrett overall.

It is possible to run amazing for a lifetime even. Phil Hellmuth is NOT good at poker. I mean I don't care what anyone says. There are just multiple examples of insane plays, tilt, losing in cash games, 3betting Q4o. I wouldn't stake him for a 2/5 live game and would bet a substantial amount of money he couldn't beat any 5/10 game in the world live. I also have numerous long-time professional friends that have played 10k+ hours and haven't had large downswings and then I have some that have. You're just wrong.

oh i admire that discipline dj, i'm about to dive back into poker and personally have no shame in my plan to start at 1-2 and regardless of results log at least a few hundred hours at 1-2 before moving up to familiar levels of the past - i don't just want to shake the rust off but just get everythign simple and automatic and work out and discover any frresh leaks of mine at stakes where losses don't hurt so much

but man... i don't possibly think i could have the results you've had and not at least take a peek at the 5/10 tables from time to time if not higher

always loved your posts in bfi as well, rip that forum

This is an extremely ambiguous statement but I'll try to make sense of it. You have had friends who play live low stakes professionally that have a breakeven stretch that lasted a year? I guess that would explain why they haven't moved up in level.

Do all your friends play identically? Same skill level? Same stack depths? Same player pool? Same level of risk tolerance? Same everything? If not, then these big downswings could possibly be the results of something other than variance.

I joined some discords so don't post much in there any more. I played some 5/T and the games were incredible at times. Other times not great. Probably a big mistake to not play it as much as possible especially when I was younger. Oh well.

Jonathan probably studied this hand very closely and ran it through a solver so I'm not going to argue against Jonathan's point that Rampage played the turn properly and played the flop and river poorly. I think we can now safely assume that Rampage is a good player, and I am a bad player. However, I would argue that it is not solely variance that results in me winning year after year despite being a bad player, when a good player like Rampage could potentially go on a 1 or 2 year breakeven stretch at 1/2. The way that hand was played, whether played 100% perfect or not, certainly illustrates how that could be the case.

Yeah, I knew that it didn't scale linearly, but was guestimating at std dev which I know will have a very high impact on my output. Here is what I'm using as inputs:

Over the last 2k hours of live play (using this sample as it most similarly composes the stakes that I will play moving forward) I have a win rate of $65/hr, and a std dev. of $707/hr. So I entered those into the win rate and std dev fields. I entered a 100000 hands time frame and ran the sim. Yep, that makes it smooth significantly moving forward. Down to a 1% chance or so of a losing year. I grant that my std dev is somewhat high, but still only around 11x the win rate. I would have thought that well within the range of reasonable for a winning LAG.

Thanks for the help with the simulator.

I do think the general conversation missed the point. My main point is that variance is very high in this game. FWIW, I notice that with the 'correct' inputs there is still a $150K draw down in the long-term results graph. As well as 15 or more $50k draw downs over 10M hands (equivalent to 50 years in the adjusted time framework ; 2k hours per year equivalent to 200,000 hands per year in the simulator since we are using 100 hands as a stand in for hours).

The point I was suggesting for discussion was that good players still stand a chance of losing a significantly large amount over a short sample (larger $$'s than apparently thought possible in this forum). My goal is to learn to better prepare my bankroll for the potential hits.

Or I suppose we can just believe that only losing players ever lose money. That variance will never catch up with you.

1,000 hr update
I recently cracked 1,000 hrs live since I last moved to a new location. It’s taken me 6.5 years and 349 individual sessions (~150 hrs / yr and 3.0 hrs / session on avg). I’ve played a mix of $3/6 limit, $1/2, $1/3, $2/3 & $2/5 NL, $2/5 & $5/10 PLO, $5/10 Big O and $3/6, $4/8 & $5/10 FO8. Obviously mixing all of these games together and spreading them over 7 years makes essentially all data analysis worthless.

But after all that I’m in at a whopping 8.40 bb/hr on average. I feel like I’m top 10 or 15% in the rooms I play.

Only reporting data so you can now easily support your decision to ignore my strat input.

It's getting a little chippy and a little off-topic in here. I moved the Garret/Rampage derail (except for the few that were directly related to the variance discussion) into that hand's thread.

Also, I remind everyone to actually read the thread rules that you click on to enter this thread. Please post constructively and respectfully.

I don't keep records in such a way that I can easily give you a calculated stat on the frequency of short session loses vs long session losses, but my experience is that longer time at the table tends to flatten the variance both due to the roll of the cards and my patience to wait for better cards.


One thing this forum has made me be more aware of is that I do get in a mindset that I need to book a win, which is a stupid way of measuring success. So I took unjustified risks towards the end of a losing session and made my overall results worse. Short sessions obviously gives me more opportunity to make that mistake. Awareness made it lots easier to fight that urge, even though it is still a part of me.

Awesome results DJ! I'm 51 hours shy of you in my 1/3 NL game but *way* shy in terms of $$$.

Any chance of posting accompanying giraffe?

Ggogogo,imoG

The fact that you said that most of your losses are a result of playing terribly, because they play so badly that you constantly have them drawing dead, is proof that you literally do not understand what it means to run bad and clearly never have.

We don't live in a perfect world where we always flop a full house against a flush draw. If you are pushing every edge possible then you are naturally going to get into situations where you are flipping or don't always have an equity advantage (coolers happen, more than people like to admit). You are going to run big bluffs sometimes for stacks, and will get called, even though it was correct to play that way.

Have you ever lost 90% of your flips over a 6 month period? Your AK always bricks vs QQ, but theirs always hits, etc... And don't try to pretend that you are so good at poker that you can avoid these spots and "wait for better ones." We are talking about pushing every edge and that means we aren't always going to have the nuts. Or lose most of your KK vs AK... Run badly at card distribution and run KK into AA a lot of times but almost never AA vs KK. Always be on the wrong end of set over set... Have people constantly flop trips vs your overpairs in single raised pots in spots where normally you should be able to value bet. Brick 90% of your big draws while having their gutshots constantly hit vs you? And when you finally have the nuts one has anything to call you? Run 6 figures below in all-in equity?

Now try having all of these happen at the same time. If you haven't experienced this, good for you, but it is still inaccurate to believe you only ever lose because of bad play.

If my last post comes across as harsh I apologize, but when someone posts stuff like "I'm pretty bad at poker but my opponents are so awful that I hardly ever lose," that is insulting to actual good players who have studied the game extensively, played tens of thousands of hours and know how variance actually works.

if you don't have an equity advantage, is it really a cooler?

I appreciate the post. Thanks.

I dont play like a nut peddling nit anymore for the past 3 years or so, and combined with playing more PLO ive had some rough stretches regarding variance. It can always be worse imo.

Sent fra min SM-G991B via Tapatalk