Questions about rankings systems...

[THAY3R]

Who would get more points in your system do you think?

A player who plays 200 tournies with a 20% ROI, or a player who plays 50 tournies with a 30% ROI?

[N 82 50 24]

Hrmm, not sure. Although there would be some level of play requirement.

I guess I need to run some tests.

[FortunaMaximus]

I gave it a shot:


BI * sqrt(runners/rank) * 2[log(winnings)] = score


20/180 win:

22 * 13.416 * 2[log(1080)] = 1791.57

18th in same tournament:

22 * 3.16 * 2[log(43.2)] = 227.4


100k win with 888 runners:

109 * 29.8 * 2[log(28824.48)] = 28973.44

81st in same tournament:

109 * 3.31 * 2[log(293.04)] = 1779.47

Sunday 250k win with 1821 runners:

215 * 42.67 * 2[log(71456.04] = 88,988.29.

18th in same tournament:

215 * 10.06 * 2[log(2075.94)] = 14,430.27

180th in same tournament:

215 * 3.18 * 2[log(437.04)] = 3,610.65

As for non-score deductions, simply subtracting the buy-ins for nonscores should suffice.

Basing this on a very small data sample though.

Minimum # of MTT's = Minimum at-bats in the MLB batting averages system?

<shrugs>

[N 82 50 24]

I will have to investigate. It seems very buyin heavy with 180th in a 1821 tournament giving you over twice as much as a 20/180 win. Of course, I'm sure the point buyin is high, but still seems a little unfair.

I do think that buyin should play a decent role though. Getting more points for a $5 FO win than a 109r win (like the TLB in some cases) is retarded IMO.

[FortunaMaximus]

Attempt 2:

This might correct the field size skew somewhat:

BI * 100[(runners/rank)/runners] * 2[log(winnings)] = score

20/180 win:

22 * 100 * 2[log(1080)] = 13347

18th in same tournament:

22 * 5.6 * 2[log(43.2)] = 403

100k win with 888 runners:

162 * 100 * 2[log(28824.48)] = 144,496.3

81st in same tournament:

162 * 1 * 2[log(293.04)] = 799.3

Sunday 250k win with 1821 runners:

215 * 100 * 2[log(71456.04] = 208723.7

18th in same tournament:

215 * 5.6 * 2[log(2075.94)] = 7988

180th in same tournament:

215 * 0.6 * 2[log(437.04)] = 681.25

$5 FO win (1627 runners):

5 * 100 * 2[log(1464.31)] = 3165.63

109r win with 194 runners (446 rebuys, 139 add-ons): (Adjusted BI = Prize pool/runners)

401.54 * 100 * 2[log(23370)] = 350838.24

A 109r win > a 250k win.

27 20/180 wins > 1 109r win.

(Also corrected 100k numbers, as BI is 162, not 109.)

Something to keep in mind:

Scores will drop as a function of non-scoring buyins, so the ratios may well be equitable in the long-term.