Drawing Conclusions from Distribution

[craigthedeac]

Obviously an ideal distribution would be 100% first place. After a signficant number of tourneys, what sort of pattern should you see (percentage-wise) from a good (but realistic) distribution.

i.e., should one be satisfied with seeing a relatively equal distribution between the 1st-3rd places? How should 4th place's distribution compare to 3rd place's? What about 5th, which SNGPT says is one of the best times to look to double up.

What sorts of adjustments do players need to make to achieve the ideal (and realistic) distribution, whatever that may be?

For example, say a player has:
1: 16%
2: 8%
3: 16%
4: 8%
5: 16%

What conclusions can you make about such a distribution and do you think it is good? How about a player with:

1: 13%
2: 13%
3: 13%
4: 13%

I think this subject has a lot of potential for good discussion and SNG strategy, go!

[splashpot]

Finish distributions are stupid.

[pineapple888]

We've covered this a bunch in the past. Your first example is unsustainable over a large sample size (N>2000). You really need at least 1000 to even start to draw conclusions.

Most winning players have something like 14/12/13 over a large sample size with those numbers decreasing slightly as you move up in levels. Most OOTM bustouts should be 5-7th place. I was 14/11/11 at the 55s before I stopped keeping detailed records.

So 1st slightly greater than 3rd slightly greater than 2nd. Any large deviation from that over a large enough sample size might indicate a leak.