|
United States12280 Posts
Also, I don't know if you're working backwards on every entry, but because SC2Ranks' match history can be delayed at times, it's usually necessary to use the official Battle.net match history. Here are the scores at the time of the snapshot for the players that have missing data:
26 Mankeyz 329-248 2674 Turaxis Yankee
32 MnMEdward 609-514 2655 Medivac Juliet
35 sixjaxNewsh 158-93 2641 Victory Bravo
36 PiQLiQ 1164-943 2639 Shuttle Nu
38 Kyhol 554-426 2638 Observer Mars
47 CellaWerra 380-231 2591 Raven Zed
54 iKillBanshee 303-240 2559 Overmind Victor
59 SPeCiaLiST 432-346 2555 Phash Kappa
61 Slurgi 463-381 2554 Talematros Eta
63 toddman 636-524 2546 Medivac Alamo
69 BrokenRhythm 881-722 2537 Medic Pepper
70 LeeHyoRi 421-327 2536 Hartley Quest
83 lovemyRM 496-436 2513 Nagglfar Zed
92 KANAKO 461-358 2805 Valerian Uncle
94 sKsYeN 536-441 2735 Arcturus Mars
94 DarkCell 245-185 2483 Zergling Quest
94 LiquidTyler 203-107 2483 Zekrath Sigma
97 Yakero 155-48 2482 Hartley Quest
98 Bamboocha 250-178 2480 Kelerchian X-Ray
99 Spades 627-508 2668 Halcyon November
100 LiquidRet 128-43 2477 Phash Kappa
100 Humble 249-189 2477 Turaxis Yankee
100 RaNGirLxD 360-289 2477 Kelerchian X-Ray
103 NeverDawn 207-139 2476 Tassadar Charlie
104 FlyingSheep 330-248 2601 Infestor Pepper
105 owmygroin 997-901 2470 Medic Mu
106 Mura 492-432 2469 Medic Mu
106 Ngry 219-146 2469 Akilae Omicron
106 Fenneth 354-290 2469 Dominion Beta
109 Ryze 967-864 2657 Hartley Nu
110 REBUKED 733-558 2529 Argo Echo
111 scvrusher 265-199 2714 Urun Charlie
112 Destiny 780-696 2458 Observer Mars
113 BigMac 190-141 2457 Horner Chi
113 LiquidNazgul 179-112 2457 Medivac Juliet
115 EnvY 658-548 2455 Observer Mars
116 stev 387-322 2454 Horner Chi
117 DalaiLameR 111-76 2453 Uraj Dixie
118 SoFtBalL 135-71 2452 Uraj Dixie
119 dayvie 486-365 2766 Scout Kilo
119 sLeepisT 402-328 2451 Immortal Yankee
121 Vtmisery 395-320 2447 Observer Mars
122 Zelniq 1034-818 2444 Medivac Alamo
123 KawaiiRice 215-106 2439 Medic Mu
124 Xog 789-704 2564 Feld Lima
125 ACERGAMEGMMA 1534-1410 2746 Mohandar Romeo
126 nGenJazBas 117-67 2429 Overmind Victor
126 Taurent 345-197 2429 Akilae Omicron
126 Toosneaky 580-497 2429 Akilae Omicron
129 fpx 138-90 2427 Shuttle Nu
130 deth 202-155 2423 Overmind Victor
130 Antimage 382-298 2423 Medic Mu
132 TaurosPKM 433-365 2547 Feld Lima
132 lime 296-233 2421 Observer Mars
134 LegendofBert 552-474 2420 Medic Mu
135 RuFF 936-804 2732 Augustgrad X-Ray
136 Android 592-472 2415 Medivac Alamo
137 Kalu 155-99 2475 Rimes Tau
137 makoplux 595-538 2412 Uraj Omega
137 Blade 330-267 2412 Tassadar Charlie
140 CombatEX 765-679 2724 Goliath Beta
141 mystik 409-351 2408 Akilae Omicron
141 NKULUNKULU 378-322 2408 Turaxis Yankee
143 InflowCoppah 553-448 2407 Medivac Alamo
144 RaNgeD 739-654 2655 Augustgrad Psi
144 Thugnificent 994-882 2466 Torrasque Eta
146 FireZerg 492-405 2402 Medic Mu
146 Science 144-106 2402 Nagglfar Zed
146 Arena 231-186 2402 Victory Bravo
146 spekkiO 136-88 2402 Overmind Victor
146 VTWhiplash 341-286 2402 Observer Mars
151 mkengyn 450-386 2401 Zergling Quest
151 MsButtloaves 160-112 2401 Uraj Dixie
151 mTwDIMAGA 84-45 2401 Medivac Juliet
151 nGenMaFia 137-94 2401 Hartley Quest
155 Polar 386-328 2400 Akilae Omicron
155 simonlegosu 493-428 2400 Observer Mars
157 ReasoN 174-106 2398 Raven Zed
158 CubonePKM 387-315 2396 Medic Mu
158 Ikki 393-335 2396 Akilae Omicron
158 BoO 530-444 2396 Uraj Omega
166 GoGoGo 100-59 2386 Overmind Victor
167 FeaR 800-724 2446 Argo Echo
167 Ocean 592-522 2383 Pylon Kappa
169 giX 276-208 2382 Immortal Yankee
169 Time 982-877 2382 Medivac Alamo
171 AhhBoxxah 151-84 2379 Talematros Eta
171 Lost 433-371 2379 Medic Pepper
173 cVen 381-320 2378 Medic Mu
174 stogez 209-132 2692 Medic Whiskey
175 jungik 241-192 2376 Moratun Sierra
175 PhiliBiRD 202-147 2376 Pylon Kappa
175 ROOTFayth 305-227 2502 Infestor Pepper
178 John 281-224 2374 Nagglfar Zed
178 Andro 682-601 2374 Observer Mars
180 NOVAstalife 407-266 2372 Medivac Alamo
180 Ganon 214-160 2498 Void Ray Tango
182 magnolia 550-506 2371 Zergling Quest
183 VTGoddeR 588-512 2622 Void Ray Tau
183 Malice 695-527 2370 Medic Mu
183 Aoratha 419-358 2370 Uraj Omega
186 Shoey 478-417 2369 Immortal Yankee
187 Pookie 464-397 2368 Immortal Yankee
188 Hawk 357-285 2682 Raven Chi
189 GplaysNet 113-79 2366 Shuttle Nu
190 MattDamon 664-578 2364 Observer Mars
190 CNBgrtmn 194-139 2364 Talematros Eta
192 Nostra 871-769 2487 Spectre Zed
192 RedAlert 336-265 2361 Lenassa Dixie
194 ZekoXPn 470-399 2674 Taarim Gamma
195 ButteryLlama 641-544 2358 Medic Mu
196 SeaofStorms 182-130 2357 Talematros Eta
197 GRITmadison 1119-1035 2417 Argo Echo
197 Jecho 352-280 2606 Raven Victor
199 ClosingTimE 308-254 2479 Infestor Pepper
200 Kvz 445-385 2352 Pylon Kappa
I'll be updating this list as I see more unbolded entries.
|
I'm not using bnet match history yet, being too lazy  . Not exactly sure how to find it. edit: nvm LOL can just click names in top 200. I'll get to it after I do all the sc2ranks ones.
I'd prefer you use true unadjusted (bnet displayed) scores, as that's what I'm listing, but I can convert it for myself.
|
United States12280 Posts
On November 23 2010 02:38 Pokebunny wrote:I'm not using bnet match history yet, being too lazy . Not exactly sure how to find it. edit: nvm LOL can just click names in top 200. I'll get to it after I do all the sc2ranks ones. I'd prefer you use true unadjusted (bnet displayed) scores, as that's what I'm listing, but I can convert it for myself.
Er sorry, I'm not using adjusted scores, I'm just correcting the scores to what they were at the time of the snapshot :> I'll clarify the post.
|
Getting bored of doing this T_T, I'll finish later today or you can go ahead if you want to. I think at this point we have enough data to completely confirm all theories, but I suppose we should check the scores near the people who's divisions you don't have confirmed data on.
|
Impressive, Excalibur.
Considering that the theory is pretty much confirmed now, can we do the same thing for the EU and KR servers?
|
United States12280 Posts
On November 23 2010 02:55 Pokebunny wrote:
Getting bored of doing this T_T, I'll finish later today or you can go ahead if you want to. I think at this point we have enough data to completely confirm all theories, but I suppose we should check the scores near the people who's divisions you don't have confirmed data on.
I can finish. I figured you stopped since I didn't see any updates =]
|
How do you do it faster than me while getting the division names at the same time >.<
|
interesting information. thats about all it is tho for now
|
I think my favorite part is how simple this is.
Blizz Top 200 Quote.
"In the meantime, we can pull some internal numbers, do a little math, and come up with a list of the best players in each region."
I like how math means (points + division offset).
So basically their top 200 is just a list of the 200 players with the most points, not highest MMR which is kinda sad.
I just want to know my relative MMR so that I can track improvement (instead of thinking that I am improving)
|
What interests me about this system is how it deals with Diamond players who improve or lose skill once they are already in a Diamond division. Imagine a player learning the game does just enough to get put into Diamond but goes into a very low Diamond division but then over time improves a lot but due to no movement between divisions of the same tier, will they not be stuck getting a lower overall rank according to Blizzard?
|
On November 23 2010 03:21 Cade)Flayer wrote:
What interests me about this system is how it deals with Diamond players who improve or lose skill once they are already in a Diamond division. Imagine a player learning the game does just enough to get put into Diamond but goes into a very low Diamond division but then over time improves a lot but due to no movement between divisions of the same tier, will they not be stuck getting a lower overall rank according to Blizzard?
Example of this: InflowMini. His displayed rating is over 3000, but he was placed in a lower-tier division. The only thing that will happen is your displayed rating will be far higher than it should be, and you'll appear lower on the top 200 than you expect. So they'll get a lower overall rating from blizzard than what they think it should be, but in reality, they'll be playing lower rated players from other divisions.
|
So if I understand correctly, for example, if inFlowMini 2,736 (3,051) plays against TTOne 2,902 (2,902) then loading screen should say that opponent is slight favored (or favored. TTOne being favored one) ?
|
Thats interesting and actually makes sense, maybe its possible to get same statistics for eu server as well?
|
On November 23 2010 03:33 Too_MuchZerg wrote:
So if I understand correctly, for example, if inFlowMini 2,736 (3,051) plays against TTOne 2,902 (2,902) then loading screen should say that opponent is slight favored (or favored. TTOne being favored one) ?
first of all its always different for both players.
for each player it is based on their own points relative to their opponents match making rating and they win and lose points based on this. if their opponents match making rating is same as their actual rating its around 12 points per win/loss which is the middle point in the scale. note that each player usually sees something different on their own screen.
example :
player 1 is rated 2500 MMR rating 2500
player 2 is rated 1400 MMR rating 2600 (mb super good player on a new account)
for player 1 it shows : opponent is slightly favored (2600 - 2500 = 100 difference)
for player 2 it shows : opponent is favored (2500-1400 = 1100 difference)
also for the actual ratings it should be based on excalibur's division adjusted rating.
the favored scale is only to show yourself how many points you will win or lose after the game.
if both players are the player 2 in the example fighting each other they should both see the other player as favored and they will probably gain 30 points per win and lose only 1-2 points if they lose themselves, so that way if they both go 50 percent vs each other they should slowly climb towards their MMR rating 2600 which is also why you can climb to your MMR rating going 50 percent as intended.
|
On November 23 2010 03:33 Too_MuchZerg wrote:
So if I understand correctly, for example, if inFlowMini 2,736 (3,051) plays against TTOne 2,902 (2,902) then loading screen should say that opponent is slight favored (or favored. TTOne being favored one) ?
Yes. StrifeCro's post is also correct.
On November 23 2010 03:37 kYem wrote:
Thats interesting and actually makes sense, maybe its possible to get same statistics for eu server as well?
Yes, if someone wants to do the work..
|
United States12280 Posts
Okay done. Editing the first post with the complete list.
|
I wonder if it's somehow possible to get the ratings of Diamond divisions below ones featuring a top 200 player. While it would not be that exciting as a general strata of data it would be a nice way of seeing where the ladder sees you as an individual for those average Diamond players.
|
On November 18 2010 18:12 Wargizmo wrote:Show nested quote +On November 18 2010 17:38 Paperscraps wrote:On November 18 2010 17:33 paralleluniverse wrote:
What is the rationale for the existence of a division modifier so that points are not comparable across divisions?
Why not just make 2000 points in division A equal to 2000 points in division B?
Why not just use the top 200 methodology everywhere? Great question. I believe most people are wondering the same thing. Your almost NEVER see a guy with 100 games ranked higher than someone with 200 games in the same division. Why? because if the guy with 100 games was significantly better he would be put in a higher tier of division to begin with.
What does it mean when this does happen? http://us.battle.net/sc2/en/profile/270391/1/Soki/ladder/1797#current-rank I always thought it was weird how I had so many fewer games played than most people around me...
|
United States12280 Posts
Okay the first post has been updated with this week's rankings, their divisions, and the division tiers have been updated as well.
Corrections to previous divisions:
Raven Chi is +315 (E), not +252 (D)
New divisions this week:
S-Rank +0
Zekrath Sigma (confirmed from last week)
Zergling Quest (confirmed from last week)
Dominion Beta
A-Rank +63
Rimes Tau
Torrasque Eta
B-Rank +126
Void Ray Tango
Raven Victor
C-Rank +189
Halcyon November
Hartley Nu
D-Rank +252
Arcturus Mars (confirmed from last week)
Void Ray Tau
E-Rank +315
Mohandar Romeo (confirmed from last week)
Raven Chi (confirmed from last week)
Taarim Gamma
|
Awesome work man, really appreciate it. Didn't find this posted anywhere in the thread, how well did the predictions agree with Blizzard's top 200? You sounded pretty happy so was it pretty much spot on for all the divisions you had information for?
|
|
|
|
|
|
EPT
