A poll about ROI

[AleoMagus]

I wanted to post this in reply to Saborion's 20+2 and 30+3 ROI question but it seems that I couldn't figure out how to put a poll in a reply...

Anyways, what are your ROI results in 10+1 to 30+3 single table SNGs. 30%+ ROI results are thrown around a lot on this forum, but I suspect a lot of them come from 100 tournament samples or less. In the long term a lot of good players might not even be doing that well. A positive ROI is great and nothing to scoff at, no matter what it is.

I have no doubt that there are 40% ROI players out there, but I think they are few and may still be on the happy end of normal fluctuations (and trust me, you can still be deviating from the norm after even 1000 sngs)

[RPatterson]

I take that back! I've played over 100 sng's just haven't kept stats!

[adanthar]

Before I came back to SNG's, I had about a 10-15% ROI at the 30's through a little over 100 tourneys. (A bad streak at the very end hurt those a bit, as well.)

After I took a break from ring games and came back, I feel like I've become more aggressive/a better player and now average around 40% through 60-70 tourneys. My overall stats are around 25% through 181, which is what I voted; I think my true actual winrate is probably in the middle, around 30-35%.

I don't multitable...but then again, I don't pay that much attention, either. Maybe I should start [img]/images/graemlins/blush.gif[/img]

[Kurn, son of Mogh]

Overall ROI after 250+ SNGs is a shade under 40%.

I don't multi-table.

[Phill S]

the first question is no problem.

but in the second i dont think lumping i dont multitable with i dont keep stats etc was hugely helpful.

imo, keeping them seperate means you can get a more meaningful stat based round how many people who do keep stats but stick to one table at a time. as it currently stands, you have a slightly unclear result.

Phill

[Peter Harris]

After 175 SNGs and small sample notwithstanding,

1st 100 went for 34% ROI
2nd 75 went for 26% ROI. This has been addressed and ironed out largely in recent games that aren't in these stats, but it was due to bad play more than bad luck. It's still comforting that my bad play merits a 25% ROI.

150 more 10+1s and i'll jump up a notch, i warrant. I'd like to see the 3rd 100 go for a 35-42% ROI range.

Regards,
Pete Harris

[AleoMagus]

Perhaps my assumptions about ROI extectations are wrong but I really think these kinds of ROI hopes are misled. I'm not directing this at one poster either, but at a very large group who I think makes these kinds of statements.

To expect ROI to settle so uniformly around 5-10% from one 100 tourney sample to the next is unrealistic.

To suggest that dropping from 36% to 24% was due to bad play or that since correcting errors you can now expect 35%+ again is somewhat wishful thinking.

Don't get me wrong, if you think you had leaks in the last 75, you probably did, but that does not in any way explain an ROI decrease like that.

I think that players on here with 30% ROI values expect results like this:

1st 100: 30% ROI
2nd 100: 28% ROI
3rd 100: 35% ROI
4th 100: 26% ROI
5th 100: 28% ROI
6th 100: 34% ROI

and so on, all converging to a solid 30% along the way.

If any of you you are getting results like this, great for you, but the uniformity of these results cannot be counted upon. To start worrying about leaks when ROI drops over 5% over the past 100 can be a big leak in itself.

If my assumptions about Sng results obeying something like normal distribution are correct (and I think they are a good approximation) then this means that a player getting a 36% ROI is only about 65% likely to earn over a 20% ROI in his/her next 100. There is about a 5% chance that they will only break even. This assumes an even quality of play.

This means that in 20 100 sng samples, a player earning an ROI of 36% can expect to get at least one sample that is break even or even losing! they can also expect one sample with an ROI of 70% or greater! one out of every three samples will be less than 20% and one out of every three will be over 50%.

In my own experience, This has been the case and I used to focus a great deal on these results, thinking that I was playing great, or playing terrible. More and more I realize that normal fluctuations were just more extreme than I imagined.

When I first started posting on these forums I was convinced that I had an ROI of 55% from a sample of over 200 sngs. I have since settled on a more realistic ROI over almost ten times as many sngs.

It can be tough, but if a person is not prepared to face a break even stretch of sngs without the temperance to remain confident in their play, tilt will destroy them.

Again, I am going off here to poor Peter, but this is really directed more at this forum as a whole.

Maybe others will offer counter arguments about the stability of their results? Do any others have LONG term samples (1000+ sngs) which have remained extremely stable throughout?

There have even been posts on here in recent weeks talking about the swingyness of sngs. Some of these posts have mentioned one 10 sng sample to the next. This is great but I wonder if these posters are prepared for a miserable 200 sng break even/losing streak. This is definitely a rare thing if you are a winning player, but certainly within the realm of possibility.

If you are 'actually' a 30% ROI player, in your next 1000 sngs, you can expect anywhere from 20% to 40%. Granted, that 20% will only happen to 1 in 20 players, but that is not all that rare. 1 in 100 players will only get 14%. Remember, we are talking about the next 1000 sngs here!

I'd be really interested in thoughts on this. Do players here feel that this is accurate or does the math miss some fundamental skill aspect of sngs that forces sng results to converge more closely around true ROI? I don't find this to be the case. My results have taken huge dips and have taken huge leaps over large samples.

Regards
Brad S

[Tosh]

I think someone with 30% ROI could easily have 200+ streaks where their ROI is not only less than 10% but negative.

[aslowjoe]

NO idea what your talking about. MY 1st 200games in PT I have an roi of 53%.



Oh yeah, my last 258 -.03%. My true roi is over 50 though. I realized that my last 258 games I have been playing in shorts but when I was winning more I was playing in long pants. I will fix that leak when summer is over.

[JNash]

LMAO