Laviak
Joined: Jul 19, 2004
|
 Posted:
Nov 17, 2008 - 09:27 |
 
|
Irgy: I was just thinking the same approach might work 
Ideally, you want to play with even teams against evenly rated coaches, but, if there aren't any evenly rated coaches, it's better to play against an evenly rated team than to try to make an adjustment to the effective team strength based on coaching ability.
Mathematically, I was thinking to model it as a pair of gaussian curves (http://en.wikipedia.org/wiki/Gaussian_function), centered at 0. You can scale them such that you end up with something like
TS diff of 0 = 90 points
TS diff of 20 = 45 points
BR difference of 0 = 10 points
BR difference of 10 = 5 points
Add them together and voila, suitability on a scale of 0 to 100... obviously the parameters would need some thought (much more than I have put in) to set effective limits on TS difference, and to set a minimum suitability factor to make a pairing sufficiently suitable.
I'm sure there are massive flaws, but it's a start, and maybe it will get some minds ticking away |
_________________
We Fink Wer Orks
--------
Help save blood bowl, foul an elf today!.
Last edited by Laviak on %b %17, %2008 - %14:%Nov; edited 1 time in total |
|
Nextflux
Joined: Jan 22, 2008
|
| Posted:
Nov 17, 2008 - 10:20 |
|
BR and BashR now works (as I understand it): Touchdown Diff is the significant factor, while Cas diff is a factor, but you need 3.33 Cas Diff to weight up for 1 TD Diff..
correct?
All in all not a bad idea.. But I suggest that BR (and BashR) (since it follows the coach), only applies to the current _active_ teams the coach have.
This to make it less attractive to create teams to lower BR and retire the team..
If it only apply to current active teams of the coach, it also probably makes it more rewarding to keep those crappy teams (and the coach have to play their crappy teams sometimes(since it is one of their active teams) to keep that BR low.
Most likely this also gives more diversity of races. |
|
|
Gritter
Joined: Jun 29, 2006
|
| Posted:
Nov 17, 2008 - 10:30 |
|
I think BR should be removed from the calculations completely.
Apart from the valid reasons already raised for this, it will also cut down on the variety of different coaches you will play if you are scheduled against people of similar ranking- there are a number of different reasons this is bad. |
|
|
koadah
Joined: Mar 30, 2005
|
| Posted:
Nov 17, 2008 - 11:04 |
|
| Mr_Foulscumm wrote: | | koadah wrote: | | Tinkywinky wrote: | | Nextflux wrote: | | conceding should be illegal in BBox, especially if we have BR |
Why? Assuming that BR isnt allowed to influence matchups it's punishment enough to lose BR? What is the difference from conceeding in ranked? |
He seems to have conceeded before a single turn was played. |
Who conceded before a single turn was played?
If you are talking about the DorfvsZon fiasco then you are very much mistaken. |
OK. Apologies.
What happened? Why are there no stats? |
_________________
O[L]C 2016 Swiss! - April ---- All Star Bowl - Teams of Stars - 2 more teams needed |
|
Gnomes FTW! (Replays...
Search
