|
|
Post by kukurukukuk on Sept 1, 2017 4:28:05 GMT -5
 Ok, so yesterday I started putting this table together, mainly to track my performance. I also thought it would be interesting to see how the MM pits players together, because why not? We all know people face champions starting from diamond, but it is interesting to see how this manifests itself from match to match. The blue/red columns represent players and letters represent leagues (i.e. C=Champion, M2=Master II etc). I reside in Expert II. Some general observations: Teams seem to be adequatly balanced. That said, a blue team with two champions will generally NOT face a red team with two champions; it will be some other combination of leagues. 23 matches in, blue team consisted of 1,65 Champions per match and red team 2,00 Champions per match. My matches so far consisted of: 47% Expert, 30% Champion, 14% Diamond, 5% Master and the rest Gold and Silver, mainly due to squading. It is also interesting that the MM picks Master players 3 times less than Diamond players. That's it. Do you find this even slightly informative? Would you care for any other stats I could deduct from this data, or do you have any other suggestions/observations you would like to make? |
|
|
|
Post by TravLar on Sept 1, 2017 5:11:01 GMT -5
Yeah, this is really interesting. I'm a fan of the MM (jeez - did I just say that out loud?) and am fascinated to see the patterns and what it does. Thanks and pls continue sir!
|
|
|
|
Post by T34 on Sept 1, 2017 5:13:06 GMT -5
Well done, real hard evidence of the MM behaviour
|
|
|
|
Post by Diablo86 on Sept 1, 2017 5:45:30 GMT -5
Ok, so yesterday I started putting this table together, mainly to track my performance. I also thought it would be interesting to see how the MM pits players together, because why not? We all know people face champions starting from diamond, but it is interesting to see how this manifests itself from match to match. The blue/red columns represent players and letters represent leagues (i.e. C=Champion, M2=Master II etc). I reside in Expert II. Some general observations: Teams seem to be adequatly balanced. That said, a blue team with two champions will generally NOT face a red team with two champions; it will be some other combination of leagues. 23 matches in, blue team consisted of 1,65 Champions per match and red team 2,00 Champions per match. My matches so far consisted of: 47% Expert, 30% Champion, 14% Diamond, 5% Master and the rest Gold and Silver, mainly due to squading. It is also interesting that the MM picks Master players 3 times less than Diamond players. That's it. Do you find this even slightly informative? Would you care for any other stats I could deduct from this data, or do you have any other suggestions/observations you would like to make? Well Done! Thank you. Did you get to look at the Win % and Average damages of Blues and Reds? If yes, can we see a comparison on Average Win % across the Blue and Red teams? Same for Average Damage. I always experienced that while playing solo, the MM puts players in such a way that the average of league rating or Win % or Average damage is more or less same for both the teams. For Example, if the Average Win % of all the Blues is (say) 54%, then the average for Reds will almost always be the same (54% for this case) Thank you again. |
|
|
|
Post by Danny Linguini on Sept 1, 2017 7:12:46 GMT -5
Very interesting. Seeing the hard numbers, you can see how close your perception is to reality. Are all those bad things we think are happening really happening as frequently as we think? Nothing like an actual recorded measurement to tell the story.
|
|
beaconator2000
Destrier
Posts: 102
Karma: 141
Pilot name: Beaconator2000
Platform: iOS
Clan: Sexy
League: Diamond
Server Region: North America
|
Post by beaconator2000 on Sept 1, 2017 8:35:12 GMT -5
Very interesting. Thanks for sharing. Matches my perception in iOS Expert 3.
Not sure why such a small pool of Master players.
|
|
|
|
Post by ♧SGT FURY 24/7♧ on Sept 1, 2017 8:58:16 GMT -5
I'm blown away by the fact that in 23 games there was only 1 no show/zero. I've had 3 (blue, of course) in 2 games this morning.
|
|
inspirace
Trusted Contributor 
Posts: 2,670
Karma: 2,959
Pilot name: inspirace
Platform: iOS
Clan: NEW
League: Champion
Server Region: North America
Favorite robot: Rog, Griff, Carn, Doc, Rhino, Haechi, Kumiho, Bulgasari, Mender, Inquisitor, Spectre, Strider,
|
Post by inspirace on Sept 1, 2017 9:05:33 GMT -5
thanks!
I would like to see whether there is a different behavior based on win %.
but you seem to have quite stable in win rate. not much streaks of wins or losses,
|
|
|
|
Post by kukurukukuk on Sept 1, 2017 10:34:36 GMT -5
Wow, thank you all for the feedback! Well Done! Thank you. Did you get to look at the Win % and Average damages of Blues and Reds? If yes, can we see a comparison on Average Win % across the Blue and Red teams? Same for Average Damage. I always experienced that while playing solo, the MM puts players in such a way that the average of league rating or Win % or Average damage is more or less same for both the teams. For Example, if the Average Win % of all the Blues is (say) 54%, then the average for Reds will almost always be the same (54% for this case) Thank you again. No I haven't, but I can start writing that down. One of these factors should be decisive, though I doubt it will be damage related. I'm blown away by the fact that in 23 games there was only 1 no show/zero. I've had 3 (blue, of course) in 2 games this morning. Actually n/a was for a player whose league wasn't determined yet, doing the 5 qualification rounds. I haven't really been taking notes on leavers/tankers, but I remember their number being extremelly small, for me at least. Something like 4 or 5 in all 23 games. Also, the silver and gold ones that appear, where clearly seasoned players with over 3000 wins each. thanks! I would like to see whether there is a different behavior based on win %. but you seem to have quite stable in win rate. not much streaks of wins or losses, Yes it was a pretty steady run, somewhere near 50%. It will be noted down from now on. |
|
Deleted
Deleted Member
Posts: 0
Karma:
|
Post by Deleted on Sept 1, 2017 12:58:16 GMT -5
Thank you, it's really interesting! I would like to see something similar for iOS.
I'm in Expert 2 too and I've always wondered why I rarely have Masters in my battles. I usually face a lot of Champions and Experts, but Masters are really rare. Maybe they are a small number, who knows.
|
|
|
|
Post by 「fluffernutter」eney on Sept 12, 2017 10:39:52 GMT -5
Thanks for documenting your match experiences; I love data analysis. It appears as though your experiences mirror my own.
I believe that the MM system is completely unfair and needs a complete overhaul. In what world does it make sense for a Diamond/Gold level player to be playing with Masters/Champions?
I've heard people complaining about the size of the player pools for Masters/Champion level players. I have an idea... start allowing people to level up. Genius! I know.
When a Diamond player is teamed up with or are playing against players that are well out of their class and the team wins they get a minimal bump with the league points. Conversely, when they lose they lose tons of league points. A Diamond player will, more times than not, always be towards the bottom of the team ranking, that is unless they get lucky. How will they ever be able to tier up when they seem to be hemorrhaging league points? How will they ever be able to level up with their meager silver earnings? It seems to me that this is by design.
I suggest that a hangar strength and a tiering algorithm/calculation be implemented. If you are in Diamond, your pool should consist of players in the Expert and Gold tiers (1 tier up and down from you) that somewhat match your hangar strength. If a player has 5 lvl 12 bots and weapons, and they happen to belong in the Gold tier, they are obviously tanking and should be dealt with accordingly. If you can only put together teams of 4 in a timely manner, then start the game with 4 players a side.
BTW, I'm actually in Expert and was simply using the Diamond tier as an example. I also don't belong to a clan, so I have not been artificially elevated into a tier that I have no business being in. I have been in Expert for many months now and have not been able to progress up through the ranks due to the poor matchmaking. I start climbing the ranks, I get met with tons of Champions players and get knocked back down. Rinse and repeat. I can just about predict when I will start climbing and then fall back down. I feel like I have been treading water, even though I have strengthened my hangar by obtaining better bots/weapons and leveling them up.
Frustrating!
|
|
|
|
Post by hi5 on Sept 12, 2017 13:24:14 GMT -5
Thanks for documenting your match experiences; I love data analysis. It appears as though your experiences mirror my own. I believe that the MM system is completely unfair and needs a complete overhaul. In what world does it make sense for a Diamond/Gold level player to be playing with Masters/Champions? I've heard people complaining about the size of the player pools for Masters/Champion level players. I have an idea... start allowing people to level up. Genius! I know. When a Diamond player is teamed up with or are playing against players that are well out of their class and the team wins they get a minimal bump with the league points. Conversely, when they lose they lose tons of league points. A Diamond player will, more times than not, always be towards the bottom of the team ranking, that is unless they get lucky. How will they ever be able to tier up when they seem to be hemorrhaging league points? How will they ever be able to level up with their meager silver earnings? It seems to me that this is by design. I suggest that a hangar strength and a tiering algorithm/calculation be implemented. If you are in Diamond, your pool should consist of players in the Expert and Gold tiers (1 tier up and down from you) that somewhat match your hangar strength. If a player has 5 lvl 12 bots and weapons, and they happen to belong in the Gold tier, they are obviously tanking and should be dealt with accordingly. If you can only put together teams of 4 in a timely manner, then start the game with 4 players a side. BTW, I'm actually in Expert and was simply using the Diamond tier as an example. I also don't belong to a clan, so I have not been artificially elevated into a tier that I have no business being in. I have been in Expert for many months now and have not been able to progress up through the ranks due to the poor matchmaking. I start climbing the ranks, I get met with tons of Champions players and get knocked back down. Rinse and repeat. I can just about predict when I will start climbing and then fall back down. I feel like I have been treading water, even though I have strengthened my hangar by obtaining better bots/weapons and leveling them up. Frustrating! I'm with you on this. Keep seeing posts with a similar theme (exhibit a: war-robots-forum.freeforums.net/thread/14008/ios-3500-needing-help ). Wondering if there is another cliff that has formed. A few months ago I had the hardest time getting out of iOS D1 and firmly into E3, then sailed up to E2 and have been bounced back and forth for over a month. Could just be a struggle like this on the cusp of all higher leagues but feels different than D3 to D2 and then D2 to D1 - but I hear others say they move quickly through Masters. |
|
|
|
Post by Jame-thon on Sept 12, 2017 16:01:29 GMT -5
Great job collecting and sharing these statistics. Always want to thank people go out of the way to share like this. I'd definitely be interested to see other research you end up collecting.
|
|
