Willy Wanka, FemmeFatale, and jsheltz take top honors in one-month, two-month, three-month lifetime TAPL competition

Forum

Overview

Advisor Blog

Discussion

Reviews

Strategies

Ideas

Replays

Willy Wanka, FemmeFatale, and jsheltz take top honors in one-month, two-month, three-month lifetime TAPL competition

The First Quarterly Lifetime TAPL Competition

Results

One Month Category:

Special Mention:

soxrock:
35 0% 0% 2% 5% 0% 28% 62% 8283
âˆ«0th.âˆ«

Soxrock set out to show that the TAPL is not a measure of skill because it rewards negative skill as well as positive skill, as in this case where in only 35 games he racked up a TAPL of 8283. Most of his finishes were last place and most of the rest were sixth place. This does stand as a measure of negative skill although it would be harder to do if several players in the same game were competing for the lowest places.

Take this to heart, O young jedi knights, to not be seduced by the dark side of the TAPL!

WillyWanka:
219 12% 21% 17% 18% 13% 8% 7% 2565
âˆ«1st.âˆ«

JNeidlinger:
157 17% 21% 18% 14% 10% 11% 6% 1828
âˆ«2nd.âˆ«

CaptSundrop:
188 19% 22% 11% 11% 11% 11% 11% 1799
âˆ«3rd.âˆ«

Tactic_Max:
162 14% 15% 17% 14% 15% 11% 10% 417
âˆ«4th.âˆ«

Two-Month Category:

FemmeFatale:
498 12% 10% 9% 17% 18% 15% 17% 2663
âˆ«1st.âˆ«âˆ«

ChelseaSucks:
147 11% 24% 18% 14% 12% 6% 11% 2046
âˆ«2nd.âˆ«âˆ«

Three-Month Category:

jsheltz:
1368 21% 13% 14% 14% 11% 11% 13% 7451
âˆ«1st.âˆ«âˆ«âˆ«

kwizatz:
1098 19% 12% 11% 10% 13% 13% 20% 6993
âˆ«2nd.âˆ«âˆ«âˆ«

montecarlo:
1007 21% 16% 15% 13% 11% 11% 9% 6809
âˆ«3rd.âˆ«âˆ«âˆ«

rnd_:
376 25% 19% 12% 8% 8% 10% 14% 6395
âˆ«4th.âˆ«âˆ«âˆ«

Integral:
1038 19% 15% 11% 12% 11% 11% 17% 4814
âˆ«5th.âˆ«âˆ«âˆ«

Vermont:
735 17% 18% 14% 16% 13% 11% 7% 4698
âˆ«6th.âˆ«âˆ«âˆ«

Cleopatra:
787 19% 14% 15% 16% 12% 10% 9% 4312
âˆ«7th.âˆ«âˆ«âˆ«

mos0911:
757 12% 15% 20% 17% 10% 11% 12% 4223
âˆ«8th.âˆ«âˆ«âˆ«

I_suck:
1216 14% 10% 12% 12% 14% 15% 19% 4185
âˆ«9th.âˆ«âˆ«âˆ«

Stormlord:
780 20% 9% 15% 15% 12% 11% 13% 4137
âˆ«10th.âˆ«âˆ«âˆ«
In the case of Stormlord I could not access his end-of-month profile so the scores here are for his latest profile that I have.

skrumgaer:
670 13% 8% 18% 18% 14% 13% 13% 3841
âˆ«11th.âˆ«âˆ«âˆ«

joivir:
634 13% 17% 20% 12% 14% 12% 11% 2660
âˆ«12th.âˆ«âˆ«âˆ«

lief79:
790 14% 12% 16% 16% 14% 14% 9% 1979
âˆ«13th.âˆ«âˆ«âˆ«

Rsquared:
1005 17% 14% 13% 12% 12% 16% 12% 1679
âˆ«14th.âˆ«âˆ«âˆ«

rugbyjoe:
465 19% 11% 14% 10% 14% 13% 14% 1600
âˆ«15th.âˆ«âˆ«âˆ«
This is from his latest profile that I have. I could not find his end-of-month one.

WayneRooney:
205 11% 22% 15% 16% 9% 12% 12% 1542
âˆ«16th.âˆ«âˆ«âˆ«

lofty58:
556 15% 19% 12% 13% 12% 13% 15% 1315
âˆ«17th.âˆ«âˆ«âˆ«

LAG_Monster
381 14% 10% 12% 13% 16% 18% 14% 1205
âˆ«18th.âˆ«âˆ«âˆ«

captainLAGER:
433 18% 15% 13% 11% 12% 13% 13% 1147
âˆ«19th.âˆ«âˆ«âˆ«

frankstur:
224 17% 19% 12% 13% 12% 11% 12% 880
âˆ«20th.âˆ«âˆ«âˆ«

truce_whore:
282 17% 11% 12% 13% 15% 11% 17% 837
âˆ«21st.âˆ«âˆ«âˆ«

Lordscamz:
528 13% 16% 17% 12% 12% 13% 14% 812
âˆ«22nd.âˆ«âˆ«âˆ«

Jdizzle77:
737 12% 16% 15% 13% 14% 13% 13% 767
âˆ«23rd.âˆ«âˆ«âˆ«

jimmymak:
886 12% 13% 14% 15% 13% 15% 15% 639
âˆ«24th.âˆ«âˆ«âˆ«

shiffy:
459 16% 15% 13% 12% 12% 14% 12% 604
âˆ«25th.âˆ«âˆ«âˆ«

OldElvis:
688 13% 16% 14% 16% 13% 13% 13% 532
âˆ«26th.âˆ«âˆ«âˆ«

I â€˜less thanâ€™ 3u:
390 12% 15% 17% 14% 14% 12% 13% 372
âˆ«27th.âˆ«âˆ«âˆ«

I appreciate all of you participating in this competition.

+ reply

Replies 1 - 9 of 9

skrum,

does these calculations reflect your recent change to sqrt(games), and if not, could you humor me with the modified standings?

do.

WAHOO I GOT A SPECIAL MENTION FROM SKRUMGAER EVERYONE!!!!!!!!!!!!!!!!!!!!

Since rnd has broadcast to the world that I plan to change the scaling of the TAPL, I will go ahead and explain the change.

I have learned more about the shi square distribution (with the help of pambo) and found out that the TAPL goes as the number of games for the variance, which means that the standard deviation goes as the _square root_ of the number of games. The standard deviation is the standard way of making comparisons, such as "x1 is so many standard deviations higher than x2". The SAT scores and other statistical measures are set up so that a standard deviation is the same number of points no matter where you are on the scale. So I will do the same for the TAPL. For all future calculations, the following formula will be used:

new TAPL = 10 X old TAPL/(square root of number of games)

The 10X is to re-scale the new TAPL to make it roughly the same size as the old. If you have played exactly 100 games, your new and old TAPL's are the same. If you have played more than 100 games, your new TAPL will be smaller; if you have played less than 100 games, it will be bigger.

Now for the point that rnd is really getting at: any scoring system that doesn't have rnd as the winner is obviously flawed. The scores in my post are according to the old TAPL. If they had been done under the new TAPL, rnd would have won first place. Only three months until the second quarterly competition, rnd!

Starting as of now, all future TAPL's are under the new definition. Past TAPL's will not be changed.

lol. sorry, i didn't think it was a secret.

congrats to the winners. ;0

hey skrum, would u mind first of all, dumbing this down as has have tried to read ur profile and getting, 1)bored (lol) and 2) confused. and then would u mind doin mine please?

thanx

xxx

sarahxxx:

your games played, lifetime profile, and lifetime TAPL under the new definition is

539 13% 16% 14% 7% 10% 17% 18% 1434

Hang in there... I am soon introducing another metric that is much easier to understand than the TAPL and it predicts _only one_ skill--the ability to score elo points.

wouldn't that, in effect, be the integral of the slope of all the tangent lines to your elo curve?

Hast hit it, rnd.

The new metric is a convolution integral: the best linear unbiased estimator of the net elo per game from the percentage profile.

+ reply

KDice - Multiplayer Dice War

KDice is a multiplayer strategy online game played in monthly competitions. It's like Risk. The goal is to win every territory on the map.

GAMES

G GPokr
Texas Holdem Poker

K KDice
Online Strategy

X XSketch
Online Pictionary