probability of a winning player winning

[LuckySlevin]

I've been thinking - if you're a winning player at the moment, and consistently making a profit over a sustained period of time (say a year).

I'm curious how likely you think it that you would win at $1.20 9 man sng's, say if you entered ten would you expect to be in profit? I know that this is a very small sample, but that's part of the reason for asking.

Cheers, Gary.

[LuckySlevin]

As most advice here prescribes 50 Bi's, am I right in concluding that you'd expect to be winning once you've entered about 50?

[spoonitnow]

This is a basic question that statistics can answer. However, I'm kind of busy at the moment.

[d0zer]

I think 30-35% ish ROI is a decent winrate for SnGs, so if you entered 10, spending $10, you'd win $3 - $3.50 ish?

Those $1.20s are so soft though I'd be surprised if a decent player couldn't hit a higher winrate.

Winrate is proportional to game softness obviously so once you get up to like the $100 SnGs, i think 10% becomes decent.

[wellrounded08]

Hmm... Here's my immediate thought on it. It may be useless however.

If we could put a number someone's quality of play, as in, Bob is better than 70% percent of the players at that level. Couldn't Bob expect to win roughly 70% of the time? Let's say bob wins 1st 2nd and 3rd an equal amount of time in the 70% he does win in 100 9 manSNG's
That's 1st 23.3 times 2nd 23.3 times and 3rd 23.3 times.
Assuming the payout goes like this, that means:
1-4.50=104.85
2-2.70=62.91
3-1.80=41.94
Total won-Total paid =89.35$ profit.
Being better thatn 70% of players however, you're probably going to lean more toward 3rd place alot so lets make it 1st-10 times 2nd 20 times and 3rd 40 times.
1-4.50=45
2-2.70=54
3-1.80=72
Total Won-Total paid=51$ profit.

This may off. I'm not a mathmatician. Or it might just not apply really. I dunno, but it's what came to mind, and it made sense when I typed it.

[Ragnar4]

Think about it Slevin. If you only won one in 50 matches, you'd go broke in about 70 matches.

you win 4.50 for first
2.70 for second
1.90 for third

Since you pay 1.20 per bout
You profit 3.30 for first
1.50 for second
and .70 for third.

If when you finish in the money, you only ever finished 3rd, You'd have to finish ITM right around 62% of the time to break even

If all you ever did was finish 2 you'd have to finish there right around 44% of the time to break even.

If all you ever did was finish first you'd have to finish there 23% of the time to break even.

So. When people talk about having a 30-35% ROI

In ten games you'd have to finish 3rd 8 times out of ten to have a 25.8% ROI

In ten games you'd have to finish 2nd 6 times out of ten to have an ROI of 35%

and in ten games you'd have to finish first 5 times out of ten for a 45% ROI (Here out of 10 games, if you finished first 4 times, you'd have an 8.75% roi)

I calculated these by using my spreadsheet calculator and just brute forcing wins from 0% till I hit somewhere around 30%

So, to answer your question Slevin, You have to finish ITM Somehwere between 23% and 62% of the time to just break even. What's more, the more skillful you are, and the more likely you are to win an SNG the less often you have to finish ITM to be a succesful player.

[Ragnar4]

What's more, the more skillful you are, and the more likely you are to win an SNG the less often you have to finish ITM to be a succesful player.

--Wow, that's pretty fucked up.

[daven]

I've been thinking - if you're a winning player at the moment, and consistently making a profit over a sustained period of time (say a year).

I'm curious how likely you think it that you would win at $1.20 9 man sng's, say if you entered ten would you expect to be in profit? I know that this is a very small sample, but that's part of the reason for asking.

Cheers, Gary.

ummm, depends:

103774939 23-Aug-08 06:42 NL Holdem $1 9 -$1.20
103772950 23-Aug-08 06:10 NL Holdem $1 9 -$1.20
103751135 23-Aug-08 04:46 TURBO NL Holdem $11 180 -$12
103654040 22-Aug-08 19:12 FL 7 Card Stud H/L $50 2 $47.50
103648637 22-Aug-08 18:52 FL 7 Card Stud H/L $20 2 -$21
102778883 19-Aug-08 03:47 TURBO PL Omaha $11 4 -$11.50
102981402 19-Aug-08 03:13 NL Holdem $21.25 2 -$22.25
102960811 19-Aug-08 03:12 NL Holdem $10 45 -$11

[animal_chin]

Here is an article about variance and this one is about variance in MTTs. Both these articles are mostly about MTTs, but you can see that with even vary large MTTs samples variance can still be a bitch. This is less so for SNGs, but I hope you see the point.

[badgers]

From my past SnGing I seem to recall the general advice was play 200 to get an IDEA of where you are at. If you are losing after 200 it's time to really think about your play. If you are winning then things are good. However, it could still be just variance etc. I think around 1k tourneys is a good estimate of your ROI.

However, it depends how many $50 limit Stud h/l tourneys you enter in the meantime.....

[spoonitnow]

There are like 2 sentences in this whole post that are anywhere close but unfortunately I don't have access to my statistics texts where I'm at right now so I'll give you a general idea. You take your average win-rate and standard deviation and do some fun shit and you can tell the chance of being up X or more (or less) in Y tournaments.

[Ragnar4]

If any of my stuff is off, I'd like to know what it is and why.

[wellrounded08]

ME>PWND

Yeah, don't pay attention to my post then.

[Fielmann]

If any of my stuff is off, I'd like to know what it is and why.

What you said was correct but you misunderstood Lucky's question. He didn't mean to say that he expects to win one tournament in 50. He asked:

1) What's the probability of being in profit after 10 tournaments?

and

2) Should I expect to be in profit after I've played 50 tournaments?

Answers:

1) What spoonitnow said. Take your ROI and standard deviation and then there's a formula in math books somewhere.

2) How do you define "expect"? You you can never be certain. My gut feeling is that an average winning player will be in profit after 50 tournaments with a probability of about 90%-95% or something like that. On the other hand: an average winning player will be in profit with a probability above 50% after 2 or 3 tournaments.

[Chopper]

ask robb about standard deviation. oh, spoon, too, since i didnt see that the first time through.

/quandry

[Jack Sawyer]

this is all crap.

this is poker, not roulette.

play bad enough, and you may drop 100 sng's in a row.

in roulette, if you actually stick to just two numbers, you still have an approx. 1/17 chance of winning a spin, notwithstanding of any other factor.