Season 3 ranking update

Everything dealing with the video game developed by Cyanide!
User avatar
dode74
Posts: 6949
Joined: 11 December 2008, 11:18
Location: Nr. Reading, UK
Contact:

Re: Season 3 ranking update

Postby dode74 » 06 March 2017, 17:29

1-0-0 = 43.20
5-2-3 = 40.74
Image

Retrocausality
Posts: 15
Joined: 25 July 2016, 09:35

Re: Season 3 ranking update

Postby Retrocausality » 06 March 2017, 21:10

dode74 wrote:1-0-0 = 43.20
5-2-3 = 40.74


Oh OK! Thanks!

Also, WEIRD! :? :lol:

User avatar
Niessuh
Posts: 70
Joined: 27 November 2015, 10:01

Re: Season 3 ranking update

Postby Niessuh » 07 March 2017, 09:11

Retrocausality wrote:
dode74 wrote:1-0-0 = 43.20
5-2-3 = 40.74


Oh OK! Thanks!

Also, WEIRD! :? :lol:


This is the reason why most people restarts teams after having a single early loss, it seems heavily punished for losing one single game

User avatar
dode74
Posts: 6949
Joined: 11 December 2008, 11:18
Location: Nr. Reading, UK
Contact:

Re: Season 3 ranking update

Postby dode74 » 07 March 2017, 09:55

Niessuh wrote:This is the reason why most people restarts teams after having a single early loss, it seems heavily punished for losing one single game
It's not, though. 5-3-2 would be above 1-0-0.
Image

Retrocausality
Posts: 15
Joined: 25 July 2016, 09:35

Re: Season 3 ranking update

Postby Retrocausality » 07 March 2017, 19:47

dode74 wrote:
Niessuh wrote:This is the reason why most people restarts teams after having a single early loss, it seems heavily punished for losing one single game
It's not, though. 5-3-2 would be above 1-0-0.


Seems a strange place to draw the line, to me.

I could understand 1-0-0 being above 0-0-10 or even 1-0-9 but I don't understand why 1-0-0 is considered "better" than even 1-1-8 let alone 5-2-3 to be honest.

User avatar
dode74
Posts: 6949
Joined: 11 December 2008, 11:18
Location: Nr. Reading, UK
Contact:

Re: Season 3 ranking update

Postby dode74 » 07 March 2017, 19:52

Where you choose to draw the line is very much subjective.
Image

Retrocausality
Posts: 15
Joined: 25 July 2016, 09:35

Re: Season 3 ranking update

Postby Retrocausality » 07 March 2017, 21:06

dode74 wrote:Where you choose to draw the line is very much subjective.


Very true, very true.

I'd be interested in hearing what the thinking behind it was, though. Presumably there's a reason why they did the formula this way; a deliberate decision to make brand new teams with only one win rank higher than older teams with more wins.

Miraskadu
Posts: 228
Joined: 08 March 2016, 03:05

Re: Season 3 ranking update

Postby Miraskadu » 07 March 2017, 21:47

Retrocausality wrote:
dode74 wrote:Where you choose to draw the line is very much subjective.


Very true, very true.

I'd be interested in hearing what the thinking behind it was, though. Presumably there's a reason why they did the formula this way; a deliberate decision to make brand new teams with only one win rank higher than older teams with more wins.



Pretty much any ranking that considers number of games played will be rather bad at judging the order at low numbers of games played.

And if you based it pure win% 1-0-0 is a 100% win rate. So it would be 'better' than 100-0-1 or 100-1-0. So how ever high you factor the number of games played in, cahnges the record which is the first one that is better than the 1-0-0 record.

User avatar
VoodooMike
Posts: 1587
Joined: 14 July 2009, 07:44
Location: Calgary, Alberta
Contact:

Re: Season 3 ranking update

Postby VoodooMike » 07 March 2017, 22:19

Retrocausality wrote:I'd be interested in hearing what the thinking behind it was, though. Presumably there's a reason why they did the formula this way; a deliberate decision to make brand new teams with only one win rank higher than older teams with more wins.

If its the formula I think it is, the basic idea is that it uses a scaling portion of your win% - scaled by the number of games you've played.

Lets say it starts at 50% and scales up until the point we feel is an "acceptable" number of games played to really rank you on... say... 20. That means a 1-0-0 record is ranked as though the person had a win% of 50%...

But it also means if your win% is mediocre and your games played is below that threshold, you may rank below that. 5-2-3 is 66% win rate, and only 9 games played. If the relationship was linear (which it isn't, but should be, but that's another topic altogether!) and we used the sample numbers from above, that'd mean your ranking would be 72.5% of 66%, or the equivalent of 47.9%, which is lower than the 1-0-0's 50% of 100% which is 50%
Friendly Reminder: Correlation does not equal Causation - tattoo it on the inside of your eyelids if it'll help.


Return to “General”

Who is online

Users browsing this forum: No registered users and 1 guest