- 314
- Denmark
[Update: The points chart has been updated because of the new 100000 DR ceiling. Having watched Tidgney's and Kie's streams tonight (April 24), the points are seemingly calculated the same way as before.]
I have made a model of the new FIA points system based on data found in posts by @breeminator (*) (that link sadly doesn't work anymore) + @Jomas (**) and online streaming of FIA races. I won't bore you with the math behind the model yet (it's in the spoiler below the chart), but present this chart first (click for a larger view):
Example: in a lobby with a Dr average of 40000 position 1 will get about 195 points and position 4 will get about 170 points.
LINK IS BROKEN: *) https://www.gtplanet.net/forum/threads/2019-20-fia-exhibition-series-race-discussion.376323/page-40#post-12925837
**) https://www.gtplanet.net/forum/threads/having-an-early-look-at-the-new-fia-points.390849/
I have made a model of the new FIA points system based on data found in posts by @breeminator (*) (that link sadly doesn't work anymore) + @Jomas (**) and online streaming of FIA races. I won't bore you with the math behind the model yet (it's in the spoiler below the chart), but present this chart first (click for a larger view):
Example: in a lobby with a Dr average of 40000 position 1 will get about 195 points and position 4 will get about 170 points.
The points p awarded to finishing position n in a lobby with an average (i.e. arithmetic mean) Driver Rating of d is
where parameters {A, B, C} = {4.85818949E-03, 1.00798190E-03, 9.0044307867E-05}.
Rationale: if you take a look at one of the pictures in @breeminator's post mentioned above (EDIT: the link is dead, somehow), it's looks like an exponentially decaying function shifted in both axes, hence the postulated A+B*exp(-C*d) form. I then used mathematical software to find parameters A, B and C that provided the best fit to the existing data.
Parameter s = -0.7916/19 (because the player finishing 20'th gets ~ 20.84% of the points awarded to 1st place (1 - 0.2084 = 0.7916)
where parameters {A, B, C} = {4.85818949E-03, 1.00798190E-03, 9.0044307867E-05}.
Rationale: if you take a look at one of the pictures in @breeminator's post mentioned above (EDIT: the link is dead, somehow), it's looks like an exponentially decaying function shifted in both axes, hence the postulated A+B*exp(-C*d) form. I then used mathematical software to find parameters A, B and C that provided the best fit to the existing data.
Parameter s = -0.7916/19 (because the player finishing 20'th gets ~ 20.84% of the points awarded to 1st place (1 - 0.2084 = 0.7916)
LINK IS BROKEN: *) https://www.gtplanet.net/forum/threads/2019-20-fia-exhibition-series-race-discussion.376323/page-40#post-12925837
**) https://www.gtplanet.net/forum/threads/having-an-early-look-at-the-new-fia-points.390849/
Last edited:
