Quote:
Originally Posted by bip!
And I think the biggest misconception - focusing on variance as a property of all ins. Variance is a part of every hand (including folding UTG for $0). Preflop and Flops probably influence variance more than all in results.
This is key. There are so many types of variance. One of the biggest, that most overlook, is flop texture. Your opponent limps AJ, you iso AT on the button, he calls and you go heads up to the flop. He has you crushed, but you're going to win this pot the majority of the time if he plays fit or fold post-flop. However, the flop texture is the variance. If this scenario plays out five times in a session, and he keeps flopping a pair, you're "running bad," even though he had the better hand each time.
On the flip side, if every flop comes down K53r and he check-folds, you ran good but most players just think, "Oh man, I'm killing it today!" Nope, you're catching good c-bet boards!
Quote:
Originally Posted by meale
Err my sims were using a sd of 70 which is perfectly reasonable for 9-10 handed live FR if you play a tighter strat.
I'm very lag, but this seems unreasonably low regardless. Mine is 95bb/hr. This does include a handful of PLO sessions, so it's a little lower in NLHE. Probably 80-85bb/hr.
Quote:
Originally Posted by bip!
The reason this trips up tons of players is that:
- minimizing risk-of-ruin for short term (say anything from 1 to 100 hours) can actually be to play a weak and losing strategy.. however, if the strat is losing, it will end with ruin 100% of the time in the long run.
- minimizing risk-of-ruin for long term (hundreds+ hours) is to maximize win rate.. but this strat might actually have higher risk of ruin on a small initial bankroll.
NOTE: this doesn't mean "play low variance until I have a roll" - because you won't ever get that roll. Poker start-up involves risk. Don't try to dodge the risk. (unless recreation is your goal - then get a job and lose slow)
It takes a long time for winning results to clearly diverge from ~break-even results. This is thanks to how big an influence variance has.
A big edge is the only thing that will keep you in the game for thousands of hours.. especially if you withdraw from your roll.
It is the great long con of LLSNL - the slow noisy bleed captures so many people. It is why small stakes buy in cap games are so healthy and will continue to be healthy. So many people trying to extend their time and so few trying to win.
Lots of wisdom in that post, but I'll disagree with one thing a little bit. Depending on the type of game, your bankroll, your winrate, etc, there can be lower variance styles of play that still have a great winrate.
Consider a couple examples.
Player A doesn't pass up any edges. If he's risking 100bb on a river bluff that is +EV by 1bb, he'll do it. He's not afraid to play huge pots with moderate hands and draws, as long as it's slightly +EV. He wins 10bb/hr.
Player B passes up those edges and plays a TAG style, he doesn't make huge bluffs or hero calls, and he tends to play his draws and moderate hands more conservatively. He doesn't get paid off as huge when he makes monsters, but he tries to avoid huge swings. He wins 7 bb/hr.
I believe, although I'll defer if you have a good case here, but I believe it's possible player B has a lower risk of ruin on a shorter bankroll. Now, of course, if you're taking a player who has a high end of 5-7bb/hr, and they're trying to further reduce variance, I'd agree with you that they will have a hard time growing their bankroll before the creep of rake, tips and negative variance take them out of the game.
Also, the type of game matters. Like, let's say both Player A and Player B have a 200bb shot taking roll for the next stake up. There's a wild, crazy game with a huge action player who is straddling every hand and getting it in light. Player A may have a higher winrate in that game, but Player B has a lower ROR on his shot-taking. At least, that's my opinion based on a small sample of personal experience and doing some serious thinking on the topic and a few calculations on ROR calculators.
Quote:
Originally Posted by browni3141
Standard deviation scales by the square root of time, not linearly. If your SD is 100BB/hour and you play 25 hands per hour, then when you convert to BB/100 that is four times as long, so multiply by sqrt(4) to get 200BB/100.
For SD:
BB/h*sqrt(100/(hands/h)) = BB/100
For the record, my live SD is about 110 BB/h, which equates to around 200 BB/100 if getting 30 hands/h.
Thanks, it's been a long time since AP Stats.