Math guys: Chances of OTM 8 times in a row?

[1C5]

For a 20% ROI player?

This has been happening to me this month about 80% of the days that I have played. It is incredible. Just wondering the math behind this happening so much. I really don't think it should be happening that much (therefore, I suck this month/new leak/really, really unlucky).

Thanks.

I think we would need an ITM % as oppose to ROI.

[deathpotato]

Your ITM would be a lot more useful than your ROI to figure this out. And it's not very unlikely, even if you cash 40% of the time you will have 8 in a row OOTM every 60 games or so.

[Seth Money]

[ QUOTE ]
Your ITM would be a lot more useful than your ROI to figure this out. And it's not very unlikely, even if you cash 40% of the time you will have 8 in a row OOTM every 60 games or so.

[/ QUOTE ]

I agree with that I had a 42% ITM and had about 6 or 7 streaks likes this +/- 1 tourney over 600 tourneys.

It sucks when its like

5- 4ths
3- 5ths
2- 6ths

but it happens. The geniouses here can give you a percentage, not I.

Seth

[MegaBet]

Welcome to the rough end of variance.

I'm estimating a 20% ROI player would probably have an ITM close to 40%, so OOTM of 60%.

Pr(OOTM 8 times of 8)=0.6^8=0.0168=1.68%

So basically if you divide the games you play into sets of 8 a player with a 40% ITM can EXPECT to finish OOTM in all 8 SNGs from the set about once every 60 sets, or 480 games.

[1C5]

Yeah my ROI is right around 40.

Thanks for the calculations.

[deathpotato]

OK I've always been a little confused about this, hopefully someone can straighten it out - you say that every 60 sets you should have an 8 OOTM streak, which is the same figure I came up with, but I'm defining the sets as every consecutive 8 games, so that any game that's not within the first or last 8 is effectively counted 8 times, while with your method each game is in only one set. i.e. if these were your stats:

ITM, OOTM, ITM, OOTM, OOTM, OOTM, ITM, OOTM, OOTM, ITM

you would have 3 sets of 8 games in that selection, rather than 1.25. It strikes me that my method makes more sense, because when you look at it the other way a run of 10 like:

ITM, OOTM, OOTM, OOTM, OOTM, OOTM, OOTM, OOTM, OOTM, ITM

would not be considered a run of 8 OOTM - you had 7 OTM in your first 8 and 1 in the next two. Am I wrong? Also, my stats seem to agree much more with my figure than yours.

Yeah my calculation is taking 8 games as seperate. So if you play in sets of 8 and you ITM first 4 and lose the next 4. And in the next set record the first 4 as OTM and then ITM for the next 4. Then my calculations wouldn't account for that.

So to take this into account, the running total of OOTM, not looking at each set individually you would have to divide by 8. But for a running total if you had OOTM of 9, it would be counted as 2 streaks of 8 and 16 OOTM would be 8 streaks of OOTM 8 times in a row. /images/graemlins/crazy.gif

So it's gotta be more than 8 OOTM every 60 games as well but less than 480. /images/graemlins/confused.gif

[ QUOTE ]
I'm estimating a 20% ROI player would probably have an ITM close to 40%, so OOTM of 60%.

Pr(OOTM 8 times of 8)=0.6^8=0.0168=1.68%

So basically if you divide the games you play into sets of 8 a player with a 40% ITM can EXPECT to finish OOTM in all 8 SNGs from the set about once every 60 sets, or 480 games.

[/ QUOTE ]

This is wrong. There are 60 sets of 8 consecutive games in 67 games. Your odds of finishing OOTM 8 in a row in a set of N games is thus given by 100*(1-(.9832)^(N-7)) where N > 8. I think that should be right.

Streaks are very common. In 100 games, expect to see a set of 8 OOTM in a row ~80% of the time. This assumes your OOTM is 60%. If it's not, replace .9832 with (1-OOTM^8).

Hope this helps.

[Ogre]

i went 0-10 then 4/3/0/1 my next set

Thanks for clearing that up Noah.

Yeah just realised that although it's a 1.68% chance of losing all 8 out of 8 for a single set, all the inferences I made from that were wrong.

[axeshigh]

It happens all the time.
In the past 2 days I've had 36 games without a win, then 1 win, then 40 games without a win, then 9 wins out of 24. (worst ootm streak only 13 though) I am somehow still sane. Or never have been /images/graemlins/confused.gif

[ QUOTE ]
Thanks for clearing that up Noah.

Yeah just realised that although it's a 1.68% chance of losing all 8 out of 8 for a single set, all the inferences I made from that were wrong.

[/ QUOTE ]

NP. Probability math sucks and should die anyway, and is always much easier to correct than do.

(I'm mainly posting on these boards now because I stayed in tonight to work on probability hw /images/graemlins/smile.gif)

[Scuba Chuck]

[ QUOTE ]
[ QUOTE ]
I'm estimating a 20% ROI player would probably have an ITM close to 40%, so OOTM of 60%.

Pr(OOTM 8 times of 8)=0.6^8=0.0168=1.68%

So basically if you divide the games you play into sets of 8 a player with a 40% ITM can EXPECT to finish OOTM in all 8 SNGs from the set about once every 60 sets, or 480 games.

[/ QUOTE ]

This is wrong. There are 60 sets of 8 consecutive games in 67 games. Your odds of finishing OOTM 8 in a row in a set of N games is thus given by 100*(1-(.9832)^(N-7)) where N > 8. I think that should be right.

Streaks are very common. In 100 games, expect to see a set of 8 OOTM in a row ~80% of the time. This assumes your OOTM is 60%. If it's not, replace .9832 with (1-OOTM^8).

Hope this helps.

[/ QUOTE ]

I've read this about 4 times now. My head hurts. /images/graemlins/tongue.gif