With only three races to go, the panorama of possible champions is clearer. Just to put numbers to it, I did some simulations to reproduce possible scores in those three races and who would be champion in each scenario.
The resulting estimated probabilities to be champion are as follows:
Code:
Moto3:
AREN 77%
OGUR 14%
VIET 6%
MASI 2%
ARBO 1%
Moto2:
LOWE 64%
BAST 29%
MARI 7%
MotoGP:
JMIR 71%
QUAR 20%
VINA 6%
MORB 2%
In each division there is a clear favorite and a main contender, cumulating more than 90% between them. Then there are one, two or three riders with an outside chance.
It is interesting that even if Ogura has just a point advantage over Vieti, the simulations, picking possible scores from past races, gives a distinctly greater probability for the Japanese.
I have considered all the past races equally relevant to forecast possible future races. It can be argued that riders' form changes through the season and, thus, the last races should be more relevant. This is true in a few cases and it is a valid point, but for most of the riders the scores through a season cannot be distinguished from random variations. This is something that I have calculated a few times in moto and car racing, however counterintuitive it seems.