Again, your standard deviation is a good indicator of how deep a downswing could be. Anyone can have a long losing streak, but if there are 2 8BB winners and one plays a high variance high standard deviation LAG type game and the other plays a lower StdDev more TAG game, the TAG will tend to have less devastating downswings (and also will tend to have less huge winning days and huge winning streaks).

Only you know what your playing style is. If you have 1500 hrs you should have a pretty good indication of what a bad losing streak looks like for you.

Figuring out bankroll requirements shouldn't be so complicated.

Don't separate "bankroll" and "living expenses." Money is money. Thinking about them as separate just confuses the matter.

1. Estimate your win-rate. Lower it a little bit to be conservative.
2. Calculate your standard deviation. You don't need a very large sample for this stat to be reasonably accurate.
3. Figure out how much you will be spending per month. Increase it a little bit because sometimes you need something unexpected or just want to spend a little. It's hard to set a $1000/month budget and follow through with it.
4. Decide how much you will play per month.
5. Decide on an acceptable RoR based on your tolerance for risk.

Now you have enough information to determine bankroll requirements.

6. Figure out your win-rate adjusted for expenses. For someone winning $15/h, spending $1050/month and playing 150h/month, this is 15-1050/150 = $8/h.
7. Use the bankroll formula or a calculator and plug in the numbers. Here is one calculator: http://www.reviewpokerrooms.com/poke...uirements.html

With an adjusted win-rate of $8/h, a SD of $200/h, and a 1% RoR, you need a bankroll of about $11,500.

This is what you are willing to lose. I would also recommend setting a stopping point before total busto so that you are not completely broke if your poker venture fails.

I wouldn't worry about emergency expenses too much. They are very rare and tough to account for. Poker variance will be the dominant factor in your RoR, not life variance. However a bigger bankroll can only help.

Be very careful when considering living expenses on a short roll. If it turns out you actually spend $1600/mo instead of $1050, you bankroll requirement about doubles with the other variables the same.

You can play more hours to minimize the effect of living expenses. Playing more than 150 hours gets hard for me, personally.

$15,000 is plenty for a single adult playing 1|2 living cheaply if they win at a good rate.

You def answered all my questions thanks so
Much will stop torturing you guys now

You should have enough expenses to pay three months rent, and lose for three months as well, maintaining a healthy poker bankroll the entire time.

If you use a thirty buyin rule and keep your life expenses separate you will be ok.

I really like this quote, well done.

I'll do some search. Thanks for the reference.

Good example is cash vs tourney.

+EV spots in cash can often be -EV spots in tourney.

I spent a while today working out my standard deviation, which is 59.70 bb/hour. Then I checked the 95% confidence interval per Happy Luck Box's post, which gives me 6.84.

My win rate to date is 6.63 bb/hour +/- 6.84 = somewhere between -0.21 and 13.47. Nice to know my initial guess about my expected win rate was pretty good but wow you need a lot of hours to get a decent confidence around it. Also would help to have a higher win rate as well.

I don't play tournies but I would have thought the opposite where busting a tourney has minimal effect on your BR but winning a massive pot can set you up to potentially make it much deeper in the tournament.

Can you give an example of your cash vs. tournament example?

The most obvious cash example for me is getting deep at 10/10 where booking the win is way more important than a small +EV spot while deep stacked.

In tournaments there is value in moving up the pay jumps, there is value to having a certain sized stack, and there is a value to your tournament life. If someone shoves all in on you in a cash game you simply consider the pot odds to determine whether to call in that spot. That's not necessarily the case in a tournament.

That's the whole point of the discussion: the EV alone of a cash game hand isn't enough info to make the decision - it depends on the risk/reward in relation to your BR because nearly everyone is under rolled.

I would highly advise against playing under-rolled for the most part.

roll advised to play a $5/$10/$25 game with 4 competent players, 1-2 whales, and the rest rec/fish?

depends on buyin cap. 25k if it's a one-off 1k cap. if you're buying in 2k, then like 30-35k. If you're buying in for 2-2.5k regularly, 50k roll

These recommendations are probably way too low for 5/10/25 NL.

I am guessing that you would need a minimum of 75k to be rolled for 5/10/25 NL.

Yeah my guess is such a game often plays 5k+ deep, so I wouldn't play that game REGULARLY with less than 100k

That was also my point.

In a tourney, it makes sense to avoid thin spots in which your tournament life is on the line. Same can be said about someone with really small BR.

So the idea of avoiding variance actually has a very wide range depending on each person's financial circumstance, more so than the EV of an individual hand.

Thus this quote is amazing in a very subtle way:

Agreed.

A giant part of poker (maybe more-so live poker because you can gauge reactions) is taking people out of their comfort zone.

Avoiding thin spots in tournaments has nothing to do with variance. Some spots might be chip EV+ but not $EV+. That's not a variance issue.

So how does avoiding thin spots in cash game have anything to do with variance?

If you don't think it does, then alright.

If you do think it does, can't you not see the parallel?

And I don't play tourney, so please enlighten me the difference between chip EV+ and $EV+.

Regarding WR and SD helping out predicting what type of downswing we could possibly go on...

... isn't this making a major assumption that the game conditions we've collected all of our data in up to this point remain the same?

And since we know that game conditions *probably* won't remain the same, aren't these stats *almost* rendered kinda useless with regards to predicting the future?

GknowsexactlywhatI'vedonetothispoint;haszeroconfid encetheymeanmuchofanythingmovingforwardG

We have no idea what your winrate is. If you aren't a winning player, then you need an infinite roll.

When we play cash, chip EV and $EV is one and the same. So our decisions are easy. The action that makes me the most chips is the one that makes me the more money and so that's what I do.

In tournaments this isn't the case. The money we make depend on our finishing place. This means that in many spots in a tournament, we are faced decisions where chip EV isn't the only thing that matters because our $EV depends on what other players not in the hand will do after the hand.

To give an example. Let's say we are in a tournament paying 15 places with 16 runners left. You are second in chips in the BB. There's a crippled stort-stack with half a BB left UTG. The chip leader is in the SB and he shoves at you. If it's a strictly chip EV+, you calculate that he may be shoving anywhere from 50 to 100% of his hands, so you call him with 20 to 40% of your hands. However, even though you may win more chips in doing so, you don't win more money because chances are that if you fold, the short-stack will get knocked soon enough and you will get ITM. Since the short stack getting knocked out in one of the next few hands happens significantly more times than the times you call and lose the SB shove, calling is chip EV+ but not $ EV+.

But that's not an issue of variance. It's isn't an issue of being able to withstand the ups and downs of the variance train. It's an EV calculation pure and simple.

That's what ICM -the Independent Chip Model- is all about. It's pretty complicated and it's probably impossible to know at each spot the difference between chip EV and $EV, but it's foolish not to take it into account at all.

Most of the time. There are definitely spots in a cash game where keeping a whale happy and spewing are +EV that might not be in a vacuum. Not as common as in tournaments where ICM is almost always a concern of course.

That's the point I was making why "avoiding variance" is as much about evaluating whether a hand is +EV as whether one's financial situation is able to bear the burden of the risk.

We're talking about the same thing, perhaps you missed the context.

Yes it is and yes they are. That's why I don't mind that I can't remember what happened more than a few weeks ago