New 200+15er has some variance questions

[Fortitude]

I'm new to SNG's having logged about 200 of the 200+15's and I'm a regular in the mid-limit hold 'em games. I've been doing relatively well and noticed that a solid 200+15 player has a similar win rate to a solid 15/30 or 1000 NL player. So now that the numbers are there I'm interested in variation and what the 'long term' is in these higher limit SNG games.

In limit hold 'em we can say that if after 25k hands you are a 3 BB / 100 winner then you can be 95% sure that you are a winning player in this game. If after 100k hands you are a 3 BB / 100 winner then you can be 95% sure that your true win rate lies within +/- 1 BB of that 3 BB / 100. (I'm not 100% sure on these numbers but they're close and if you want to know exactly run a search as there are many threads on it)

What is the SNG equivalent of these kinds of statistics? How many SNGs do I need at what ROI before I know I'm beating the game? How many SNGs do I need at what ROI before I'm getting somewhat close to my true win rate?

Thanks guys.

[mike28]

a million

[Al P]

1000 - 5000.

Think of it this way. A SnG lasts say 70 hands max. Most of the time you won't make it that many hands. One of the most important places is on the bubble (4-5 players left). This will usually last 10-15 hands or so. So even after 1000 SnG's you haven't even played 10,000 bubble hands (since you won't make it to the bubble every time). Now you need even more SnG's before your in the money (hand distribution/winning %) starts to even out because you probably only make it there 35% of the time or less.

And none of the even considers running good or bad /images/graemlins/laugh.gif

[deathpotato]

If one is only making the money 25% of the time one should really re-evaluate one's game of choice /images/graemlins/shocked.gif

[Al P]

I'd take it at the 200's.

[deathpotato]

You'd be happy making the money 25% of the time when those make up 30% of the possible finishes?

[Al P]

Oops, was thinking ROI lol. Doh, time to edit.

[Apathy]

What is your party SN/SN's?

[Fortitude]

1000-5000 is a pretty wide range. Isn't there a more solid consensus on this matter? And did you mean 1000-5000 before you know whether or not you're a winning player in that game or 1000-5000 before you're starting to know what your win rate is at? I'll assume the former. Either way I'm a little shocked nobody has any solid answers for these types of questions.

[Voltron87]

winrates are overrated.

[BradleyT]

Controversial post to get you started (http://forumserver.twoplustwo.com/showflat.php?Cat=&Number=2302700&page=0&view=colla psed&sb=5&o=186&fpart=1)

[lorinda]

I think it's something like 75,000 to know your ROI within 1%

Your ITM converges a heck of a lot quicker and you can start to make some guesses at things after 1000-2000, but ROI takes forever.

Lori

[Freudian]

[ QUOTE ]
1000 - 5000.

Think of it this way. A SnG lasts say 70 hands max. Most of the time you won't make it that many hands. One of the most important places is on the bubble (4-5 players left). This will usually last 10-15 hands or so. So even after 1000 SnG's you haven't even played 10,000 bubble hands (since you won't make it to the bubble every time). Now you need even more SnG's before your in the money (hand distribution/winning %) starts to even out because you probably only make it there 35% of the time or less.

And none of the even considers running good or bad /images/graemlins/laugh.gif

[/ QUOTE ]

Not to mention most players have an evolution of their game during those first 1000-2000 SnGs, making the numbers hard to interpret.

[Fortitude]

I don't see how my SN's are at all relevant to my post.

[Apathy]

[ QUOTE ]
I don't see how my SN's are at all relevant to my post.

[/ QUOTE ]

I might be able to tell you if I think you can beat the game or not.. that was your original question was it not?

[the shadow]

Check out the threads at section 5.2 of the favorite threads (http://forumserver.twoplustwo.com/showthreaded.php?Cat=&Board=singletable&Number=191 8735&fpart=1&PHPSESSID=) list.

[Paul2432]

If ROI is 10% then after 1000 you get be 95% sure you are better than break even.

After 5000 95% +/-5%.

edit to add:

[ QUOTE ]
I'm not 100% sure on these numbers but they're close and if you want to know exactly run a search as there are many threads on it)


[/ QUOTE ]

Paul