Monday, October 1, 2018

Predicting the 2018 USTA League Nationals semi-finalists and champion - 18 & Over 4.0 women

I've written about my simulations of the new format for USTA League Nationals, and hinted at who some favorites are, but now it is time to actually use the simulations to predict who the semi-finalists will be and of those, who the favorite to win it all is.

As a reminder, the simulations use the top-10 average for each team using my estimated ratings, and then account for some variation in play by each team and run through a million iterations to see what the most common result might be.

Next, we'll look at the 18 & Over 4.0 women that is coming up this weekend in Arlington, TX.  Here is the link to the earlier simulation write-up.

The most likely teams to make the semis are:
  • Hawaii - 93%
  • New England - 92%
  • Northern California - 85%
  • Midwest - 82%

Note that these percentages are really very high.  I am perhaps not allowing for enough variability in my simulations, but I'm going with this and we'll see how well it predicts.

Note also that these are not the top-4 rated teams, as schedule strengths are a factor and two of the above teams slips in to the semis without being one of the highest rated 4 due to an easier schedule while two of the top-4 rated teams has a tougher schedules.

The chances of these teams advancing is very close for the top-2 and 3rd/4th, so tie-breakers will come into play for the seeding in all likelihood, but if they are seeded as I list them, in the semis, Hawaii would be favored over Midwest and New England and Northern Cal would be a pick'em match, New England getting the nod for having the better chance of making the semis.  In the final Hawaii would be the pick.  This one will be interesting as Hawaii's ratings can sometimes be off a bit bit relative to other sections because they are, well, an island.

Should one of the above teams falter, Middle States, Florida, or Eastern are most likely to take advantage of it and advance, Southern Cal is also dark horse to find their way in if the stars align.

If you are on one of the teams, or any team going to Nationals, it is not too late to get more details from my simulation or a flight report to scout opponents and plan your line-ups.  While the above is what the simulation says, the ratings aren't perfect between sections, players have good and bad days, not everyone will make the trip or players other than the top-10 will play, so some smart captaining can still go a long way and my reports can really help.  Contact me if interested in any reports.

Update: Corrected a few percentages but the predictions stayed the same.

No comments:

Post a Comment