|
On June 20 2012 20:42 skeldark wrote:Show nested quote +On June 20 2012 20:39 korona wrote:On June 20 2012 19:48 skeldark wrote:On June 20 2012 18:28 korona wrote: GM league for season 8 has now been opened on all servers except Kr. Hopefully there is enough data collected through the tool to determine what were the lowest GM season entry MMRs for each server. if all 200 gm player use my tool... Even if, what to do with this information? I think its not very important whats the 200th highest mmr at week 1 at session 8. it change all the time. I just take from every server a gm player out of the database if i find on and take his mmr as promotion line. if hes very high i put it a little bit lower It is just interesting data as likely the easiest way to get in GM is to play & win just when the GM opens and have high enough MMR moving average. There are much more players near the promotion barrier when 200 spots are open than in midseason when only couple of spots are open and the barrier is much higher. This data would also be interesting to those high master players who are wondering if they are close to the promotion barrier in the beginning of a new season. Regarding seasons: It would be nice to have a line separating each season in the MMR graph as it is hard to remember the exact point when the season changed by just looking at the graph. actually its other way round. in midseasson there are several hundred master players who have higher mmr than half of gm. And that means the entry barrier in midseason is most likely higher than in the beginning of the season. If there is one spot open in the midseason in GM, it is 'reserved' for the active player who has the highest MMR moving average at that point of time and is not in GM? One thing for sure it is much harder to get in GM in midseason than in the beginning when 'fast' players who have just 'borderline' mmr moving average can get in.
|
On June 20 2012 20:57 korona wrote:Show nested quote +On June 20 2012 20:42 skeldark wrote:On June 20 2012 20:39 korona wrote:On June 20 2012 19:48 skeldark wrote:On June 20 2012 18:28 korona wrote: GM league for season 8 has now been opened on all servers except Kr. Hopefully there is enough data collected through the tool to determine what were the lowest GM season entry MMRs for each server. if all 200 gm player use my tool... Even if, what to do with this information? I think its not very important whats the 200th highest mmr at week 1 at session 8. it change all the time. I just take from every server a gm player out of the database if i find on and take his mmr as promotion line. if hes very high i put it a little bit lower It is just interesting data as likely the easiest way to get in GM is to play & win just when the GM opens and have high enough MMR moving average. There are much more players near the promotion barrier when 200 spots are open than in midseason when only couple of spots are open and the barrier is much higher. This data would also be interesting to those high master players who are wondering if they are close to the promotion barrier in the beginning of a new season. Regarding seasons: It would be nice to have a line separating each season in the MMR graph as it is hard to remember the exact point when the season changed by just looking at the graph. actually its other way round. in midseasson there are several hundred master players who have higher mmr than half of gm. And that means the entry barrier in midseason is much higher than in the beginning of the season. If there is one spot open in the midseason in GM, it is 'reserved' for the active player who has the highest MMR moving average at that point of time and is not in GM? Exact. and as long as he did not play or is consider as inactive the gm spot stay open. The hole gm system is poorly designed.
Wait a second.... We can find out how long you need to get inactive by this!
|
On June 20 2012 20:39 skeldark wrote:Show nested quote +On June 20 2012 20:22 fapy wrote: After updating my MMR is randomly changing by 3 in between some games, was this intended?
eg 1886 1903 (win) 1900 (no game played) 1916 (win) i dont understand. you did not play a game but he shows you a game? and this with 3 less mmr? can you give me a picture?
MMR: http://i.imgur.com/WnLfz.png Last 10 games: http://i.imgur.com/Kp7TE.png
Starting at 98 on the graph I win first 5 games between 101 and 103 it adds extra game where i lose 3 MMR
This is F analyser
|
On June 20 2012 21:02 fapy wrote:Show nested quote +On June 20 2012 20:39 skeldark wrote:On June 20 2012 20:22 fapy wrote: After updating my MMR is randomly changing by 3 in between some games, was this intended?
eg 1886 1903 (win) 1900 (no game played) 1916 (win) i dont understand. you did not play a game but he shows you a game? and this with 3 less mmr? can you give me a picture? MMR: http://i.imgur.com/WnLfz.pngLast 10 games: http://i.imgur.com/Kp7TE.pngStarting at 98 on the graph I win first 5 games between 101 and 103 it adds extra game where i lose 3 MMR I see i see. I dont know where the game comes from but the diffrence of 3 shows how inaccurate i am 3! can you send me your games.dat file? or just pm the last 10 lines out of it?
|
On June 20 2012 21:06 skeldark wrote:Show nested quote +On June 20 2012 21:02 fapy wrote:On June 20 2012 20:39 skeldark wrote:On June 20 2012 20:22 fapy wrote: After updating my MMR is randomly changing by 3 in between some games, was this intended?
eg 1886 1903 (win) 1900 (no game played) 1916 (win) i dont understand. you did not play a game but he shows you a game? and this with 3 less mmr? can you give me a picture? MMR: http://i.imgur.com/WnLfz.pngLast 10 games: http://i.imgur.com/Kp7TE.pngStarting at 98 on the graph I win first 5 games between 101 and 103 it adds extra game where i lose 3 MMR I see i see. I dont know where the game comes from but the diffrence of 3 shows how inaccurate i am  3! can you send me your games.dat file? or just pm the last 10 lines out of it?
Where is this file?
|
On June 20 2012 21:08 fapy wrote:Show nested quote +On June 20 2012 21:06 skeldark wrote:On June 20 2012 21:02 fapy wrote:On June 20 2012 20:39 skeldark wrote:On June 20 2012 20:22 fapy wrote: After updating my MMR is randomly changing by 3 in between some games, was this intended?
eg 1886 1903 (win) 1900 (no game played) 1916 (win) i dont understand. you did not play a game but he shows you a game? and this with 3 less mmr? can you give me a picture? MMR: http://i.imgur.com/WnLfz.pngLast 10 games: http://i.imgur.com/Kp7TE.pngStarting at 98 on the graph I win first 5 games between 101 and 103 it adds extra game where i lose 3 MMR I see i see. I dont know where the game comes from but the diffrence of 3 shows how inaccurate i am  3! can you send me your games.dat file? or just pm the last 10 lines out of it? Where is this file? Sc2gears\User Content\Plugin file cache\mmr.plugin.MMRPlugin
|
On June 20 2012 21:01 skeldark wrote:Show nested quote +On June 20 2012 20:57 korona wrote:On June 20 2012 20:42 skeldark wrote:On June 20 2012 20:39 korona wrote:On June 20 2012 19:48 skeldark wrote:On June 20 2012 18:28 korona wrote: GM league for season 8 has now been opened on all servers except Kr. Hopefully there is enough data collected through the tool to determine what were the lowest GM season entry MMRs for each server. if all 200 gm player use my tool... Even if, what to do with this information? I think its not very important whats the 200th highest mmr at week 1 at session 8. it change all the time. I just take from every server a gm player out of the database if i find on and take his mmr as promotion line. if hes very high i put it a little bit lower It is just interesting data as likely the easiest way to get in GM is to play & win just when the GM opens and have high enough MMR moving average. There are much more players near the promotion barrier when 200 spots are open than in midseason when only couple of spots are open and the barrier is much higher. This data would also be interesting to those high master players who are wondering if they are close to the promotion barrier in the beginning of a new season. Regarding seasons: It would be nice to have a line separating each season in the MMR graph as it is hard to remember the exact point when the season changed by just looking at the graph. actually its other way round. in midseasson there are several hundred master players who have higher mmr than half of gm. And that means the entry barrier in midseason is much higher than in the beginning of the season. If there is one spot open in the midseason in GM, it is 'reserved' for the active player who has the highest MMR moving average at that point of time and is not in GM? Exact. and as long as he did not play or is consider as inactive the gm spot stay open. The hole gm system is poorly designed. Wait a second.... We can find out how long you need to get inactive by this! Agreed. The GM system has its flaws. People can maximize their MMR/MMR moving average and stop playing and wait. And then try to get in GM just when it opens by winning one match. MMR of these 'maximiers' will likely fall after this to their 'normal level', which also explains why there are lot of GM:s in midseason who have lower MMR than hundreds of high master players at that point of time. This is also why I was interested in the entry barrier just when the GM opens, as at that moment it is the 'easiest' to get in GM even if you are not a typical GM-caliber player.
The 'active' player regarding the demotion from GM is a player who has less than 180 unused bonus points. Not sure what the 'activity limit' regarding the league entry is. It would be logical that 'activity limit' regarding league entry is more strict as only ~90 bonus points has been accumulated when GM opens one week after the start of new season. If it would be the same '180 unused bonus points', then people could get in GM just by playing 2 matches if they have high enough MMR moving average.
|
On June 20 2012 21:37 korona wrote:Show nested quote +On June 20 2012 21:01 skeldark wrote:On June 20 2012 20:57 korona wrote:On June 20 2012 20:42 skeldark wrote:On June 20 2012 20:39 korona wrote:On June 20 2012 19:48 skeldark wrote:On June 20 2012 18:28 korona wrote: GM league for season 8 has now been opened on all servers except Kr. Hopefully there is enough data collected through the tool to determine what were the lowest GM season entry MMRs for each server. if all 200 gm player use my tool... Even if, what to do with this information? I think its not very important whats the 200th highest mmr at week 1 at session 8. it change all the time. I just take from every server a gm player out of the database if i find on and take his mmr as promotion line. if hes very high i put it a little bit lower It is just interesting data as likely the easiest way to get in GM is to play & win just when the GM opens and have high enough MMR moving average. There are much more players near the promotion barrier when 200 spots are open than in midseason when only couple of spots are open and the barrier is much higher. This data would also be interesting to those high master players who are wondering if they are close to the promotion barrier in the beginning of a new season. Regarding seasons: It would be nice to have a line separating each season in the MMR graph as it is hard to remember the exact point when the season changed by just looking at the graph. actually its other way round. in midseasson there are several hundred master players who have higher mmr than half of gm. And that means the entry barrier in midseason is much higher than in the beginning of the season. If there is one spot open in the midseason in GM, it is 'reserved' for the active player who has the highest MMR moving average at that point of time and is not in GM? Exact. and as long as he did not play or is consider as inactive the gm spot stay open. The hole gm system is poorly designed. Wait a second.... We can find out how long you need to get inactive by this! Agreed. The GM system has its flaws. People can maximize their MMR/MMR moving average and stop playing and wait. And then try to get in GM just when it opens by winning one match. MMR of these 'maximiers' will likely fall after this to their 'normal level', which also explains why there are lot of GM:s in midseason who have lower MMR than hundreds of high master players at that point of time. This is also why I was interested in the entry barrier just when the GM opens, as at that moment it is the 'easiest' to get in GM even if you are not a typical GM-caliber player. The 'active' player regarding the demotion from GM is a player who has less than 180 unused bonus points. Not sure what the 'activity limit' regarding the league entry is. It would be logical that 'activity limit' regarding league entry is more strict as only ~90 bonus points has been accumulated when GM opens one week after the start of new season. If it would be the same '180 unused bonus points', then people could get in GM just by playing 2 matches if they have high enough MMR moving average.
I know that with this tool and the trick you describe its way easier to get into gm. Or not easier but more luck depended. ( if you have a winningstreak in the first week) But dont forget that there are only few players that are able to push them self so high. So being gm dont mean top 200 but it still means top on the server.
![[image loading]](http://i.imgur.com/nAqrl.png)
You have to be close to the orange mark and like you can see, there is not much population left.
The active bonuspool rule for dropping out of gm is not the active server rule.
A logic system would be:
(Spend_bonus +X) smaller total_bonus = inactive
with x = bonuspool you can get in Y days. Would count for a season start and filter everyone who is not playing much. Or even easyer
inactive = no game played in Y days.
|
United States12224 Posts
Any thoughts about releasing a website that aggregates all of these user histories, or adding it to SC2Ranks or something? I think we could be reaching a point where we could create a browsable unified ladder for users of this tool while simultaneously expanding the number of users.
I know you have a webserver that already hosts this data, I meant making it publicly viewable, by region, etc.
|
On June 21 2012 01:58 Excalibur_Z wrote: Any thoughts about releasing a website that aggregates all of these user histories, or adding it to SC2Ranks or something? I think we could be reaching a point where we could create a browsable unified ladder for users of this tool while simultaneously expanding the number of users.
I know you have a webserver that already hosts this data, I meant making it publicly viewable, by region, etc. at the moment the webserver only have the plain data. The caclulation and analyse part is only on the client. I have to put the plain server data into my debug version and this one analyse user by user when i watch him. I could write a tool that analyse all users but for useable website i have to put it into a database and than prog a web frontend...
-- btw you are mod so perhaps you can help me, I can not connect to the ts server whatever i do. It dont even awnser my packet `(i sniffed it) EDIT problem solved hisself. Strange
|
For some reason the option to sort by e and e+ (and now F i assume) is no longer an option after I updated. The menu at the bottom is now the Display drop down menu, moving average, opponents, dots. Did you I do something that prevents this from sorting now? Thanks for any assistance and again thanks for this amazing program!
|
On June 21 2012 03:52 Beren wrote:For some reason the option to sort by e and e+ (and now F i assume) is no longer an option after I updated. The menu at the bottom is now the Display drop down menu, moving average, opponents, dots. Did you I do something that prevents this from sorting now? Thanks for any assistance and again thanks for this amazing program!  sorry i dont understand. Why sorting?
e e+ f are different algorithm to analyse the data. I removed the old ones.
|
On June 21 2012 04:22 skeldark wrote:Show nested quote +On June 21 2012 03:52 Beren wrote:For some reason the option to sort by e and e+ (and now F i assume) is no longer an option after I updated. The menu at the bottom is now the Display drop down menu, moving average, opponents, dots. Did you I do something that prevents this from sorting now? Thanks for any assistance and again thanks for this amazing program!  sorry i dont understand. Why sorting? e e+ f are different algorithm to analyse the data. I removed the old ones.
Oh ok so F is just default now? There use to be a choice and its no longer there so I wasn't sure if it was done intentionally.
|
you can still see the old e+ if you enable debug options in the settings. But in all cases where e and f disagreed it was f who was right. So i decided to remove e , e- was shit anyway. I kept him in, so people see something when there is not enough data for e+. It turns out this confused the players more than it helped.
|
Is there anyway that you will be able to make the tool work when your opponent has 0 ladder points? I've had a good amount of those and it always gives an error and skips that game and then the mmr gets messed up.
|
Woah this is really cool =)
|
United States12224 Posts
On June 21 2012 09:00 EPICWINZ wrote: Is there anyway that you will be able to make the tool work when your opponent has 0 ladder points? I've had a good amount of those and it always gives an error and skips that game and then the mmr gets messed up.
Probably not, because your opponent doesn't lose any points if he already has 0, and without that value you can't determine your own MMR. If he lost -1 point then you'd know your MMR is way above his points, if he lost -23 then you'd know it was way below his points. If he's at 0 then he doesn't lose anything, which creates a bad data point.
|
On June 21 2012 09:39 Excalibur_Z wrote:Show nested quote +On June 21 2012 09:00 EPICWINZ wrote: Is there anyway that you will be able to make the tool work when your opponent has 0 ladder points? I've had a good amount of those and it always gives an error and skips that game and then the mmr gets messed up.
Probably not, because your opponent doesn't lose any points if he already has 0, and without that value you can't determine your own MMR. If he lost -1 point then you'd know your MMR is way above his points, if he lost -23 then you'd know it was way below his points. If he's at 0 then he doesn't lose anything, which creates a bad data point.
Yeah but you could use the amount of non-bonus pool points the player gained to figure out how many points they would have lost if they did not have zero points.
Wouldn't it just be 24- (amount of non-bonus points I gained) ?
|
My MMR plugin for some reason doesn't show the graph. It keeps saying that the player quality is 75%. How do I get it to 100%?
|
On June 21 2012 10:22 EPICWINZ wrote:Show nested quote +On June 21 2012 09:39 Excalibur_Z wrote:On June 21 2012 09:00 EPICWINZ wrote: Is there anyway that you will be able to make the tool work when your opponent has 0 ladder points? I've had a good amount of those and it always gives an error and skips that game and then the mmr gets messed up.
Probably not, because your opponent doesn't lose any points if he already has 0, and without that value you can't determine your own MMR. If he lost -1 point then you'd know your MMR is way above his points, if he lost -23 then you'd know it was way below his points. If he's at 0 then he doesn't lose anything, which creates a bad data point. Yeah but you could use the amount of non-bonus pool points the player gained to figure out how many points they would have lost if they did not have zero points. Wouldn't it just be 24- (amount of non-bonus points I gained) ? No. How much he lost have nothing to do with how much you win. This to values are total independent from each other. If he have After the game 0 points, i can not tell how much he would have lost and there is no information in the system about it.
the new versions should not skip the game if he can not collect data and the graph should not be messed up because of this. can you show a picture what you mean?
On June 21 2012 09:02 haegN wrote: Woah this is really cool =) Thank you 
On June 21 2012 14:25 HexSCII wrote: My MMR plugin for some reason doesn't show the graph. It keeps saying that the player quality is 75%. How do I get it to 100%? Not in your hand. It depends on the skill diffrence between you and your opponent and his league and tier compare to yours.
|
|
|
|