The 36 "most predominant" players
  • 76 posts
  • Page 2 of 6
The_Bishop wrote:
So, here it is. As I said I call the old parameter "absolute predominance" (or alpha-predominance) and the new one "balanced predominance" (or beta-predominance).

The only difference is that instead of considering the number of games played in the last month / session, here I have considered the square root of that value. This is equivalent to a geometric mean where the "rating" parameter weighs twice compared to the "last month games" parameter. Well, these are the results (except for possible errors).

--  Top 36  beta-predominant  players  --
  June 19th, 2020

Position---  Rank  &  username  ---Rating --- Total games -Last month games (sqrt) ---α-Predominance ---β-Predominance
...
342nd Lieutenant @PT2716830448 sqrt = (21.166)1216.76857.487
5Lieutenant Colonel @Joeyseven49951321110 sqrt = (10.488)549.45052.388
90Serg Major @allthingsmustpass204329135519 sqrt = (22.782)1060.31746.544
332nd Lieutenant @Texx272816481278 sqrt = (16.673)758.38445.484
7Major General @Deepdaleduck4719453979 = (8,888)372.80141.942
31Private @1949midden2784488219 sqrt = (14.799)609.69641.200
...
98Sergeant Major @HalSucks2009965417 sqrt = (20.421)837.75341.026
78Warrant Officer @ElectricBlue214425365354 sqrt = (18.815)758.97640.339
10Lieutenant Colonel @simomosi39874538102 sqrt = (10.100)406.67440.268
9Colonel @huskers014294340081 sqrt = (9.000)347.81438.646
151Command Serg @2BeeStings1722634501 sqrt = (22.383)862.72238.544
8Brigadier Gen @ProblemChild96  4452260567 sqrt = (8.185)298.28436.440
...
23Captain @Riskk32434870116 sqrt = (10,770)376.18834.927
11Lieutenant Colonel @maafi3880488369 sqrt = (8.307)267.72032.231
3Lieut General @hooboy1153261673236 sqrt = (6.000)191.73631.956
47Chief War Officer @Hoodlum240735672172 sqrt = (13.115)414.00431.568
84Sergeant Major @superDENI20805572221 sqrt = (14.866)459.68030.921
2General @slackbatter5765254027 sqrt = (5.196)155.65529.955
...
51Lance Corporal @McNulthy2366285159 sqrt = (12.610)376.19429.835
64Warrant Officer @vodkamikebid22488155170 sqrt = (13.038)382.16029.309
42Private @abuzees2564294118 sqrt = (10.863)302.55227.853
69Private @MrTrotsky22034553148 sqrt = (12.166)326.04426.802
282nd Lieutenant @roy123453036130275 sqrt = (8.660)227.70026.292
113Private @Andies19347404184 sqrt = (13.565)355.85626.235
...
18Lieutenant @The_Bishop3437171957 sqrt = (7.550)195.90925.949
62Warrant Officer @wailuen842256910125 sqrt = (11.180)282.00025.222
81Sergeant Major @bluewabbit2109385138 sqrt = (11.747)291.04224.775
137Private @letitbe18021591189 sqrt = (13.748)340.57824.773
19Captain @KOE_KittyKat3434573352 sqrt = (7.211)178.56824.763
65Warrant Officer @DrJardi22411239122 sqrt = (11.045)273.40224.752
...
239Corporal @options1350822333 sqrt = (18.248)449.55024.635
108Sergeant Major @TomC19474358159 sqrt = (12.610)309.57324.552
15Major @Henris_136571061245 sqrt = (6.708)164.56524.531
24Captain @DurandalBlue3241267157 sqrt = (7.550)184.73724.470
40Chief War Officer @blackrabbit259474288 sqrt = (9.381)228.27224.334
128Sergeant Major @BigBrain9418602274171 sqrt = (13.077)318.06024.323
...
beta-Predominance, 37th to 100th (click to show)
«God doesn't play dice with the World» ~ Albert Einstein
The_Bishop wrote:
Clearly this ranking can change quite a lot from one month to the next depending on the number of games played. It can be considered a sort of "last month's performance". However, the data reported here are no longer updated since they refer to June 19th, but this was just an example.

What should this parameter represent? First of all it is a recognition towards users who play many games, they are those who make the site work and make it popular. If in addition to playing many games you are also good then easily you will be on this list. For a talented player, not much effort is required to appear in the list, just play 30 or 40 games a month. On the contrary, a player with a high rating who does not want to play much in order not to risk ruining his rating, will have difficulty entering this ranking. In short, those who dare to face the challenge every day are the most rewarded! :)
«God doesn't play dice with the World» ~ Albert Einstein
AlexCheckMate wrote:
Responded instantly on the message I got, then waited some before replying here. I'm still very intrigued by the list; I believe it shows exactly that what it's supposed to show; dominance, which could indeed be captured as a formula based on two variables; skill/rating and visibility/games played.

The higher your rating is and in the more games you've played (& thus got noticed by more people), the more you dominate.

I like that there's some more emphasis on rating rather than #of games and the new pre-fixes; alpha/absolute & beta/balanced - very clever.

At this point though, if I wanted to "dominate" this list, I could likely create a bunch of LT 2p games with people >X rating. It would keep my rating rather stable/safe, while adding extremely to my "visibility"-parameter. So as dough_boy pointed out, it'd be good if 2p games could be filtered out (I realise that's currently not possible in an easy way (you'd have to go check everyone's stats manually and deduct # of 2p games played between the updates)).

The assumed favouritism for live games is fine by me. I believe live games have a greater impact on visibility anyway.
PS: 36 is coooool :roll:
“Gravitation cannot be held responsible for people falling in love. How on earth can you explain in terms of chemistry and physics so important a biological phenomenon as first love? Put your hand on a stove for a minute and it seems like an hour. Sit with that special girl for an hour and it seems like a minute. That's relativity.”

― Albert Einstein
Virtuosity98 wrote:
I agree with Alex - if the list is designed to measure a combination of a player's ability and exposure/activity, then it works well.

I do have a couple of points of opinion on how the list could more accurately measure that combination of variables:
1. The way the list is measured inherently favours exposure over ability. The vast majority of players (97.4%) have a rating between 4000 and 500. So, the most skilled players have a predominance 8x higher than the least skilled players (ceteris paribus). And for 95% (within 2s.d. of mean) of players, the ratio quickly falls to only 6.38x.

By taking figures from the same percentiles we can see that being active is rewarded much more than being skilful. 97.4% of players played between 221 and 5 games. So, the most active players have a predominance 44.2x higher than the least active players (ceteris paribus). For 95% of players, the ratio is still highly overweighted at 34.2x.

Thus, being active has a far more significant impact on your predominance than being skilful. I think the intention was to have both variables influence the score equally, as implied by the formula predominance = rating x games played.

So-called beta-predominance attempted to rectify this by square-rooting the last month games variable:
97.4% of players played between 221 and 5 games, square rooted to 14.87 and 2.24, respectively. So, the most active players have a predominance 6.6x higher than the least active players (ceteris paribus). For 95% of players, the ratio is 5.8x.

So by taking a square root of the last month games, the difference between the relative weightings of the two variables has been reduced:
alpha predominance - 6.38:34.2
beta predominance - 6.38:5.8

So, even though choosing to use the square root was actually quite an arbitrary decision, it quite luckily turned out to be quite effective for equalising the weightings of the two variables :)

2. However, the two variables are still unequally weighted, now rating is actually over-represented in the beta predominance variable (6.38:5.8) by 10%. Also, since the time between D12 lists varies from month to month, so too will the disparity between rating and new games played. When the time between lists is longer, Playing more games will become more influencial on beta-predominance than rating, and vice versa. Thus, the measuring system is slightly inconsistent between lists.

Consider a ranking system where all players who qualify for the list are ranked in order of rating, and in order of last month games, then the two rankings are multiplied. If such a ranking system was used instead of absolute values, it would present two distinct advantages. Firstly, the weighting of the two variables (skill and exposure/activity) would always be weighted exactly equally, at 1:1, which is what we want if indeed we believe in the formula predominance = rating x last month games. Secondly, longer months would not distort the importance of the two variables.

3. I believe that, when talking about dominance, games last month is not as telling a variable as no. of players played with last month. If it were possible to collate the data where the number of players in each game was taken into account, then this measure would be more indicative of exposure. Clearly, someone who has played 10 9-player games has had more exposure than someone sho has played 10 2-player games. I realise however that collecting such data would be almost impossible.




Oof, I might need to edit this to make it more readable. What do you think of these points? Agree/disagree?
It is now Day 8. Please submit your Lynch vote, as well as any Role-specific Day actions you wish to perform (countdown).
Day Actions:
• #LYNCH [player], #NO LYNCH, #ABSTAIN in forum thread.
• Role-specific actions (via PM with V98).





The_Bishop wrote:
@AlexCheckMate, I am glad for your interest and appreciation!

@Virtuosity98 same, adding this: when things are simple let's make them more complex! :D kidding
Some answers in short (click to show)

@All
Head-to-head lovers are certainly advantaged in this list, while capped cards players are disadvantaged. In fact, Dima had something to complain about, he expressed it his own way but basically he is right. Nothing I can do though. As I said I would have liked better to use the turns taken instead of the games played, but I do not have a record of that data.
In my opinion the perfect parameter for measuring a player's activity fairly would be the sum of all alive opponents for each turn he took in the last session. However, it doesn't seem any feasible, so I play with the data that I have.

Plus, the rating too is not an effective measure of a player's skills because it has too strong fluctuations. For example, on the list of May 16th I had 2800 points, in that of June 19th I had 3400 .. Wow, have I improved so much in a month?! No, I haven't. Using a metaphor I can say that in the first case Maafi took a picture while I was sitting, in the second case he took a picture while I was jumping, but in reality I am always the same "tall".
It wouldn't be very useful to make one parameter perfectly representative of reality if we do not do the same with the other. We can say that both are approximations, but they are easily available data. Overall, it seems to me that the "beta-predominance" comes close to representing what I wanted, so I'm personally satisfied. :)

Anyway, I will draw up the list also in the coming months, so maybe we will have a broader view.


«God doesn't play dice with the World» ~ Albert Einstein
The_Bishop wrote:
--  Top 36  beta-predominant  players  --   July 24th, 2020

Placement. . .  ---  Rank  &  username  ---  . . .Rating --- Total games -Last session games (sqrt) ---β-Predominance  -count-
...
51Warrant Officer @2BeeStings23181136502 sqrt = (22.405356)51,936-1-
4Lieutenant General @Joeyseven5080142695 sqrt = (9.746794)49,514-2-
43Chief Warrant Officer @Hoodlum246736072400 sqrt = (20.000000)49,340-3-
69Unranked @clydesdale444210131097448 sqrt = (21.166010)44,470-4-
2General @slackbatter6078259050 sqrt = (7.071067)42,978-5-
11Dominator @ProblemChild9639172720115 sqrt = (10.723805)42,005-6-
...
94Sergeant Major @abutia19703475407 sqrt = (20.174241)39,743
36Chief Warrant Officer @Texx259716699218 sqrt = (14.764823)38,344-8-
99Sergeant Major @allthingsmustpass192729488353 sqrt = (18.788294)36,205
41Chief Warrant Officer @PT25081023193 sqrt = (13.892443)34,842-10-
20Lieutenant @TP_knighty31643782104 sqrt = (10.198039)32,267
1General @AlexCheckMate629366526 sqrt = (5.099019)32,088-12-
...
12Major @KOE_KittyKat3658580976 sqrt = (8.717797)31,890
103Sergeant Major @SmokinHog1897404282 sqrt = (16.792855)31,856
44Private @1949midden2431653165 sqrt = (12.845232)31,227-15-
38Chief Warrant Officer @Riskk25605011141 sqrt = (11.874342)30,398
64Private @letitbe21821763172 sqrt = (13.114877)28,617
139Command Sergeant @SN00717221055271 sqrt = (16.462077)28,348
...
77Private @Andies20607592188 sqrt = (13.711309)28,245
262nd Lieutenant @Blagoje_Jovovic300813587 sqrt = (9.327379)28,057-20-
39Chief Warrant Officer @phillberto25592283117 sqrt = (10.816653)27,680
167Master Sergeant @HalSucks15561277312 sqrt = (17.663521)27,484
19Lieutenant @simomosi3170461375 sqrt = (8.660254)27,453
90Sergeant Major @God_of_War2001295188 sqrt = (13.711309)27,436
...
111Private @Dominator31866569203 sqrt = (14.247806)26,586-25-
158Private @MrTrotsky16274813260 sqrt = (16.124515)26,235
30Chief Warrant Officer @TimTheGreat278626887 sqrt = (9.327379)25,986
10Lieutenant Colonel @maafi3967492239 sqrt = (6.244997)24,774
61Private @cdravis2222241117 sqrt = (10.816653)24,035
52Private @HannibalTunsi23152746101 sqrt = (10.049875)23,265-30-
...
219Sergeant @GeneralHalsey13573273279 sqrt = (16.703293)22,666
7Brigadier General @Deepdaleduck4665456223 sqrt = (4.795831)22,373
151Command Sergeant @Mdhaglund1655328178 sqrt = (13.341664)22,080
128Command Sergeant @genghis_burke1799181147 sqrt = (12.124355)21,812
157Master Sergeant @IsolatedBadger1628524179 sqrt = (13.379088)21,781-35-
322nd Lieutenant @roy123452683136563 sqrt = (7.937253)21,296
...
beta-Predominance, 37th to 100th (click to show)
«God doesn't play dice with the World» ~ Albert Einstein
Dima wrote:
what i dont like about the stat, is that it doesnt take into account the player amount of the finished games. having 7 finished 1vs1 games is like one finished 7p game. in both cases you have fought against 7 players, but in one case its one single game and in another its 7 games. therefore the predominance is in the second case much higher. it seems that these stats are supporting people who play many short + live games.


it would be much more interesting to have a statistics, that shows how many and how often you beat other players. something like "amount of defeated players"
"vorple: the real strategy comes when you cant just win cuz you got lucky and got the big card stack"
Virtuosity98 wrote:
I think The_Bishop updates it whenever a new d12 list is made. If he has time, that is! :D
It is now Day 8. Please submit your Lynch vote, as well as any Role-specific Day actions you wish to perform (countdown).
Day Actions:
• #LYNCH [player], #NO LYNCH, #ABSTAIN in forum thread.
• Role-specific actions (via PM with V98).





The_Bishop wrote:
I'm sorry for the delay.
I'll probably publish it tomorrow..
«God doesn't play dice with the World» ~ Albert Einstein
The_Bishop wrote:
Almost there... Calculations are done, but making the tab requires time. I had to figure out how to automatize it, but didn't.
Sorry, probably sunday or monday I'll publish it. I have to go out a couple of days.
I can reveal that @dimceto is first, followed by hooboy11, braveheart1, clydesdale444, allthingsmustpass, Joeyseven and... all the others.
«God doesn't play dice with the World» ~ Albert Einstein
The_Bishop wrote:
You guys are just lucky that I missed my train... So here we are.
Later than late, but still sooner than never! ;)

--  Top 36  beta-predominant  players  --   August 28th, 2020
an>
Ranking. . .  ---  Rank  &  username  ---  . . .Rating --- Total games -Last session games (sqrt) ---β-Predominance  -Count-
...
4Lieutenant General @dimceto51801159177 sqrt = (13.304134)68,915-1-
3Lieutenant General @hooboy11521116873130 sqrt = (11.401754)59,415-2-
45Chief Warrant Officer @braveheart12420597482 sqrt = (21.954498)53,130-3-
54Unranked @clydesdale444230631627530 sqrt = (23.021728)53,088-4-
332nd Lieutenant @allthingsmustpass264529859371 sqrt = (19.261360)50,946-5-
5Lieutenant General @Joeyseven5143150579 sqrt = (8.888194)45,712-6-
...
122Command Sergeant @lima1760605496 sqrt = (22.271057)39,197
322nd Lieutenant @Mdhaglund2687536208 sqrt = (14.422205)38,752-8-
14Major @1949midden3566770117 sqrt = (10.816653)38,572
21Captain @KOE_KittyKat32585921112 sqrt = (10.583005)34,479-10-
232nd Lieutenant @Blagoje_Jovovic2995259124 sqrt = (11.135528)33,351
10Lieutenant Colonel @maafi3881499573 sqrt = (8.544003)33,159-12-
...
12Major @B4rny3624163481 sqrt = (9.000000)32,616
118Com Sergeant @Gottalottabling1775610294 sqrt = (17.146428)30,435
58Private @Andies22577770178 sqrt = (13.341664)30,112-15-
49Chief Warrant Officer @God_of_War2360457162 sqrt = (12.727922)30,038
2General @slackbatter5831261626 sqrt = (5.099019)29,732
34Chief Warrant Officer @dwcalvert25932261130 sqrt = (11.401754)29,565
...
13Major @Henris_135971071767 sqrt = (8.185352)29,443
67Warrant Officer @Hoodlum218336251179 sqrt = (13.379088)29,207-20-
272nd Lieutenant @TomC27564558111 sqrt = (10.535653)29,036
89Private @cdravis1966459218 sqrt = (14.764823)29,028
37Chief Warrant Officer @Texx256416824125 sqrt = (11.180339)28,666
92Sergeant Major @elysium519558301200 sqrt = (14.142135)27,648
...
81Master Sergeant @lucky232012331183 sqrt = (13.527749)27,218-25-
1General @AlexCheckMate658168116 sqrt = (4.000000)26,324
18Private @Drac3382227459 sqrt = (7.681145)25,978
302nd Lieutenant @blackrabbit271888689 sqrt = (9.433981)25,642
9Lieutenant Colonel @OldDogGen3896353941 sqrt = (6.403124)24,947
78Master Sergeant @Barbaro2078437144 sqrt = (12.000000)24,936-30-
...
19Captain @ProblemChild96 3353277454 sqrt = (7.348469)24,639
136Command Sergeant @Newb191671433215 sqrt = (14.662878)24,502
98Command Sergeant @ggg1231925626160 sqrt = (12.649110)24,350
53Private @Feej2308 920111 sqrt = (10.535653)24,316
216Sergeant @nperry541321802320 sqrt = (17.888543)23,631-35-
48Chief Warrant Officer @TimTheGreat237336698 sqrt = (9.899494)23,492
...
beta-Predominance, 37th to 100th (click to show)
«God doesn't play dice with the World» ~ Albert Einstein