Simulating 2024 USTA League Nationals - 4.0 Men

USTA League Nationals for 2024 begin this weekend!  The schedule is more compressed this year, normally Nationals would have started two weeks ago, but this year we start later and have more events each weekend.

This weekend has five events, 40 & Over 3.5 in Orlando, and then 18 & Over 2.5W and 5.0 in Scottsdale, 3.0 in San Diego, and 4.0 in Surprise.

I've been doing simulations since 2018 and have found that the four semi-finalists are usually in the top six to seven projected teams, and more often than you might think, I project the four semi-finalists perfectly.  There are surprises sometimes and a mid-pack team overachieves and advances, that is why the matches are played on the courts and not on a computer.  Still, it is interesting to do these simulations and see how the projections do.

Why do these simulations you ask?  The primary reason is that the format for USTA League Nationals is now a flight-less random round-robin where each team plays four other random opponents.  This introduces significant variations in schedule strength, the possibility of an undefeated team not making the top-4, and teams vying for the top-4 perhaps not having played head-to-head and unfortunate tie-breakers being used.  The simulations aim to educate folks on how it all works and look at what may happen.  Also see this write-up for some things to know about Nationals.

Because of the more compressed schedule, I likely won't have the time to do previews/simulations of all events, but will try to do a few.  Let me know which you'd like to see!

Like I did last year, I'll be primarily just giving the data from the simulation and not writing a bunch of words.  Most of it is self-explanatory, but the favorites and contenders will be listed in alphabetical order.  The strengths are based on top-10 averages.  Contact me if you want to learn more.

On to the projections from the simulation, next, the 18 & Over 4.0 men.

Teams: 17
Chance of 4 undefeated: 4.0%
Chance of 5 undefeated: <1%
Chance of tie for last spot / most likely size: 94% / 4
Strongest / weakest team strength: 4.08 / 3.82
Toughest / easiest schedule: 4.00 / 3.94
Favorites: Intermountain, Midwest, Missouri Valley, SoCal
Contenders: Middle States, NorCal, Pacific Northwest, Southern

The simulation likes the top-4 teams, but just one has a most likely record of 4-0, but three of them have a 38% or greater chance of going 4-0.  The contenders aren't far behind most also projected at 3-1.  Court records and the tie-breakers will be important.

What do you think?

Edit: Found an error and corrected it that adds Southern as a contender

Simulating 2024 USTA League Nationals - 4.0 Women

USTA League Nationals for 2024 begin this weekend!  The schedule is more compressed this year, normally Nationals would have started two weeks ago, but this year we start later and have more events each weekend.

This weekend has five events, 40 & Over 3.5 in Orlando, and then 18 & Over 2.5W and 5.0 in Scottsdale, 3.0 in San Diego, and 4.0 in Surprise.

I've been doing simulations since 2018 and have found that the four semi-finalists are usually in the top six to seven projected teams, and more often than you might think, I project the four semi-finalists perfectly.  There are surprises sometimes and a mid-pack team overachieves and advances, that is why the matches are played on the courts and not on a computer.  Still, it is interesting to do these simulations and see how the projections do.

Why do these simulations you ask?  The primary reason is that the format for USTA League Nationals is now a flight-less random round-robin where each team plays four other random opponents.  This introduces significant variations in schedule strength, the possibility of an undefeated team not making the top-4, and teams vying for the top-4 perhaps not having played head-to-head and unfortunate tie-breakers being used.  The simulations aim to educate folks on how it all works and look at what may happen.  Also see this write-up for some things to know about Nationals.

Because of the more compressed schedule, I likely won't have the time to do previews/simulations of all events, but will try to do a few.  Let me know which you'd like to see!

Like I did last year, I'll be primarily just giving the data from the simulation and not writing a bunch of words.  Most of it is self-explanatory, but the favorites and contenders will be listed in alphabetical order.  The strengths are based on top-10 averages.  Contact me if you want to learn more.

On to the projections from the simulation, starting with the 18 & Over 4.0 women.

Teams: 17
Chance of 4 undefeated: 5%
Chance of 5 undefeated: <1%
Chance of tie for last spot / most likely size: 91% / 4
Strongest / weakest team strength: 4.02 / 3.81
Toughest / easiest schedule: 3.96 / 3.89
Favorites: Eastern, Florida, Midwest, Southern, Texas
Contenders: Caribbean, Intermountain, New England, Southwest

The simulation likes the top-5 teams, the most likely records being 4-0 for just one and 3-1 for the other four.  One of the contenders is likely to go 3-1 as well, so could be a decent size tie-breaker there.

What do you think?

Is the USTA trying to boost Safe Sport trained numbers? Get trained, get your year-end rating early

If you pay attention to your e-mails from the USTA, you probably got one in the past few days promoting Safe Sport training with a perk being if you take one of the three on-line training courses, you will get your year-end NTRP rating level via e-mail early.

Specifically, the e-mail notes that year-end ratings should be published in the first few days of December, but if you want to get your rating sent to you via e-mail early on November 29th, they will do so if you take one of three Safe Sport training courses by November 11th.

The three courses are:

• Parent’s Guide to Misconduct in Sport
• Abuse Prevention for Adult Athletes
• SafeSport for Volunteers

I will admit I've been aware of Safe Sport and have always thought it was largely applicable primarily to juniors, so found it interesting the USTA was seemingly sending this to all USTA League players (those that care about their year-end level) as it wouldn't really be applicable to this audience.

I did take the 2nd and 3rd course above to see what the content was, and indeed they are both very much focused on juniors or youth/collegiate teams and scenarios related to that.  Certainly what is covered can apply to any sporting situation and USTA League does have teams, but none of the content addressed those sorts of scenarios.

The cynical part of me thinks the USTA has Safe Sport training metrics they need to meet and is looking to boost the numbers by using the perk of getting your year-end rating early.  There is nothing wrong with that, and some USTA League players have kids playing sports or interact with high school or college teams, but those metrics will need to be looked at with a lens knowing they were perhaps artificially boosted with members the training doesn't really apply to.

Am I being too cynical?

It is also interesting that the USTA apparently will have the ratings ready by 11/29 but it choosing to hold on to them and not publish for four days.  This is perhaps by design, wanting to wait until the work week to publish and have staff available to deal with the influx of questions and complaints, but if a lot of people take the Safe Sport training and get their rating e-mailed to them early, it is really no different than just publishing the ratings on TennisLink on 11/29.

In fact, it could cause more confusion as players get an e-mail with their new rating and then go to TennisLink and still see their old rating.  It all seems a little odd.

And to be honest, it really would be a service to league players to just publish a little earlier as some leagues start immediately in January and have registration deadlines in early December which really puts a crunch on captains to figure out who can captain which team and figure out rosters.  The four extra days could really help.

What do you think?

USTA League Nationals is less than a week away

It is the middle of October and if this was like prior years years, we'd be several weeks in to USTA League Nationals events.  But this year the schedule has been compressed into four weeks and things get started next week.

The events starting this coming Friday the 18th are:

• 18 & Over 2.5 women - Scottsdale
• 18 & Over 5.0 - Scottsdale
• 18 & Over 4.0 - Surprise
• 18 & Over 3.0 - San Diego
• 40 & Over 3.5 - Orlando

Thoughts go out to all those in Florida who had to deal with a sequence of hurricanes, and the National Campus in Orlando did close for a few days for each one, but thankfully the forecast for this weekend, and the following which will also have an event there, looks good.  There is the normal Florida chance of a shower, but otherwise looks like 80's for both weekends.

Good luck to all the teams!

Well, it happened again, a 4-0 team was sent home at Florida 18 & Over Mixed Sectionals

A few days ago I wrote about the use of the unflighted round-robin format being used at 18 & Over Mixed Florida Sectionals, and the particular implementation allowing for five teams to go 4-0 with only spots for four to advance.  The possibility existed in both the 6.0 and 9.0 flights.

In the 9.0 flight, the possibility had disappeared after just the first day and first two matches, and by the end of the round-robin there were only two 4-0 teams still standing, along with a 3-1 team, and the rarity of a 2-2 team advancing even happened!  In fact there were three 2-2 teams vying for the 4th spot, two of them were 6-6 on courts and both were from the same area (Orange/Seminole) and had both finished 4-2 in local league play, but the one that won local league (Siqueira) lost one fewer set and got the spot.  All in all a competitive flight and the format worked fine.

Unfortunately it didn't work fine for the 6.0 flight.  As a reminder, here is what the visualization of the round-robin schedule looked like.

You can see five teams across the top all playing the five teams across the bottom.  If all the teams in one of the rows won, they'd all five be 4-0 and with only four advancing, one would be left out.

Well, it happened, and not only did it happen, it happened in pretty spectacular fashion.  I'd noted in my earlier post that because it could happen, the teams would need to go all out to ensure they positioned themselves to win the standings tie-breaker and not get left out, and they all did.  Here were the standings:

Team	Wins	Losses	Ind. Wins	Ind. Losses	Sets Won	Sets Lost	Games Won	Games Lost
Marion	4	0	12	0	24	1	148	71
Orange	4	0	11	1	22	2	141	56
Miami	4	0	11	1	23	2	142	60
Alachua	4	0	11	1	22	3	145	82
Broward	4	0	11	1	23	4	141	61

No team lost more than one court across 12 courts played.  No team lost more than four sets across all the matches.  Every team won more than 140 games (over 11.75 per match).  The "worst" of the five teams by the standings actually had the 3rd best game winning percentage with two other teams losing 10 and 21 more games respectively.  So all these teams appear to have been putting their best foot forward and trying to win as easily as possible.

But in the end, a 4-0 / 11-1 / 23-4 team didn't advance to the semis!  Their one court loss was in a super tie-break, and they happened to win two other courts in super tie-breaks, and losing four sets total was enough to put them behind the other teams resulting in finishing 5th.

I do not know how this went over with the Broward team, but I can't imagine it went well.  Hopefully they knew it was possible so weren't surprised, but it is still tough to play that well and seemingly dominate, and then have four other teams finish ahead of you.

This was avoidable if an actual random schedule had been used.  With just 10 teams, playing four matches is more than enough to decide things fairly, but the one format that would cause problems was used and it happened, similar to how it happened before.

The point of this point is not to say I told you so, but more to educate players about the pros and cons of the format to be aware of it when it is used and not be surprised, but also to be able to go to your area/district/section coordinators and ask for the best formats to be used.  I'd also hope that those coordinators are learning from this experience and won't make the same mistake in the future.

To be fair, they did change the format from two years ago to take the top-4 teams to semis instead of just the top-2 to a final, but that wasn't sufficient and they should have known it wasn't sufficient and ensured this scenario couldn't happen.

My condolences go out to the Broward team, but hope once again this can be a lesson to those using the unflighted round-robin format of the importance of using it in the correct way.

Has USTA Florida botched the random round-robin format again?

In 2022 at Florida 18 & Over Mixed Sectionals, a disaster happened where two undefeated teams were sent home in the 9.0 flight.  Might it happen again?

This happened because the unflighted round-robin format was used with 10 teams with two sets of five teams on each side playing four teams on the other side.  This means in theory if all the teams on one side win, there could be five 5-0 teams, and they only took two teams to a final.  It is all explained here.

What is worse is they had five strong teams on one side and five weaker teams on the other, and sure enough four teams went 4-0 and two were sent home.

You would think an event like this happening would be a great learning opportunity and steps would be taken to prevent it from happening again.  Well, sort of.

This year, again at 18 & Over Mixed Sectionals, there are again 10 teams at both the 6.0 and 9.0 levels, and unflighted round-robin is being used with each team playing four matches.  The good news is it appears they will take the top-4 to semis instead of just the top-2.  Phew, we are safe then right?

Not so fast ...

Both levels appear to have had the short-sighted "two sides of five teams each" approach to scheduling, so there still remains a chance of five undefeated teams!

Here is what it looks like for 6.0 flight if you visually look at the schedule.

Yep, there are clearly two groups of five teams, the top row plays the bottom row, and it is possible either row could go 5-0.

And for the 9.0 flight.

We see the same thing here.

Now, will it happen?  The chances are probably low, but the mere fact the format is used in a way that allows for it to happen when it could easily be avoided is a gaffe.  All one has to do is actually make a random schedule instead of constructing a schedule that explicitly allows for the worst case scenario.

I don't know if anyone from the USTA Florida reads my blog, but if so, I volunteer my services for free to review your unflighted round-robin schedules in the future so this can be avoided.  I would prefer to nip this sort of thing in the bud before it is rolled out and can't be changed.  It isn't fair to the players involved.

Let's hope the worst case doesn't happen for the players' sake.  If the disaster is avoided, that doesn't mean the format is ok, it just means they got lucky.

Even the National format with 17 teams and a schedule that actually appears random has a chance of sending an undefeated team home and it has happened (also here).

Good luck to all the teams involved.  Make sure you smash your opponent lest you go 4-0 and lose out on the standings tie-breaker.

The World Tennis Number (WTN) changed again

Being someone very interested in ratings, I've followed the World Tennis Number (WTN) as an interested party, including doing some early analysis on how it mapped to NTRP.  You can see all my posts on WTN here.

The ratings themselves have gone through a few "adjustments", most recently just over a year ago, but a notice was sent out of another impending adjustment a few days ago, and it appears with today's update, it has gone into effect.

Both the USTA and LTA sent e-mails with a heads up about the change, from the LTA's e-mail, the goal was to:

• Improve accuracy by adjusting where some players start on the scale
• Make every match count whilst recognizing the difference in ability between you and your opponent
• Faster movement of a player's WTN to reflect their level

The USTA's e-mail said more.  On where players start:

Based on the data, it’s become increasingly apparent that some players are being initialized too high. We want junior players to experience appropriate improvements to their rating as their game improves. This change will ensure players see this reflected in their rating as their level improves.

Regarding faster movement:

Player ratings will transition to their ‘true’ skill level faster from the initial rating given to them. This will mean that when a player is given an initial rating that is too low or too high, they will update to a more accurate rating much faster from their first few matches.

And about matches counting:

Every match counts towards a player's rating but it’s important that the rating change reflects the difference in ability. Players will experience smaller changes to their WTN in matches when losing to a superior player. However, players will still see a sharp increase to their rating should they beat a player with a substantially better rating.

The USTA does have a document highlighting the changes with an FAQ as well.

This all seems good, but we'll see how it plays out.

On a personal level, the adjustment a year ago saw my rating get worse, going from around 21 one all the way up to 29 (remember, the scale is from 40 to 1, 40 being the lowest/worst rating).  That was a pretty major shift, seemingly because the USTA felt as a group, adult league players were rated way too high.

This time around the shift went back the other way, but was far smaller, I went from a 27 to 25.

Is this better?  I'm not sure, it is difficult to do a thorough analysis on WTN as the data isn't readily available.  My rating getting better feels nice, but it is just a number, and one league players as a whole don't really seem to follow closely.

What do you think?  Do you care about your WTN?  Do you check it each Wednesday when the new ratings are published?