|
On November 01 2010 10:26 Rea wrote: according to ur formula someone with more points, including bonuspool, has a higher mmr... thought its well known since early beta that this is not the case
just go through any matchhistory and compare the players where one got +12 or -12p before bonuspool and compare their ladderpoints, they wont always be similar
This. The fact that a strict ordering is maintained when the reality is nowhere near strict means there is something missing. Hard to make an mmr calculator without taking into account opponents and other things. If anything, an estimator based solely on visible variables (opponent histories not being visible as far as I know) would probably have to have win-loss differential as a variable.
|
On November 01 2010 13:15 Excalibur_Z wrote:Show nested quote +On November 01 2010 10:40 Medzo wrote: I just want to point out that your hidden rating isnt determined by your shown rating at all. If you look at the top 200 a lot of those people are not in the top 200 on scranks (shown rating ranks). It is because their hidden ratings are higher than other people with higher point ratings. Basically what it seems like to me is you are ranked in your division by your hidden rating and the system will correct itself and give you the correct amount of shown rating points so that this is accurate.
I have 2 reasons for this hypothesis. First, In my own division pylon kappa last week everyone rank 9 and above was in the blizzard hidden rating top 200 in the same order of their ranking. The 9th ranked person was 1965 points which is about rank 400 in shown rating. My second reason is just from playing at a ladder and profiling my opponents when im favored or unfavored. If you have a high hidden rating you will fight other players who you are favored over who have a higher shown rating, they are usually rank 1 or 2 in their division with about 2k points but have hidden ratings similar or lower than 1600-1800 in harder divisions. At the same time you can play another player with lower or equal shown and hidden rating in equally hard or harder (medivac alamo) divisions.
This is easily noticeable when you have a very high hidden rating. While I don't necessarily think it's possible to determine MMR from displayed rating concretely, I think your reasoning is wrong here. The Top 200 is a list of players by points with division influences removed, it's not by hidden rating. That is confirmed by Blizzard. The formula posted by the OP is also incorrect because it considers points even across all divisions, which we know from Blizzard not to be the case. We also don't know whether there is a division hierarchy (where Medivac Alamo > Medic Mu > Observer Mars for example) or just different tiers of divisions (where Medivac Alamo, Medic Mu, Observer Mars are Tier 1, Augustgrad X-Ray and Scout Kilo are Tier 2, etc). For example, the first week of the Top 200 had mostly Medivac Alamo, Medic Mu, Observer Mars, and Akilae Omicron members listed. The third week added a few new divisions popping up in the list: Immortal Yankee, Infestor Pepper, Pylon Kappa. If you compared their rankings in the Top 200 with some old SC2Ranks snapshots taken within a day of the Top 200 announcement, most of the scores lined up. However, there were some outliers. Dayvie is the most prominent. He had more points than most other players but was below them on the Top 200 list, and that's because his division Scout Kilo is not weighted the same as the original "top" divisions.
As far as i am aware you don't switch between divisions ever unless you are promoted or demoted (correct me if i'm wrong). There are other things to consider. The top 200 might reflect not the highest rated, but could be a combination of rating and recent win% or biggest gains. Also you might consider that the top 200 could be 1-2 weeks old before they post it.
|
On November 01 2010 13:15 Excalibur_Z wrote:Show nested quote +On November 01 2010 10:40 Medzo wrote: I just want to point out that your hidden rating isnt determined by your shown rating at all. If you look at the top 200 a lot of those people are not in the top 200 on scranks (shown rating ranks). It is because their hidden ratings are higher than other people with higher point ratings. Basically what it seems like to me is you are ranked in your division by your hidden rating and the system will correct itself and give you the correct amount of shown rating points so that this is accurate.
I have 2 reasons for this hypothesis. First, In my own division pylon kappa last week everyone rank 9 and above was in the blizzard hidden rating top 200 in the same order of their ranking. The 9th ranked person was 1965 points which is about rank 400 in shown rating. My second reason is just from playing at a ladder and profiling my opponents when im favored or unfavored. If you have a high hidden rating you will fight other players who you are favored over who have a higher shown rating, they are usually rank 1 or 2 in their division with about 2k points but have hidden ratings similar or lower than 1600-1800 in harder divisions. At the same time you can play another player with lower or equal shown and hidden rating in equally hard or harder (medivac alamo) divisions.
This is easily noticeable when you have a very high hidden rating. While I don't necessarily think it's possible to determine MMR from displayed rating concretely, I think your reasoning is wrong here. The Top 200 is a list of players by points with division influences removed, it's not by hidden rating. That is confirmed by Blizzard. The formula posted by the OP is also incorrect because it considers points even across all divisions, which we know from Blizzard not to be the case. We also don't know whether there is a division hierarchy (where Medivac Alamo > Medic Mu > Observer Mars for example) or just different tiers of divisions (where Medivac Alamo, Medic Mu, Observer Mars are Tier 1, Augustgrad X-Ray and Scout Kilo are Tier 2, etc). For example, the first week of the Top 200 had mostly Medivac Alamo, Medic Mu, Observer Mars, and Akilae Omicron members listed. The third week added a few new divisions popping up in the list: Immortal Yankee, Infestor Pepper, Pylon Kappa. If you compared their rankings in the Top 200 with some old SC2Ranks snapshots taken within a day of the Top 200 announcement, most of the scores lined up. However, there were some outliers. Dayvie is the most prominent. He had more points than most other players but was below them on the Top 200 list, and that's because his division Scout Kilo is not weighted the same as the original "top" divisions.
I looked into this and something important to consider is that there is still inflation going on in MMR because of the newness of the game. As players getting better and noobs....well stay noobs, then the higher ranked players will be keep climbing. Although this is probably the most important point. I will look into. Part of me believes that the ladder is screwed up, possibly points not converging to MMR fast enough.
|
United States12204 Posts
On November 01 2010 13:21 darmousseh wrote:Show nested quote +On November 01 2010 13:15 Excalibur_Z wrote:On November 01 2010 10:40 Medzo wrote: I just want to point out that your hidden rating isnt determined by your shown rating at all. If you look at the top 200 a lot of those people are not in the top 200 on scranks (shown rating ranks). It is because their hidden ratings are higher than other people with higher point ratings. Basically what it seems like to me is you are ranked in your division by your hidden rating and the system will correct itself and give you the correct amount of shown rating points so that this is accurate.
I have 2 reasons for this hypothesis. First, In my own division pylon kappa last week everyone rank 9 and above was in the blizzard hidden rating top 200 in the same order of their ranking. The 9th ranked person was 1965 points which is about rank 400 in shown rating. My second reason is just from playing at a ladder and profiling my opponents when im favored or unfavored. If you have a high hidden rating you will fight other players who you are favored over who have a higher shown rating, they are usually rank 1 or 2 in their division with about 2k points but have hidden ratings similar or lower than 1600-1800 in harder divisions. At the same time you can play another player with lower or equal shown and hidden rating in equally hard or harder (medivac alamo) divisions.
This is easily noticeable when you have a very high hidden rating. While I don't necessarily think it's possible to determine MMR from displayed rating concretely, I think your reasoning is wrong here. The Top 200 is a list of players by points with division influences removed, it's not by hidden rating. That is confirmed by Blizzard. The formula posted by the OP is also incorrect because it considers points even across all divisions, which we know from Blizzard not to be the case. We also don't know whether there is a division hierarchy (where Medivac Alamo > Medic Mu > Observer Mars for example) or just different tiers of divisions (where Medivac Alamo, Medic Mu, Observer Mars are Tier 1, Augustgrad X-Ray and Scout Kilo are Tier 2, etc). For example, the first week of the Top 200 had mostly Medivac Alamo, Medic Mu, Observer Mars, and Akilae Omicron members listed. The third week added a few new divisions popping up in the list: Immortal Yankee, Infestor Pepper, Pylon Kappa. If you compared their rankings in the Top 200 with some old SC2Ranks snapshots taken within a day of the Top 200 announcement, most of the scores lined up. However, there were some outliers. Dayvie is the most prominent. He had more points than most other players but was below them on the Top 200 list, and that's because his division Scout Kilo is not weighted the same as the original "top" divisions. As far as i am aware you don't switch between divisions ever unless you are promoted or demoted (correct me if i'm wrong). There are other things to consider. The top 200 might reflect not the highest rated, but could be a combination of rating and recent win% or biggest gains. Also you might consider that the top 200 could be 1-2 weeks old before they post it.
The Top 200 is points only with division weighting factored out, we have that quoted. The rankings are taken each Monday morning hours before they are posted, though to be fair that could mean up to 3 or 4 hours of posting/processing time. 3-4 hours isn't going to be enough to change the point values by that much, though, especially considering not all of these players are playing during that translation period.
|
I just checked this with a few of my friend's profiles... and it's pretty blatantly wrong. The highest rated ones were, without exception, those who hadn't played much in the last month.
|
Holy crap, I am in love with this thread
|
On November 01 2010 14:20 Obscura.304 wrote: I just checked this with a few of my friend's profiles... and it's pretty blatantly wrong. The highest rated ones were, without exception, those who hadn't played much in the last month.
One, did you test this on his site, two, did you use the entire equation (including that jazz about bonus pool)
|
i with they would just show us our mmr instead of making us guess. stupid system. at least in wow u could see it
|
Does it get Bliz's top 200 right? What percent error? Knowing this I think we can know how well this formula works/ how close it gets to Bliz's ranking that they hide.
|
Well, that certainly answers that. At least if you're going to try and push people so you can make a quick buck, come up with a little more inventive of a formula :/
|
On November 01 2010 14:29 Slaytesics wrote: One, did you test this on his site, two, did you use the entire equation (including that jazz about bonus pool) There is no point to including the "jazz" about bonus pool in the equation. It literally makes no difference to the final sum.
|
It's impossible for this number to be remotely accurate because of two simple reasons:
1) MMR is your 'skill', but it's volatile which is why we don't see it, instead we get a magic number called points, since we already used magic poney points, it's now called magic pink kitty points or MPKP for short. 2) The amount of MPKP you gain or lose is based on your 'real' points (NOT the battle.net or division points) compared to your opponents MMR every time a match ends
You play two games, and you've already made it impossible to go from MPKP -> MMR because the data you have (total wins, total losses, current points) are only overall stats, not specific ones. In order to reverse MPKP -> MMR, you need to have statistics on every single persons points and league.
Even then, I don't think you could do it. You would likely need to only play against people you could repeat the same process for and had it all done in a bubble.
And even then, then, you don't have any way of verifying the number as being the true MMR.
[edit] Removed current theory part
|
Good job! Will see how accurate this system is in the future (after tweaking and all)
|
according to this i'm top 100 us mmr, but on sc2ranks.com i'm not even top 1000, and my mmr doesn't match me regularly against anyone over 2000 points. feels kind of inaccurate.
|
according to the main page i am in the top 100 MMR in the world but im merely a 1500 NA diamond player... what gives? flopflop.779
|
On November 01 2010 16:30 flopflop wrote: according to the main page i am in the top 100 MMR in the world but im merely a 1500 NA diamond player... what gives? flopflop.779
The website is just used so that you can get your rating, it doesn't list all of the players.
|
On November 01 2010 16:52 darmousseh wrote:Show nested quote +On November 01 2010 16:30 flopflop wrote: according to the main page i am in the top 100 MMR in the world but im merely a 1500 NA diamond player... what gives? flopflop.779 The website is just used so that you can get your rating, it doesn't list all of the players.
Haha thats good, I had like a ~3400 hundred rating and saw idra at ~3700 and was like hmmmmm. I am no where near his skill level.
Very cool though OP, best of luck tweaking this into perfection. I do think the ladder system is kinda weird when trying to determine skill. It is more a relative type tool than an exact one. So trying to make something exact(true MMR) with something is that isn't so exact(ladder rating) is definitely a challenge haha.
|
On November 01 2010 17:13 Paperscraps wrote:Show nested quote +On November 01 2010 16:52 darmousseh wrote:On November 01 2010 16:30 flopflop wrote: according to the main page i am in the top 100 MMR in the world but im merely a 1500 NA diamond player... what gives? flopflop.779 The website is just used so that you can get your rating, it doesn't list all of the players. Haha thats good, I had like a ~3400 hundred rating and saw idra at ~3700 and was like hmmmmm. I am no where near his skill level. Very cool though OP, best of luck tweaking this into perfection. I do think the ladder system is kinda weird when trying to determine skill. It is more a relative type tool than an exact one. So trying to make something exact(true MMR) with something is that isn't so exact(ladder rating) is definitely a challenge haha.
IIRC Idra's only played 100+ games on his account on US and is like 100-20
|
I am still not following... do I not show up on the main page of http://sc2stars.com/ for you guys? this makes no sense to me.
edit: I figured it out. It adds names to the list as people search for them. Thanks guys.
|
Just a request: Could you hide the friend code in your resultset?
|
|
|
|