re: Tennis 2020 -- The Unofficial Non-GS Thread

Imagine trying to create a rating system that could beat the betting markets. What type of factors would you have to consider besides skill level? Well, just off the top of my head...

1. Current form
2. Surface
3. Court speed
4. Weather
5. Fatigue
6. Accumulated fatigue
7. Head to head
8. Motivation
9. Injuries
10. Record vs. certain player types

It would be very difficult to quantify any of these things, let alone all of them... for each individual match.

You might be better off using Elo ratings in a reverse sort of way compared to betting odds. For example, if Elo says Thiem should beat Nishikori 60% of the time, but the money line (-300) is implying that Thiem wins 75% of the time, maybe Thiem would be the play. We could surmise that oddsmakers and/or bettors are seeing a big difference in the other factors not related to skill level, and we could/should trust in the wisdom of crowds.

I haven't tried this, but it seems like a decent idea. I might give it a shot when they start back up next month.