I understand most of you think 10,000 hands is a good baseline number to eliminate variance in ring games, but how many tournaments do you guys believe are needed to determine accurate ITM and ROI numbers? Does it change depending on the number of entrants (I assume so, since an MTT will have a much higher variance than 10 person SNGs)? For the sake of this question, give your opinions on 5, 10, and 30 person SNGs.

oh yeah, :rockpoker

I don't think its possible to play enough MTTs to get a true count. SnGs, maybe 250 for an early check up?

__________________
“There's no sense in being precise when you don't even know what you're talking about.” - John von Neumann

Actually, I dont think 10k is enough, and game conditions change some much I am not sure there is a number.

You can get confidence intervals, but I don;t have experieince in calculating them.

You can probably get an idea if you like the things or not after 50 (doh kills my 3 table thread), and get a sketch of a picture of whats going on. Spme think this is too small, but I think its a good start.

Also remember, ROI and ITM are based on histoical figures and really cant be used to be predict futures trends (as much as I would like with the 3 tables). You can do this with historical WR's to say if I play x of hours of ring, I expect to make y $. While you can say with my ROI, I expect to make x $ of my y $ investment, you can't expect to see this rsult each time due to the hgh variant nature of SNG's. MAybe You can, but I just dont play SNG's like a do ring, I sit and play 4 hours of ring to grind.

__________________

You could not place in a 100 straight multitables and have a -100% ROI and then win a big tournament and have a huge ROI. Its nearly impossible to caculate ROI for MTT's. AS for SnG's Id say 150/200 would give you an accurate number.

From the maths point of view you consider all the permutations of holes cards, flop (not every flop,- there are about ten types of flop), and table (really there's only about four distinct type of players).

When you go through the permutations you need about 100k hands before you can see if your playing low pp an AT correctly. You can make money out of these but they are small +EV.

Actually the same applies to tournaments. You might argue a bit about the the number of people on a table but you need the same number of hands to fully sample all the permutations.

It is too easy to win a Tournament by getting AA or KK a couple of times and winning the odd coin flip. With 100 ppl you only have to win 4 all ins to get to top table. If there are 200 ppl then you only need to win 5.

If you consider that in an average tournament you will play 100 hands (including early bust outs and going all the way) then you need to play 500 tournaments before you know if you're a winner.

Tournaments have very high variance. The sites love them because anyone can win -good players will generally do better than bad players - simply because they are more willing to hit the all-in button on marginal calls.

It's got to be said as well that most tournament structures on the net do not allow a good player to 'get hold' of a table as they can in ring. It takes 20-30 hands to start to 'work' a table in ring. Tournament players don't do this simply because in a MTT you will change table too often.

Invariably this is why 'tournament' players get killed in ring (and why many pros make thier money in ring).

Tournaments are very exciting and whenever I feel like I need a break from poker... I will play one...

__________________
See me playing $10/$20NL like it was play money

http://video.google.co.uk/videoplay?...405&q=xxdemexx

Doberman: "but Sarge, isn't poker gambling and just luck?"
Sgt. Bilko:" not the way I play it"

I have no idea what my MTT ROI is. I must have played hundreds and have had several large wins which more than make up for the deserts in between.

Whether I am any good or not is at this point definately not statistically proveable with 30k hands in PT. I make the money about 1 in 4 (with 10% of the field paying) and have made about 20 final tables. I'm certain my game is decent, but I could easily have made my profit of the back of the probability curve rather than through skill.

SnGs you can get a sense of where you are after 100 but a lot more to really establish how you are doing, and of course assuming conditions don't change.

Ah yes all the wonderful talk of mathematics and variance in SNG'S but if i am not mistaken SLP i played with you on a SNG table $5 turbo last night at prima and you called my ALL IN RAISE WITH Q-K UNSUITED AND SUCKED OUT WITH A QUEEN ON THE FLOP.I of course had A-K.Can you please tell me what your mathematics were on this call as you called the tightest player at the table all in with this crap.If slp718 is not you disregard this post.lol

Well as long as you're not bitter about it.

You seem angry about maths being applied to poker. Here and in the Doyle V Sklansky thread.

I don't think anyone is saying that maths is the be-all and end-all of the game, it's just an important tool among others in starting to understand it.