Question about finish distribution.

[djj6835]

I'll try to make my question as clear as possible. I'm mainly asking this of posters who have several thousand tournaments worth of records.

Say I have played my last 1000 tournaments with a 29.5% ROI and the following finish distribition, 1st-13.7%, 2nd-15.8%, 3rd-13.5%. Also assume that this group of 1000 tournaments is about a year and a half after I started playing SNGs, and not my first 1000 games played.

Ideally I would like to get a better distribution of finishes for my next 1000 that I play. I was thinking basically swapping my current 1st and 2nd distribution would be great. This would mean about a 4% increase in ROI.

Im wondering what kind of progress some other players have made as they played more games. Is this type of goal reasonable? Did your stats begin to flatten out at a certain point so that you really didnt improve your results from onle large block of data to another?

Thanks for any responses.

[Lori]

137 wins vs 158 losses heads up results is such a small sample that there is almost zero way of telling if you have improved over the next 1000 games, even if you come out of it with 158 vs 137.

If your HU battles were all coin flips, there is a 24% chance that one of you would have won 158 or more of those flips. (12% chance it's your opponent)

For more info on calculating HU probabilities, you can mess with this (http://faculty.vassar.edu/lowry/binomialX.html) site and work out confidence levels that you will be happy with.

Lori

[axeshigh]

[ QUOTE ]


Ideally I would like to get a better distribution of finishes for my next 1000 that I play. I was thinking basically swapping my current 1st and 2nd distribution would be great. This would mean about a 4% increase in ROI.


[/ QUOTE ]

I think you should focus on your general game and/or moving up with that kind of stats. What buyin are you playing? There's really no point of focusing on your finish distribution as it's not something you can affect directly, and even if you had some kind of superpowers and managed to switch your firsts and seconds, you might have less thirds and more fourths and get a drop in hourly rate.

[djj6835]

[ QUOTE ]
There's really no point of focusing on your finish distribution as it's not something you can affect directly, and even if you had some kind of superpowers and managed to switch your firsts and seconds, you might have less thirds and more fourths and get a drop in hourly rate.

[/ QUOTE ]

I'm not really worried about changing my results. Whatever happens, so be it. Just curious as to what kind of statistical improvements other players have shown, but as Lori stated, the actual HU sample is so small the results don't actually mean anything.