Saturday, February 13, 2010

Can someone explain variance to me?

I don't understand variance at all can someone explain it?

I understand the basis of it but somethings don't make sense.

If i was tossing a coin 1000 times then after lots of flips the amount of heads and tails would even out. However if i was betting someone completely random amounts each time that we fliped the coins then how can that ever even out?

In poker today. AA first hand or a tourney, all in and i lose and the result is i'm out that SNG and have lost $3. Later on different tables I've been dealt AA twice more, winning small pots from low stacks going all in both times. Now even though i have a few more chips I haven't won anything from those results. Now If Aces will win a certain percentage of the time, what if that percentage you lose by is worth 10 times the amount you win by? How does that come into Variance?

If, when heads up at the final table of the WSOP, you are equal with chips with the other player and you get dealt AA and get beaten, the difference in prize money is millions of dollars. Would someone make that back up in the future by making the right decisions and believing in variance?

So if you keep making the right decisions things will work out? I can understand that if every decision was equal, but they arn't.

I also think it's easier to lose money at poker than win it. Again if you dump a buy in a NL10 with AA, how many times would you need to dealt that hand again to win the money back? What if everytime you lose with AA you get stacked and the times you win you win a small pot? How does that come into variance?

I really don't understand it

People at he $3 limits make it a minefield going all in with trash and calling raises with anything. I will let any of you log into my account and play 10 Five seater $3 SNG and see if it is possible to win consistantly from it. Just ask and i'll give you my log in details...

15 comments:

Anonymous said...

long term making the right moves pays off. wsop main event, equal chips, heads up your oppo goes all in, you have aces and call. he turns over xx and hits another on the river you millions down. play that over ten times and who is more likely to win more money?


beckford19minutes

Meteoric said...

ok, I played 4 with the intentions of playing 10, but got way too hacked off with it and had to go and do something more useful (like buying some beer)

Got one 1st and two 2nd places so at least I made a profit, but the shove/fold crap was really annoying.

On the last one I was HU and blinds were only 100/200 but my opponent had just two buttons - shove and fold. After a dozen flips he got it. Actually both 2nd places were similar - pure bingo for 1st place).

Don't know why you bother with them...

Having said that, getting itm wasn't that hard to be honest.

United113 said...

yes but you won't get the chance to play that ten times and thats my point. Everything is situational, if he played 10x and the time he lost just happened to be heads up at the final table of the WSOP then he'd be miles down.

I understand that making the right moves pays off etc etc just not how amount won or lost is figured into the whole variance idea.

Meteoric said...

Oh I see what you're saying.

The cash win is factored into ICM isn't it? It takes into account whether the risk of the shove/call is sufficient given the potential gain.

Variance based on just the cards + pot odds is just for cash games where you will play the same hand many times in the future.

United113 said...

see this is what i was trying to understand!!

Meteoric you said - The cash win is factored into ICM isn't it? It takes into account whether the risk of the shove/call is sufficient given the potential gain.

Can you expand?? If i'm honest thats gone straight over my head and think if you explain that, would make me understand!!

John said...

Variance is a dirty word

United113 said...

it is!! people talk about it and i understand making the right moves long term etc etc but not a clue about ICM etc that meteoric said...

United113 said...

Meteoric

but got way too hacked off with it

Yeah thats probably my problem with them to be honest..It really is push/fold alot of the time

Meteoric said...

ah, you MUST buy Kill Everyone - an awesome book for tourney's and SNGs (link on right of my blog, 3rd book down, lol I'm such an affiliate whore :-) ). Seriously though, it is a good book.

Here's a description of ICM in the mean time;
http://www.sitandgoplanet.com/sitandgo/sng_bubble/Introduction_to%20ICM.html

ICM uses your stack size as equivalent to how much of the prize pool you deserve - ie you're past the bubble and total prize money is $100 shared between you, if you have 30% of the chips then your tournament equity is $30 (30% of the prize).

You then have AT faced with a shove from a bigger stack - you think you have a 60% chance against his range - ICM calculates whether that 60% chance is low enough risk considering you're staking your 30% equity of the total prize.

Not exactly something you add up in your head, but there are plenty of calculators out there, and Kill Everyone is full of charts showing which hands are worth shoving or calling with dependent on your 'M' (or CSI as they call it). I guess you know what M is, but don't forget what it really means - it's your equity of the tournament (which is why when it's really low you go crazy shove fest because you literally have nothing to lose).

Meteoric said...

I guess it get's a bit more rational for higher buy-ins - like I said we were 5k in chips each with blinds 100/200 - highish, but not enough to spaz out and shove every third hand.

I haven't really played many tournaments tbh - it's only this husng stuff that's got me back onto them...

Anonymous said...

for true variance its important to be playin the same stakes over what ever sample size used..thats how roi is worked out.then said situation's mean nothing

United113 said...

yeah thats a good point, i tend to flit between buy in amounts and always seem to be outdrawn the time i move up

Anonymous said...

for true variance its important to be playin the same stakes over what ever sample size used..thats how roi is worked out.then said situation's mean nothing

Anonymous said...

Great blog as for me. I'd like to read more concerning this matter. Thank you for posting this information.

Anonymous said...

Not sure where to post this but I wanted to ask if anyone has heard of National Clicks?

Can someone help me find it?

Overheard some co-workers talking about it all week but didn't have time to ask so I thought I would post it here to see if someone could help me out.

Seems to be getting alot of buzz right now.

Thanks