SNG statistics

[costanza_g]

[ QUOTE ]
...murmur, murmer....150 games......murmur, murmur....

[/ QUOTE ]

BURN HIM!!! BURN HIM TO HELL, OFF WITH HIS HEAD!!!

[Indiana]

[ QUOTE ]
Maybe I've misunderstood something, but I used a 95% confidence interval to calculate that: http://en.wikipedia.org/wiki/Confidence_interval

<p - z sqrt(p(1-p)/n), p + z sqrt(p(1-p)/n)> = <.19 - 1.96 sqrt(.19(1-.19)/150), .19 + 1.96 sqrt(.19(1-.19)/150)> = <.13, .25>

Don't you think he's got any idea wether he's a winning player when he's played 150 SnGs with a ROI of 19%?

[/ QUOTE ]

This is not an appropriate CI calculation for ROI. It would be ok for ITM though. Do you see why?

Indy

[Hakon]

Maybe beacause ITM is only true or false, while ROI is an average of each game's ROI? I dunno, why is it? [img]/images/graemlins/smile.gif[/img]

[zipppy]

[ QUOTE ]
Well I have no idea what that crazy math means, but I can say that I'm a winning player and I can easily go over 150 sngs with a negative roi or a ridiculously inflated one.

[/ QUOTE ]

That's not as important as knowing whether or not a losing player can go 150 SNGs with a 19% ROI.

If the OP had a negative ROI and was wondering if he was truly a losing player after 150 SNGs, then it would be relavent to say that a winning player has done the same.

So...any losers among us? Anyone? Bueller?

...

zip

[Indiana]

[ QUOTE ]
Maybe beacause ITM is only true or false, while ROI is an average of each game's ROI? I dunno, why is it? [img]/images/graemlins/smile.gif[/img]

[/ QUOTE ]

By George u've got it. ROI is not a proportion so you cannot use this CI. Another poster has touched on the nature of correlated data, but ignoring that you could use something like a normal or exact CI.

Indy

[citanul]

I AM LOCKING THIS THREAD NOW BUT YOU SHOULD KNOW FOR FUTURE REFERENCE THAT I HATE YOU.

CITANUL