(NC) Why anything but Speeds?

[Wolfram]

I never tried the regulars. I run goot at the speeds, so why switch, eh...

[gumpzilla]

[ QUOTE ]
[ QUOTE ]
If your ROI at regulars is higher, the variance is indeed higher.

[/ QUOTE ]

I don't plan on making this a regular part of my posts, but this time it actually is fitting:

FYP

[/ QUOTE ]

Chill out. It is very clear that you guys are talking about completely different things when you're talking about "variance." One of you is using the word in a technical statistical sense. The other pretty clearly seems to be talking about experiencing lengthy downswings as a winning player.

The latter - which is a very standard colloquial use of "variance" around here - is going to be less likely if you have a higher ROI. It seems pretty clear that a 40% ROI player is less likely to go on a breakeven streak over 200 games than a 5% ROI player, even though his standard deviation is larger per game. If you accept that people generally have higher ROIs at regulars than speeds, then you should probably accept that they are less likely to have lengthy downswings. This is what people are talking about.

[KennyBanya]

[ QUOTE ]
If your ROI at regulars is higher, the variance is indeed lower.

[/ QUOTE ]
FeNeF,

The variance is exactly the same because the payouts are exactly the same. I believe you may be confusing more or larger downswings with variance.

KennyBanya

[kevkev60614]

[ QUOTE ]
Quote: "Not really, but it just makes them go at a slower pace. It is lower variance though."
Even though this has been posted 2234890235 times before,
the speeds have a variance that is the same as the regulars.
Please stop posting otherwise.
Your earn is lower and they feel like a crapshoot when there are 8 people left with blinds at 200/400, so you feel like the swings are greater, but the variance (or its square root, the standard deviation) are the same.
Also, all pros should play the speeds, the regulars are for fish.

[/ QUOTE ]
1) Technically, dollar variance will be different because the vig is different.

2) I think what a lot of people fail in explaining is that the same standard deviation applied to a lower population mean leads to a higher probability of a negative sample mean. In other words, having a lower true ROI at the speeds than the regulars given the same variance for each will net a higher likelihood of a short-term negative ROI at the speeds.

3) wpr101 was absolutely wrong about the difference in variance. And it's clear that you know what you're talking about. But don't be a cock. There's a way to disagree without being rude. This sort of post would be totally inexcusable even if a pooh-bah made it.

[ QUOTE ]
[ QUOTE ]
I play only regulars for two (main) reasons.
1) ROI is currently more important to me than $/hr.
2) I feel that I learn more about poker playing the regulars than I do the speeds.

[/ QUOTE ]

Can you explain why?

[/ QUOTE ]
I recently moved up with only 25 buy-ins, and I don't want to move down. My br hasn't affected my game, but it has affected my game choice. Maybe this is -EV and I should be playing a lower, speed game. But in addition to learning more from the regulars, I enjoy playing them more, and I play very few a week and want to win in the short-run, too.

[Dr_Jeckyl_00]

I was playing well ($33, single table) before Party changed its structure, bank roll was growing, life was good. Then they changed, and I played the speeds and tried to learn to 2 table for the 1st time. I almost lost my entire bankroll. Finally I switched to regulars, dropped to $11's, have learned to 3 table, bankroll is back to 75% of what it was before the change. I could not beat the speeds, I don't know why. I try them occassionally. Getting the highest ROI possible, with the shortest downswing and continuing to learn are what is important to me. If I could play speeds successfully, I would, but I can't, yet, so I don't.

[billxo1b]

how many tables do you play concurrently if you play speeds? 6 tables seem maximum for me.

[craigthedeac]

I can do 6 pretty comfortably, working on adding a couple.

[KennyBanya]

[ QUOTE ]
how many tables do you play concurrently if you play speeds? 6 tables seem maximum for me.

[/ QUOTE ]

I think 6 is my max for speeds, at least with one screen.
I have felt rushed and frazzled the few time I have tried 7 tables. For some reason that one extra table really makes a difference.

For me personally to play 7 or 8 speeds, I would either need an additional screen or at least learn to use some of those "auto hot key" features of MTH. Otherwise I am really rushed to act on those tiny tables.

KennyBanya

[1C5]

I tried 9 speed son my laptop, pretty tough.

Got my 30" Dell monitor and 8 or 9 tables is not doable as they are all on one monitor and close to full size.

[MadScientist]

Quote:" 1) Technically, dollar variance will be different because the vig is different.

2) I think what a lot of people fail in explaining is that the same standard deviation applied to a lower population mean leads to a higher probability of a negative sample mean. In other words, having a lower true ROI at the speeds than the regulars given the same variance for each will net a higher likelihood of a short-term negative ROI at the speeds.

3) wpr101 was absolutely wrong about the difference in variance. And it's clear that you know what you're talking about. But don't be a cock. There's a way to disagree without being rude. This sort of post would be totally inexcusable even if a pooh-bah made it."

The VIG is the same for the speed and the regulars since you pay the same fee, so I don't understand point 1.

As for point 3, I feel that if you don't take the time to understand the statistical definitions of the quantities of interest relevant to poker, you will not improve. By saying that the variance is greater at the speeds, (when it is virtually identical) it masks the true effect which is that your return, ROI, is lower. In fact, for many posters on this board, it is in fact negative.
The other effect, that I was trying to get at is that it feels more random because you get into these crapshoot situations where you feel like you have to push into a full table or go through the blinds with a short stack, so it feels like the variance should be greater.

I haven't seen anyone state what the mathematical definition of variance is, so here it is:
Var(X) = E(X^2)-E(X)^2
standard deviation~Var^1/2
E -> expected value
So if you take a spreadsheet with your 1,2,3, and ootm finish dist. And take X=5,3,2, and -1, respectively, if P(X) is the fraction of the time you finish X, then
E(X^2)=SUM P(X)*X^2 over all X and
E(X)=SUM P(X)*X
The finish distributions are about the same for regs and speeds (the slight difference doesn't change the results much), so the variance is the same.

Good luck at the tables.