Forum
golden ratios and golden tables
|
skrumgaer wrote
at 2:13 PM, Tuesday May 8, 2007 EDT
In regard to the current discussion about the appropriate gaps for the tables, I thought I would make some mathematical observations and set up a Quality-of-Opponent Adjustment (QOOA) that does not have to be re-calibrated. The QOOA and the TAPL that it modifies are discussed on my profile page.
The distribution of the number of tables of each level follow approximately the Golden Ratio (reference http://en.wikipedia.org/wiki/Golden_ratio), known from the time of the ancient Greeks. This ratio is 1.618034, and the conjugate and inverse of it is 0.618034. The number of kdice tables varies from time to time, but the ratio of the number of zero-limit tables to all higher tables, and the ratio of 1500-limit tables to all higher tables, and the ratio of 1700 limit tables to all higher tables, etc, seem to approximate the golden ratio. There is something aesthetically pleasing about this ratio for the tables, because regardless of your level of play, the proportion of the time that you can reach the next higher table is constant. Therefore I will set up a QOOA that does not have to be re-calibrated when new levels of tables are added. It is designed for six levels of tables, which is sufficient when the proportion of games played at each level is rounded off to a whole percent. The QOOA is QOOA = p0/371 + p1/142 + p2/54 + p3/21 + p4/08 + p5/03 where p0 is the percentage of games you have played at the zero limit tables, expressed as an integer, p1 for the next higher level of table, p2 next higher level, and so on. The table limits are to be disregarded; what is important is the level of table. The level of table is to be determined at the time you sit down. For example, if you sit down at a 1900 table when there are zero-limit, 1500-limit, and 1700 limit tables available, that particular game is to be counted as part of p3. If you sit at a 2000 table and a 1900 table is open, that game is counted in p4, but if you sit at a 2000 table and there are no 1900 tables open, that game would be counted in p3. Some sample calculations: Suppose you have played 50% of your games at the zero limit tables and 50% at the 1500-limit tables. Your QOOA would be 50/371 + 50/142 or 0.49. Suppose you have played 20% of your games at the zero limit tables and 80% of your games at the 1500-limit. Your QOOA would be 20/371 + 80/142 or 0.62. Your QOOA increases when you play more at the higher level tables. The QOOA multiplied by the TAPL gives you your TAPL adjusted for quality of opponents. Incidentally, the denominators in the QOOA formula closely approximate alternative numbers from the Fibonacci series 1, 1, 2, 3, 5, 8, etc. The Fibonacci numbers and golden ratio are closely linked. |
|
montecarlo wrote
at 3:35 PM, Tuesday May 8, 2007 EDT integral ate a sandwich today. he said the bread might have been a touch moldy.
he told me to post this vital information on this thread, since it seems at least as jaw-dropping as skrum's revelations of the golden rule. |
|
montecarlo wrote
at 3:36 PM, Tuesday May 8, 2007 EDT integral, if you were to post your sandwich story under the idea's page, then i would vote for it. it is more than pertinent.
|
|
redsox5445 wrote
at 2:14 PM, Wednesday May 9, 2007 EDT skrumgaer you seem to have a lot of time on your hands to do these wierd seemingly random math equations...just what i have noticed
|
|
_rnd wrote
at 8:14 PM, Thursday May 10, 2007 EDT i think its time for some interum may TAPL calculations.
|
|
_rnd wrote
at 8:25 PM, Thursday May 10, 2007 EDT I hope you don't mind skrum, i've taken the liberty.
RANK TAPL IDIOT ELO ELO_RANK 1 1 rnd 2337 (1st) 8 2 Stoudemire 2206 (9th) 2 3 LUCK 2253 (3rd) 6 4 montecarlo 2217 (8th) 5 5 integral 2241 (6th) 7 6 Canarioz 2124 (49th) 25 7 Mrdanger 2054 (149th) 11 8 Mikeypoo 2042 (171st) 9 9 wishbone 2159 (20th) 12 10 wowimbad 2119 (56th) 22 11 Vohaul 2075 (112th) 23 12 Vermont 2107 (71st) 10 13 petomni 2094 (85th) 24 14 jfdis 2146 (30th) 17 15 Kehoe 2089 (92nd) 15 16 uukrul 2078 (109th) 19 17 SarahPlaying 2057 (142nd) 20 18 Optical 2117 (59th) 16 19 rabbidmonkey4 2123 (51st) 13 20 Tuttle 2134 (36th) 4 21 Rsquared 2066 (126th) 3 22 riser 2264 (2nd) 18 23 joex 2039 (177th) 21 24 clomax 2173 (12th) 14 25 Sweden 2083 (98th) |
|
skrumgaer wrote
at 9:03 PM, Thursday May 10, 2007 EDT Nice try rnd, but you have to use Yates' Correction if the expected number of games per category is less than 5 (which is true if you have played less than 35 games). I refer you to the collected works of skrumgaer at my profile page.
|
|
_rnd wrote
at 9:34 PM, Thursday May 10, 2007 EDT umm... i'm first either way. run the numbers.
|
|
skrumgaer wrote
at 8:48 AM, Friday May 11, 2007 EDT rnd: just for you.
The TAPL's for you and riser have had the Yates' Correction made. 8 (5.11) Stoudemire 0 2206 (8th) 47 17% 29% 23% 14% 8% 4% 2% 1998 2 (9.50) X LUCK X 0 2253 (2nd) 82 20% 19% 18% 20% 8% 9% 2% 1863 6 (5.84) montecarlo 0 2217 (7th) 75 26% 9% 6% 21% 14% 12% 9% 1666 5 (6.04) integral 0 2241 (5th) 51 29% 21% 9% 15% 5% 7% 9% 1658 1 (9.78) rnd* 0 2337 (1st) 11 49% 31% 14% 5% 5% 5% 5% 1444 7 (5.46) Canarioz 0 2140 (34th) 118 24% 16% 16% 10% 11% 11% 10% 1332 25 (2.23) Mrdanger 0 2054 (151st) 139 12% 15% 18% 21% 12% 7% 11% 1311 11 (4.23) Mikeypoo 49 2148 (25th) 343 11% 14% 16% 17% 16% 14% 9% 1250 23 (2.41) Vohaul -10 2009 (236th) 202 13% 15% 11% 11% 9% 16% 20% 1226 9 (4.72) wishbone 0 2159 (20th) 92 21% 14% 18% 4% 11% 16% 13% 1176 13 (3.25) wowimbad -20 2080 (104th) 125 22% 12% 9% 13% 18% 13% 10% 1134 21 (2.47) jfdis 0 2152 (23rd) 55 27% 12% 12% 12% 12% 7% 14% 919 10 (4.30) petomni 0 2094 (88th) 140 15% 20% 16% 16% 12% 9% 10% 900 12 (3.32) uukrul 95 2166 (17th) 196 18% 11% 13% 16% 14% 15% 9% 784 20 (2.56) SarahPlaying 0 2076 (109th) 130 19% 10% 15% 18% 14% 10% 11% 771 19 (2.62) Kehoe -17 2072 (115th) 116 18% 13% 12% 9% 13% 20% 11% 762 24 (2.34) Vermont -12 2076 (109th) 95 14% 18% 15% 15% 17% 10% 6% 726 22 (2.45) Optical 0 2117 (60th) 79 12% 18% 21% 11% 13% 12% 8% 672 4 (7.52) Rsquared 0 2062 (131st) 129 19% 17% 11% 13% 9% 15% 13% 654 17 (2.68) rabbidmonkey4 0 2123 (53rd) 75 17% 17% 17% 17% 12% 13% 5% 654 14 (3.10) Harry Tuttle 0 2134 (37th) 52 23% 15% 15% 17% 9% 11% 7% 650 18 (2.64) joex 23 2062 (131st) 105 14% 13% 20% 13% 13% 9% 16% 510 16 (2.70) clomax 0 2106 (74th) 118 16% 14% 11% 18% 10% 11% 16% 489 15 (2.82) Sweden 34 2117 (60th) 102 10% 14% 17% 16% 16% 10% 12% 386 3 (8.45) riser * 0 2201 (9th) 18 14% 14% 14% 14% 11% 11% 14% 23 |
|
skrumgaer wrote
at 9:14 AM, Friday May 11, 2007 EDT rnd: to make you feel better, the lifetime TAPL for your profile in which you have the 11 games in May is 5417.
|
|
rnd wrote
at 10:18 AM, Friday May 11, 2007 EDT skrum, you can't just pick and choose like that. first of all, using "5%" is a complete misread of the yates correction.
second; when using a normalization technique like yates, you have to apply it to the entire dataset; you can't skew one or two because you feel the data 'looks wrong'; thats not math, its magic. |