Right, so as the number of bets approaches infinity the probability of going broke approaches 100%. This is easily seen when considering the strategy of betting 100% of the BR on each bet. Even if you win 90% of the time, you are sure to go broke quickly despite having a very high expectation.

Build a big bankroll. Treat it as a funny money. Have enough to withstand challenges of fate. Do not spend the winnings. The variance in NL cash games is extremely high, due solely to big pots. Tight play does not provide a solution to this problem, although it somewhat tones it down. The obvious answer is a big enough bankroll. Do not adopt any strategies designed to reduce variance.

aaronbrown is sort of wrong. you're not betting 10 percent of your roll, you're betting a buy in, a static amount, or their buy in, a variable amount, or your bet, which may or may not be static, depending on how you play. unless you literally buy in for a set percent every time and go all in within the first few hands, aaronbrown's point is not quite valid.
if you shortstack, winning once then losing once is often busting, but if you bigstack or medium stack, then it depends on the size of the opponents' stacks relative to each other and you that'll determine whether it's winning or losing, although long term, you'll probably run into enough big stacks on the second loss to make it a bad play long term.

Let's consider my argument of lost equity. Consider a 500 max buy in, you are sitting with 1k, and villain has infinity. You play back to back coin flips.

25% lose both - down $1500
25% lose first, win second - down $500
25% win first lose second - down $1000
25% win both - up $3000

So what we truly have is a situation where over time, you will retain your EV, even though 75% of the time! you will end up down in that moment.

This goes to the heart of my argument. I'd rather play the 50-50 EV spot in a situation with a 50-50 win rate, as opposed to 50-50 EV with only 25% win rate.

Plus, if you treat your buy-in as your bankroll, issues like volatility drag most definitely come into play.

It's easy to say, build a bankroll and not worry about variance. My bankroll is fine. All I am saying is that there must be a way to reduce your variance by a factor greater than reducing your EV by adjusting your range and picking your spots. Surely there is a mathematical way to back up this theory.

And I'll make another point.., the odds of being covered on back to back all ins is very unlikely... So your EV in that spot may very well be negative in practical application.

I don't believe in it but I see people do it all the time because they are nitty...

i said buy in instead of stack

I was thinking about this today.

Do we overestimate the long term? The old adage 'if you get it in good you will win in the long run'...but what exactly is the 'long run'? Is it as long as our bank roll lasts? Is it a chronological time period?

The 'long run' is also always challenged by new variables,m new villains, old foes who wisen up, stacks that ****** the natural flow of our bank roll...is it really possible for anyone to take advantage of the 'long run'?

The long run is your poker playing lifetime?

Also if you regularly give up EV that long term may not be so long term?

But then again it hurts to approximately flip for stacks over and over so IDK I am not sure...

Poker playing lifetime?

As long as you keep playing poker. For Doyle decades and for some other winners it will be less time.

Lets say you play correctly for five years straight, consistently get it in good...but go broke. Could it be said that your poker lifetime met its natural end?

IMO if you play good for 5 years with an adequate bankroll using good bankroll management and you are losing you are a losing player. It's not always about getting it in good either. There is always picking up the loose pots. You cannot just wait to make the nuts on people and "get it in good." If you are playing correctly using all of the things above and you go broke you could say that. Then you could also get a job and get back in action with increased skill recover you losses and some and play until you look like Doyle. In that case your poker playing lifetime would be when you quit playing poker.

yes. 100%

I wouldn't say I totally disagree with this either. When you are trying to actualize every edge it is very easy to go over the edge as well and that opposed to the other way around could be disastrous.

Ok yah I agree, I can see some merit to this point of view.

If everyone at the table is a $500 stack, and the fish on your immediate right is $1k stack, and you have a $1k stack then I can see times where maybe it's not worth taking that 2% edge with some other reg that has a $500 stack.

This is a hard situation to quantify. As the other poster said, if you win you'll be $1500 stack and although the fish currently only has $1k he could get up to $1500 as well.

and that's only in the bogus scenario he shoves all in and shows his hand face up. chasing 2% all the time gives you little room for error.

Well we all know that we don't know with certainty where we stand... but even so, we can still apply the logic I am espousing to drawing hands and more clear cut situations or situations in which your reads are at a high performing rate.

If I'm in late position with JJ and someone ahead of me shoves pre-flop... then yeah, there's a significant spread of whether I'm a huge favorite or a huge dog. There it's about feel, and not as much about playing a slight edge.

I think this example is not a good way to look at your issue, it's a bit too far from reality to help much.
Cash tables don't often get capped at sub 100bbs levels so for this to happen you need for someone to be open shoving 200bbs at you. I do understand that you may want to preserve your stack to get a better shot at stacking a weak opponent later and you should but it's a hard line to judge.
Really if you did have $1000 on a $500 max buyin table the decision is whether you are good at deep stack poker rather than the difficulty in finding the cut-off point for a volatility drag effect.
When you are deep by the time the allin decision comes along usually there will be plenty of overlay in the pot and less in the stack, so a smaller proportion of your br in action and the more the overlay the closer this optimal b-r drag point will get to the 50%.
I think you would have to take every hand as a different case as there won't be a simple equation to locate the optimal point.
This point is going to vary with many factors:
(i) size of the wager at the decision point
(ii) size of your BR
(iii) ease at which a fish will hand their stack over (eg, can wait for a better spot)
(iv) chance the villain will double and leave.
(v) size of the rake
and a I guess a few more.
In the current poker game variance isn't bad really it's just the lack of preparation about it that causes problems and BRM does help.

I think that trying to extract the most you can out of the game protects you best from the worst swings and this is really what most people are trying to achieve who try to adjust to reduce variance.
Perhaps an interesting experiment to try is to use your db to find the bb/100 of yourself and a good few others in your db. I would think as a cash player you have quite a few players with 1000+ hands - see if there is much variation in the variance and if this has any pattern. Do good players seem to have higher or lower variance?
(the data is not completely unbiased but it may still be revealing in some way - do filter for the same table size though)
Some style of play may be less variance than others, maybe the so called small-ball approach is lower (?) but I suspect a lot of the variance is simply locked in by the type of game you play, full ring, HU etc.

As you can see I still think ignoring the variance of a poker situation and just trying for your best game is better.
I thought this from the earlier post by Aaron was very good and if you do adjust you are gifting lower variance to your opponents.

The example uses betting a constant fraction of bankroll each time for simplicity to make a point. The principle applies to all betting situations.

When you evaluate a long series of bets, say all the bets you will take in your lifetime, you have to consider both EV and volatility drag. Volatility itself is likely to average out if you're sensible, you are likely to end up somewhere near your median outcome.

Say you have $500 in front of you, $500 in the pot, and someone goes all-in on the river. You think you have a 35% chance of winning if you call, so your EV of calling is +$25.

If the $500 is all the money you will ever have to play poker, and you plan to keep playing poker for a long period (assuming you don't bust) and think you're good enough to find plenty of other +EV opportunities, you should not call this bet. It doesn't matter how you feel about risk, any strategy that accepts this bet will almost certainly do worse than a strategy that accounts for volatility drag.

In this extreme case, when you lose, you lose not only $500, but all the profit you could have made from playing poker forever.

But say you have another $500, so your total poker playing bankroll is $1,000. This bet is still too risky, the volatility drag it creates is more than the EV.

You need a total bankroll of $9,926 to make this bet a good one. You don't necessarily need to have that in cash today, but you have to be willing to lose that much playing poker without giving up the game.

If you had more than that, say $25,000, then you should take the bet to maximize your long-term outcome. You might still decline it because you dislike volatility and poker is just a hobby. But with less than $9,926 taking the bet is a long-term losing proposition.

Every bet you make in poker or other risk-taking activities is subject to the same force.

So some would argue that you shouldn't be playing at a table where any bet with positive odds yields a long term negative return based on your bankroll.

But I agree that's not a sensible approach either.

All of the conversation in this thread has just reinforced my understanding and belief that positive EV or pot/implied odds does not always equate to playing the hand.

Everything I read indicates that the only rational exception is when you are playing too high for your bankroll.

surprised no one has mentioned insurance yet

if you buy health insurance, etc. that's what you're doing!