View Full Version : TB rating system
Olykiller
2-07-2013, 14:13
Hi all
In this topic I’d like to point out some things about the TB rating system. It has been from summer 2010 since I rated, as you all know at that moment Splatterladder was the rating system. So this week I started to play some maps serious again and I noticed a remarkable aspect of the TB rating system.
Take for instance this map
http://et.trackbase.net/session/368/
As you can see dtec had a -0.33 on this map while he had almost 22ppm on FA2 which is good. The reason is because the average ppm is 12,19 which is rly high.
Other example:
http://et.trackbase.net/session/92160/
had a 100-22 score that map. Still got minus, why? Average ppm too high (9,59)
The counter example:
http://et.trackbase.net/session/61486/
huge + with only 16,45 ppm, reason here: low average ppm (6,16)
my point is:
play vs less skilled players and you get much +
play vs better players (more raters playing same map) and everyone gets – because average ppm is high.
In my opinion it should be the other way around; play good vs other raters and get much +, play vs less skilled players and get less +
(this was the case with SL. I still remember when we played with 3 or 4 .GoD| players on same map so average was rly high and everyone gets much +)
Some stuff to think about for next tb update :)
Grtz
Oly
Hi all
In this topic I’d like to point out some things about the TB rating system. It has been from summer 2010 since I rated, as you all know at that moment Splatterladder was the rating system. So this week I started to play some maps serious again and I noticed a remarkable aspect of the TB rating system.
Take for instance this map
http://et.trackbase.net/session/368/
As you can see dtec had a -0.33 on this map while he had almost 22ppm on FA2 which is good. The reason is because the average ppm is 12,19 which is rly high.
Other example:
http://et.trackbase.net/session/92160/
had a 100-22 score that map. Still got minus, why? Average ppm too high (9,59)
The counter example:
http://et.trackbase.net/session/61486/
huge + with only 16,45 ppm, reason here: low average ppm (6,16)
my point is:
play vs less skilled players and you get much +
play vs better players (more raters playing same map) and everyone gets – because average ppm is high.
In my opinion it should be the other way around; play good vs other raters and get much +, play vs less skilled players and get less +
(this was the case with SL. I still remember when we played with 3 or 4 .GoD| players on same map so average was rly high and everyone gets much +)
Some stuff to think about for next tb update :)
Grtz
Oly
So first to the last sentence: You said that good players vs good players would get - rating, but how could this session be made then? http://et.trackbase.net/session/61486/ 3 players with ~ 20 rating did play and get a + rating.
So now to the other two sessions.
1) http://et.trackbase.net/session/368/
dtect got a minus rating of course because the av ppm was higher than normal, especially on fa#2 but also because he (with ~28rating points) did have the same ppm like a player with only 14 rating points.. As well there were pretty much guys with ~15 rating who did also pretty good ppms.
2)http://et.trackbase.net/session/92160/
You did probably get minus rating because you had only 4 ppm more than a player which hasn't even got 10 rating points.
Of course the k/d ratio is important but you need to take into account which stats other players have had and how much their rating is.
For Example:
Player 1 rating 20 20ppm
Player 2 rating 4 16ppm
Average rating 8 14ppm
Player one would get -rating, player 2 +rating and the other players would need to be sorted between and above/below these two players.
We don't want to be like SL for sure and imo it would make players with a high rating getting even more rating. And it's not really like that, it depends on ppm and way more stuff, not only about rating.
But take always into account the following things:
Your rating
Your ppm
Your k/d ratio
the rating of the second best player and his rating
Average rating / ppm
If you have got more questions about this or other users want to initiate their opionion, go ahead and post it below:)
(we're always open for sth new and feedback!)
Olykiller
2-07-2013, 16:40
So first to the last sentence: You said that good players vs good players would get - rating, but how could this session be made then? http://et.trackbase.net/session/61486/ 3 players with ~ 20 rating did play and get a + rating.
the average ppm is only 6 which means there were a lot of noobs on the server
So now to the other two sessions.
1) http://et.trackbase.net/session/368/
dtect got a minus rating of course because the av ppm was higher than normal, especially on fa#2 but also because he (with ~28rating points) did have the same ppm like a player with only 14 rating points.. As well there were pretty much guys with ~15 rating who did also pretty good ppms.
i see
2)http://et.trackbase.net/session/92160/
You did probably get minus rating because you had only 4 ppm more than a player which hasn't even got 10 rating points.
Of course the k/d ratio is important but you need to take into account which stats other players have had and how much their rating is.
hm i understand but i go TB to check how much + i had cuz i did good map only to find out i got - :p
the rating of the second best player and his rating
also: a good player could use a random name with low rating and do good ppm so the raters get - :) this is the way to get nr1 :D
thx for infos Jonny :)
Your k/d ratio
the rating of the second best player and his rating
These actually matter for rating? Thought it was only about rating and ppm compared to other peoples rating and ppm.
About the session, yea the avg ppm is the reason I got minus but might also be affected by the fact that TB was temporarily down during that time and it got back up during the map, the map was 30min but TB only tracked 19mins (just saying don't take it as whine or complaining).
I think the top/top percentage (10-15%?) of players on a server should get at least + if not a neutral 0 (how much based on average rating, ppms etc - or some clever way :p). Someone could be top on e.g sKy-e and be say 1.5x ppm of average compared to the same person going on FA1 and getting 2-3x average ppm and they'd get more rate on FA and may even lose rate on sKy-e. It seems if I get a little over 2x average PPM I will always get a +, REGARDLESS of anything else. Someone shouldn't lose rating being top killer/xp gain on say NBS because they only got 1.4x average ppm...
These actually matter for rating? Thought it was only about rating and ppm compared to other peoples rating and ppm.
About the session, yea the avg ppm is the reason I got minus but might also be affected by the fact that TB was temporarily down during that time and it got back up during the map, the map was 30min but TB only tracked 19mins (just saying don't take it as whine or complaining).
It's not the way we're doing it but it's a good compare. Always check out the rating / ppm of the second best player, that's how i do it and get less deaths than time got minutes + (depends on rating) get a multiplier of the time in kills
We're very sorry about the downtime we had a few days ago, but it concerned every player playing in this time, so we can not consider a single player.
I think the top/top percentage (10-15%?) of players on a server should get at least + if not a neutral 0 (how much based on average rating, ppms etc - or some clever way :p). Someone could be top on e.g sKy-e and be say 1.5x ppm of average compared to the same person going on FA1 and getting 2-3x average ppm and they'd get more rate on FA and may even lose rate on sKy-e. It seems if I get a little over 2x average PPM I will always get a +, REGARDLESS of anything else. Someone shouldn't lose rating being top killer/xp gain on say NBS because they only got 1.4x average ppm...
It depends on average rating, might be that sky-e has more visitors with a high rating in the time you rated than fa#2 got, which is currently always full of players with lower rating (that's the reason why raters visit fa#2).
Olykiller
2-07-2013, 22:48
I think the top/top percentage (10-15%?) of players on a server should get at least + if not a neutral 0 (how much based on average rating, ppms etc - or some clever way :p). Someone could be top on e.g sKy-e and be say 1.5x ppm of average compared to the same person going on FA1 and getting 2-3x average ppm and they'd get more rate on FA and may even lose rate on sKy-e. It seems if I get a little over 2x average PPM I will always get a +, REGARDLESS of anything else. Someone shouldn't lose rating being top killer/xp gain on say NBS because they only got 1.4x average ppm...
its a public secret that FA#2 is the best rating server ;)
These actually matter for rating? Thought it was only about rating and ppm compared to other peoples rating and ppm.
About the session, yea the avg ppm is the reason I got minus but might also be affected by the fact that TB was temporarily down during that time and it got back up during the map, the map was 30min but TB only tracked 19mins (just saying don't take it as whine or complaining).
#1 TB spoken. /lock thread
http://wiki.trackbase.net/wiki/Wolfenstein_Enemy_Territory
The rating system is nothing more than a little bit more advanced ELO System, this means to get exactly +0.00 you are expected to do an x amount better than others based on your rating.
In the following example we take the following averages and expected results:
Average rating: 10, your rating: 20
Average ppm: 15, your ppm: 20
You must get at least (your rating / average rating ) * average ppm, in this case that would mean (20/10)*15 = 30 PPM. If you get lower than this average (in this case 20) you will get minus, if you would have got more than 30 ppm you would have gotten a positive rating for the round.
It's simple as that.
For the normal xp: this is just the xp you get in game
For the points: it's a combination of revives/kills/deaths/accuracy and some others things - I'll try to look them up.
Powered by vBulletin® Version 4.2.5 Copyright © 2024 vBulletin Solutions Inc. All rights reserved.