Analyzing 2023 USTA NTRP year-end ratings - Distribution of players across levels by gender

2023 year-end ratings were published a few days ago and I've started doing my stats with bump rates by section and gender, bump rates by level and gender, strange bumps, and player counts with year-end ratings.  Next up, distribution of players by level and gender.

This analysis looks at the players with year-end C ratings for 2023 and for comparison 2022.  The bar represents the number of players and the annotation is what percentage that number is of the total.

First up, players by level overall.

We see the standard "normal" distribution with just some small changes from last year.  There was some growth in raw numbers at lower levels, 2.5 thru 3.5, but the 4.0 and 4.5 levels shrunk slightly.  However, while there were more 3.5s in 2023, the percentage that are 3.5s went down slightly due to the significant growth of 3.0s.

Since we saw that generally speaking more players are bumped up than down, this is likely due to new players coming in at lower levels and getting year-end ratings while the higher levels perhaps have some players no longer playing USTA League.  I will have to do some analysis on this to see if that is the case.

Next, the distribution of women by level.

Similar story here, but the increased number of 3.0s was a bit larger and the 3.5s slightly less.

Last, the distribution of men by level.

The men had noticeably more 3.0s and 3.5s, but from a percentile distribution standpoint, over 1% fewer 4.0s and nearly 1% fewer 4.5s.

Stay tuned for more!

Analyzing 2023 USTA NTRP year-end ratings - Are the number of players with ratings going up or down?

Earlier this year, I wrote about the USTA's press release on tennis growth and in that post I introduced a new chart showing participation based on the number of players who receive year-end ratings.  Now that ratings are out, I can add 2023 to the mix.

Here is the chart showing the number of players with year-end ratings since 2012.

We see the metric has gone up slightly from 2022 (231K to 238K), it is still lower than all the other years from 2012 to 2021.

There are a number of possible reasons the number has gone up including:

• We are still in the post-COVID recovery period where players continue to come back to league play
• The COVID tennis boom we heard about from the TIA and USTA is finally resulting in some of those new players joining league play
• Some sections with secondary leagues that didn't count for ratings in the past are now counting these leagues, so players that play in them are now getting year-end ratings.

The bulk of the increase from 2022 is with C rated players which are up around 6K on their own.

2024 will be interesting to see where we end up!

Analyzing 2023 USTA NTRP year-end ratings - What strange or unexpected bumps were there?

2023 USTA NTRP year-end ratings are out and I've taken a look at bump rates by section and by level, and before slicing the data more, I thought I'd take a step back and just look for edge case bumps we don't see often or appear to be glitches in the system.

First up, the double bump.  Bumps up and down happen, in typical years about 15% of players are bumped, around 9% go up and 6% go down.  It varies from year to year but that is roughly what you can expect.  Bumps up or down just one level are the most common of course, but do double bumps happen?  If so, how often?

For bumps up, I show 29 players whose 2023 year-end C rating was at least 1.0 higher than their C rating they were using for 2023.  So it does happen, but not a lot.  What about a triple bump up?  I found one, sort of, but it was actually a glitch (see below).

For bumps down, I only found one player that has this, sort of.  I say sort of because they were a 2022 year-end 4.5C, but were able to appeal down to 4.0, and at 2023 year-end received a 3.5C.  So they were a double bump down, but "benefited" from appealing down and being able to play 4.0, although they still played 4.5 and only played one Tri-Level match on the 4.0 court.  To be honest, the 3.5C doesn't appear warranted, but the USTA gave it to them.

There is always interesting year-end ratings around some players that go to Nationals.  Here are a few I've seen that are a little baffling:

1. A team finishes 3rd at Nationals, and no one is bumped up other than a player who went 4-5 on the year.  Others on the roster won 70+% of their matches on the year including 3-0, 4-1 or 5-1 at Nationals and weren't bumped up.
2. A 5.5C takes time off and comes back as a 5.0S and goes 9-0 including 4-0 at Nationals and gets a 4.5C.

For what it is worth, my ratings don't agree with either of the above.

How about some bumps that appear to be glitches or bugs or due to something other than the algorithm?

• A 5.0C takes time off and comes back as a 5.0S and goes 1-2 playing 5.0.  What is their year-end rating?  2.5C!  Clearly something amiss here, I'm guessing it will be fixed and changed at some point, and in fact as I just checked as I write this it has been, so good on the USTA for fixing it.  I saw a few more similar to this that have also been fixed.
• A 3.5S played in 2022 and was DQ'd to 4.0 mid-year, but somehow got a 2.5C year-end rating.  Clearly a glitch and it was fixed, but not for months it seems so the 4.0C they got this year looked like a triple bump but wasn't once the fixed 4.0C from last year was recorded.
• A 4.5S plays only Mixed, loses two matches early, then goes 14-1 and gets an M rating as expected, but it is a 4.0M!  My ratings had them closer to a 5.0 than a 4.0.
• A 5.0T self-rates as a 5.0 and plays 5.0, does go 0-3, but ends up as a 3.5C!

For some of these that are glitches, the USTA does correct them as I note so that is good.  But a few are outstanding and make one scratch their head.

What crazy bumps have you seen?

Analyzing 2023 USTA NTRP year-end ratings - Bump rates by level and gender

The 2023 USTA NTRP year-end ratings have been out for just over a day now and I started doing analysis with bump rates by section and gender, now on to slicing it another way, by level and gender.

First, here are the bump rates just by level.

This isn't at all unexpected, lower levels have more bump ups, higher levels have more bump downs.

Breaking it out by gender, first, the women.

The women have slightly more bumps down as the level gets higher, but looks very similar to the general trends.

And here is how the men look.

After seeing the women, as you might expect, much the same, just with fewer bumps down as the level goes up.

Stay tuned for more!