Back before Hangtown, over 3 months ago, we made our predictions as to who would come out on top
for the 2016 MX Outdoor Nationals season. We took into account past performance, age, injury history, and even (as a minor factor) the
2016 Supercross results. And this is what we came up with, listing the actual finish on the right:
Right or Wrong?: 2016 450MX Pre-Season Predictions
Now, because of all the injuries, things went a little haywire. No matter what, the number 1 pick was going to be wrong -- Ken Roczen dominated 2016 outdoors, and I think most everyone will agree that this would have been true even if Ryan Dungey, Jason Anderson, Trey Canard--and many others--were healthy all season long.
Because of these riders with significant time lost to injury, the full-season results are all over the place. To look at their respective performance for the time the riders were healthy and riding, we can look at results on a per-week basis. These numbers include only the weeks that the racers did ride:
For instance, Dungey was predicted to earn 45 points per week, and he earned 43.7, for a difference of 1.4 fewer. Broc Tickle was predicted to earn 18.3 points per week, and he earned 21.4, for a difference of 3.1 more. This "per-week" comparison still has its flaws since the actual points are inflated -- the thinning of the field helps those who managed to ride the whole season, for instance Phil Nicoletti, who had one of the biggest variances (that said, staying healthy is partly a skill [citation needed ;) ], so credit to those who rode 12 weeks).
The "per-week" Rank is a little better, I think, at comparing between all the riders. The top 4 were all within 1 spot by this measure. Blake Baggett scored much lower than predicted, though he had 2 DNF's out of 10 races, so that weighed his rank down heavily. Actually, only 5 of the predicted Top 20 had a variance in their rank by more than 2 spots, which seems pretty good in retrospect.
Last, we'll look at those who the predictions missed entirely. Benny Bloss was predicted at 22nd, but he finished 6th overall with 19 points per week (predicted at 4 points per week). As mentioned, some of that was due to injuries to other riders, but even if everyone had stayed perfectly healthy and he'd finished, say, 15th, that still would have been much better than the prediction.
Matt Bisceglia was mistakenly listd as one of the 250 riders for the predictions -- had he been listed with the 450 guys, he would have slotted in at 14th. He actually finished 15th, partly due to injury though -- but, he was 12th in points per week, which is still pretty close to the prediction of 14th.
Dean Wilson was left off of the predictions because he wasn't expected to return until the second half of the season at best, and probably not at 100% right away -- meaning he wouldn't earn enough points to break into the Top 20. If he were expected to have a full season, he would have been predicted at 12th -- he actually finished 19th overall but was 8th in points per week, a good deal better than the prediction. Actually, 8th is pretty good for someone coming off an ACL tear and having to jump in mid-season.
Similar analysis for the Lites bikes. The first thing that's apparent is the #1 prediction, Jeremy Martin, had a season to forget, at least for someone who was a heavy favorite going in to the season. Aaron Plessinger also didn't live up to his prediction, but that has been much discussed here. In the Top 10, 5 guys finished within 2 spots of their prediction. I don't know if that's particularly good or bad, but it doesn't seem that great:
Also much discussed in previous articles is how the 250 predictions haven't been that great all season. At least the "per-week" rankings, the numbers look a little better, especially in the Top 10. Overall, 11 out of 20 were within 2 spots of their prediction (I bumped out Matt Bisceglia since he didn't actually ride 250's after all), which seems at least decent, while 7 out of the top 10 were within 2 spots:
Compared to their predicted per-week rank, Alex Martin was the biggest positive standout (no surprise), while Mitchell Oldenburg also had a much better ranking than predicted. On the negative side, Christian Craig did much worse than predicted, but he only had 4 motos, 2 of which he finished in 40th because of the crash that ended his season -- without the DNS from the second of those 2, he would have earned 7 points per moto, which is better than his prediction. Other than that, the underperformer would be Jordon Smith who was predicted to be right in the middle at 11th but performed on a per-week basis considerably worse, in 16th.
So, there we have it. The accuracy of the predictions is in the books for Motocross season one at MotoXGraphs! We'll put up a way-too-early prediction for 2017 Supercross and Motocross soon. Should be fun, even if a bit trivial -- though we might even learn something.