Thursday, October 31, 2019

Simulating 2019 USTA League Nationals - 55 & Over 8.0 Women

As I did last year, I will be doing simulations of each Nationals to predict who the most likely four teams are to make the semi-finals, and also look at other interesting things that may occur.  Last year, the simulations did a pretty good job predicting who would advance to the semis including having all four teams predicted correctly several times.

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.

Nationals are approaching the fifth weekend of competition, the 55 & Over 8.0 women's event being held in Arizona.  Here is what the simulation says is likely to happen.

As a reminder, my simulations are done using my Estimated Dynamic NTRP Ratings and looking at the average rating for the top group of players on each team, the actual schedule each team will play, and then doing a million simulations of the matches with some random variation in each team's expected result.

First, the schedule with 17 teams will nearly always allow for five teams to be undefeated and one sent home (<1% chance), but four undefeated would be a nice and tidy result with no drama, but there is just a 2% chance of that so we are more than likely to have some tie-break scenarios with only a 14% chance of three undefeated.

That leaves a very good chance, 97%, that there will be a tie for the last spot and it come down to tie-breakers.  That tie is likely at a 3-1 record.  The most likely size of the tie is four at 27%, but five is nearly as likely at 24%, six is still 20%, and a 3-way tie just 14%.  The chances of larger ties diminish pretty quickly, but an 11-way tie is theoretically possible, but it appears a multi-way tie is a very high probability and it could be reasonably large.

Should there be a tie on team record, it comes down first to who has the best court record, then head-to-head (if applicable), then to who lost the fewest sets, then who lost the fewest games, and finally percent of games won.  There is a 51% chance it comes down to the sets lost tie-breaker, and there is a 37% chance that is between two teams and a healthy 40% chance it is between three teams,  If the controversial tie-breakers are going to come into play, this is the likely spot.

The schedule strengths are moderately close, resulting in the high likelihood of ties, the team with the easiest schedule having an opponent average of 3.93 while the team with the toughest schedule having an opponent average of 4.06.  This leads to an easier road for some teams than others, one of the predicted semi-finalists getting there due to an easier schedule.

So who is most likely to come out in the top-4?  Middle States, New EnglandHawaii, and SoCal are solid choices, Middle States the most likely.  Florida is in a virtual tie for fourth most likely, and NorCalMoValley, Southwest and Southern will step in if someone slips up.

For those interested, I offer a variety of reports to make Nationals more fun and help captains prepare.  I have a Simulation Report that has all of the details of the simulation including the average ratings for each team, each team's schedule strength, the most likely record for each team, and the chance of each possible record for each team.  I also offer reports to help teams scout opponents in more detail, both a Flight Report with full roster averages, top-8 averages and played by court averages for each team, as well as full Team Reports with detailed ratings for each rostered player and stats who who plays with who and on which court and how they do together.  Contact me if interested in any of these reports.

Monday, October 28, 2019

Simulating 2019 USTA League Nationals - 18 & Over 9.0 Mixed

As I did last year, I will be doing simulations of each Nationals to predict who the most likely four teams are to make the semi-finals, and also look at other interesting things that may occur.  Last year, the simulations did a pretty good job predicting who would advance to the semis including having all four teams predicted correctly several times.

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.

Nationals are approaching the fifth weekend of competition, the 18 & Over 9.0 Mixed event being held in Orlando.  Here is what the simulation says is likely to happen.

As a reminder, my simulations are done using my Estimated Dynamic NTRP Ratings and looking at the average rating for the top group of players on each team, the actual schedule each team will play, and then doing a million simulations of the matches with some random variation in each team's expected result.

First, the schedule with 17 teams will nearly always allow for five teams to be undefeated and one sent home (2% chance), but while a schedule can be done that precludes it, it is possible there could be six undefeated, although very unlikely with about 500 of the million simulations ending that way.  Four undefeated would be a nice and tidy result with no drama, but there is just a 19% chance of that so we are more than likely to have some tie-break scenarios with a 41% chance of three undefeated.

That leaves a pretty good chance, 79%, that there will be a tie for the last spot and it come down to tie-breakers.  That tie is likely at a 3-1 record.  The most likely size of the tie is four at 28%, but three is nearly as likely at 26%, five is still 22%, and a 2-way tie just 11%, 6-way 9%.  The chances of larger ties diminish pretty quickly, but an 11-way tie is theoretically possible, but it appears a multi-way tie is a very high probability and it could be reasonably large.

Should there be a tie on team record, it comes down first to who has the best court record, then head-to-head (if applicable), then to who lost the fewest sets, then who lost the fewest games, and finally percent of games won.  There is a 32% chance it comes down to the sets lost tie-breaker, and there is a 59% chance that is between two teams and a healthy 32% chance it is between three teams,  If the controversial tie-breakers are going to come into play, this is the likely spot.

The schedule strengths do vary a lot, the team with the easiest schedule having an opponent average of 4.42 while the team with the toughest schedule having an opponent average of 4.63.  This leads to an easier road for some teams than others, one of the predicted semi-finalists getting there due to an easier schedule.

So who is most likely to come out in the top-4?  NorCal, NorthernSoCal, and Eastern are solid choices, NorCal the most likely.  PNW, TexasMid-Atlantic, and Florida will step in if someone slips up.

For those interested, I offer a variety of reports to make Nationals more fun and help captains prepare.  I have a Simulation Report that has all of the details of the simulation including the average ratings for each team, each team's schedule strength, the most likely record for each team, and the chance of each possible record for each team.  I also offer reports to help teams scout opponents in more detail, both a Flight Report with full roster averages, top-8 averages and played by court averages for each team, as well as full Team Reports with detailed ratings for each rostered player and stats who who plays with who and on which court and how they do together.  Contact me if interested in any of these reports.

Sunday, October 27, 2019

What is on tap for week 5 of USTA League Nationals? More 55 & Over and Mixed gets going, what about the weather?

Week 4 of USTA League Nationals is complete, so it is time to look forward to week 5 and what we can expect.

The first 55 & Over event was just completed this weekend, and that division continues with the men and women at the 8.0 level playing in Surprise, AZ.

With 18 & Over and 40 & Over complete, it is now time for Mixed to get going and it starts with the 18 & Over 7.0 and 9.0 levels playing in Orlando.

We are still over 4 days away from play commencing, but we do need to take a look at the weather forecast given the weather issues this weekend and last weekend.

Arizona's forecast calls for sun all three days and upper 70s to lower 80s.  No chance of rain shown on either side of the weekend either so this looks great for the 55 & Over players headed there.

Orlando's forecast calls for isolated thunderstorms on Friday, a 40% chance of afternoon showers on Saturday, and mostly cloudy on Sunday.  This forecast is not dissimilar to the forecast for this weekend a week ago, so let's hope it rain and delays don't happen again.

More 2019 USTA League National Champions Crowned - Week 4 - A 7-way tie!

Week 4 of USTA League Nationals is complete and we have four more champions.

The 55 & Over 7.0 men and women played in Arizona and with nice dry conditions the event went off without a hitch.

The women saw NorCal beat Southern in one semi and Hawaii beat Texas in the other both by 2-1 scores, and then NorCal took the final 3-0.

The men's semis had SoCal beat Southern 2-1, and Northern beat Florida also 2-1, Northern winning by final by the same score.

The 40 & Over 4.5+ men and women was played in Orlando at the National Campus, and unfortunately for the second time in the two weeks the site has been used, there was rain that disrupted matches, this time forcing a short format on Saturday resulting in the majority of round-robin matches using no-ad and starting all sets at 2-2, so first to four sets, and some very late finishes Saturday night.

And for the women, those late finishes had meaning as there was a 7-way tie for 2nd thru 8th at 3-1 to determine the last three teams to advance.  When it was all decided, Eastern had beaten PNW 3-2 and Midwest beat Southern 4-1, and Midwest won the final 3-2.

The men had just 16 teams so fewer matches to get in and didn't finish as late but still had a 5-way tie for 3rd thru 7th, and when all was said and done, Southern beat Texas 4-1, PNW beat Mid-Atlantic 3-2, and Southern won the final 3-2.

Nationals are not done, we still have 55 & Over events and Mixed will get started this coming weekend.

Saturday, October 26, 2019

USTA League Nationals Week 4 Semi-finalists - What tie-breakers had to be used? 7-way tie for 2nd!

Week 4 of USTA League Nationals has completed the round-robin, despite some delays due to rain and shortened format matches for the 40 & Over 4.5 men and women in Orlando, and we have four sets of semi-finalists.

In Orlando, the women's semi-finalists are Eastern, who was 4-0, then Midwest, Southern, and Pacific Northwest all at 3-1 who had to beat out four other teams in a 7-way tie for 2nd.  Florida was tied with PNW at 3-1 / 12-8 and they each had lost 18 sets as well.  So despite PNW winning one more set (29 to 28, the USTA does not look at sets won or sets differential), it went to games lost and PNW had a big lead there, so thankfully deservedly got the spot.  But there were three more teams with 11-9 court records, so it was really close.  The simulation had two of the four semi-finalists predicted, and a third semi-finalists was mentioned as lurking.

The men 4.5+ semi-finalists are Southern and Pacific Northwest, both undefeated, joined by Mid-Atlantic and Texas both 3-1.  Three others were 3-1, but at least two courts worse than Texas so no strange tie-breakers.  My simulation did predict three of the four semi-finalists.

The 55 & Over 7.0 in Arizona had no weather issues and completed play without any drama or shortened formats.

The women's semi-finalists are NorCal and Hawaii, both 4-0, joined by Texas and Southern.  SoCal was tied with Southern at 7-5 on courts, but Southern had won head-to-head so they got the spot despite SoCal losing fewer sets and winning more sets!  And losing fewer games and winning more games too!  I have no argument with this tie-breaker, it is just remarkable that Southern won just the right courts to pull it off while SoCal won their other matches a lot easier.

The men's semi-finalists are SoCal, Florida, and Northern, all 4-0, and Southern got the last spot being one court better than Texas.

Good luck to all the teams on Sunday!

Weather update 3: Week 4 of Nationals in Orlando

As we arrive at Saturday morning, word is that in Orlando where the 40 & Over 4.5+ Nationals are being played, they did not complete Friday's matches and will be playing catch-up today.  And with more rain forecast late in the day, it appears they will be forced to change matches to a short format to have hope of getting them in.

If TennisLink is up to date, it appears the women's 3:00 and 4:30 matches from yesterday (four total), and men's matches in the same time-slots also remain (seven total).  To make matters a bit worse, it appears they can't get started until 9:00 am today.  Here is the schedule:

  • Friday 3pm matches will now be at 9am
  • Friday 4:30pm matches will now be at 10am
  • Updated Saturday Schedule:
  • Saturday 7:30am matches are now at 11am
  • Saturday 9:30am matches are now at noon
  • Saturday 11am matches are now at 1 pm
  • Saturday 1pm matches are now at 2 pm
  • Saturday 3pm matches will remain at 3pm
  • Saturday 4:30pm matches are now at 4pm

Astute readers will note they only have hour slots for these matches, so yes, a short format will be used.  They will play 2 out of 3 short sets, first to four games, win by two, no ad. At 4 games all, they will play a seven point tie break. If you split sets, you will play a match tiebreak to 10 points.  This is effectively normal format, but with no-ad and played as if you are starting at 2-2 each set.

Now, the challenge is that the forecast calls for rain to come back starting around 3pm.


If the rain comes at 3pm, and assuming matches are done quickly enough to stay on schedule, that still leaves 10 matches currently scheduled to be played that may be affected.  That is 50 court hours needed and even if they can get the courts dry at 8pm, and also use the limited indoor courts, it could be a late night to get everything in.

Now, credit to the USTA for communicating with the players and doing what they reasonably can do adapt to a challenging situation.  They are also helped a bit by the men's event having just 16 teams, otherwise they'd have a few more matches they'd have to be getting in.

Let's hope it goes as well as it can and the format doesn't have to be shortened any further.

Friday, October 25, 2019

Weather update 2: Week 4 of USTA League Nationals in Orlando

The weather report yesterday wasn't promising, and it turns out it rained quite a bit in Orlando today and has disrupted play significantly.

As I understand it, at least some 11am matches have not been completed yet as of 6pm, but they are hoping to get them complete or back in progress by 7pm.  Then:

  • 1pm matches plan to go on at 7:00 pm
  • 3:00 pm matches at 9:00 pm
  • 4:30pm matches at 10:30 pm

The handful of indoor courts will be used at 7:00 pm but they hope to be playing outdoors as well.

Word is they will try to get back on schedule and play tomorrow's matches as planned, but there is a 50% chance of Thunderstorms again.

Update: Friday's matches were not completed, short-format on Saturday.

Thursday, October 24, 2019

Weather update: Week 4 of Nationals in Orlando

When I did the start of the week weather check, weather was great in Arizona and some possibility of rain in Orlando.

As of Thursday night, the forecast still calls for a decent chance of rain.  With a 40-50% chance of thunderstorms every day, it looks like there is a good chance play gets interrupted.  Hopefully it will be better than the forecast or the interruptions be brief.



Simulating 2019 USTA League Nationals - 40 & Over 4.5+ Women

As I did last year, I will be doing simulations of each Nationals to predict who the most likely four teams are to make the semi-finals, and also look at other interesting things that may occur.  Last year, the simulations did a pretty good job predicting who would advance to the semis including having all four teams predicted correctly several times.

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.

Nationals are approaching the third weekend of competition, the 40 & Over 4.5+ women's event being held in Orlando.  Here is what the simulation says is likely to happen.

As a reminder, my simulations are done using my Estimated Dynamic NTRP Ratings and looking at the average rating for the top group of players on each team, the actual schedule each team will play, and then doing a million simulations of the matches with some random variation in each team's expected result.

First, with 17 teams playing a random four opponents and there will always be scenarios where five teams are undefeated, but less than 1% of them.  Six is even possible but extremely unlikely.  Four undefeated would be a nice and tidy result with no drama, but there is just a 22% chance of that so we are more than likely to have some tie-break scenarios with a 42% chance of three undefeated.

That leaves a very good chance, 76%, that there will be a tie for the last spot and it come down to tie-breakers.  That tie is likely at a 3-1 record.  The most likely size of the tie is a three or four at 31%, but three has a 27% chance, five 21%, and two 10%.  The chances of larger ties diminish pretty quickly, but an 11-way tie is theoretically possible, but it appears a multi-way tie is a very high probability and it could be reasonably large.

Should there be a tie on team record, it comes down first to who has the best court record, then head-to-head (if applicable), then to who lost the fewest sets, then who lost the fewest games, and finally percent of games won.  There is a 20% chance it comes down to the sets lost tie-breaker, and there is a 75% chance that is between two teams and a 22% chance it is between three teams,  If the controversial tie-breakers are going to come into play, this is the likely spot.

The schedule strengths do vary a bit, the team with the easiest schedule having an opponent average of 4.46 while the team with the toughest schedule having an opponent average of 4.58.  This leads to an easier road for some teams than others, one of the predicted semi-finalists getting there due to an easier schedule.

So who is most likely to come out in the top-4?  Florida, Intermountain, Eastern, and Southern, Florida the most likely.  Several others are in a group waiting for one of those to slip up including Middle States, Northern, and Midwest.

For those interested, I offer a variety of reports to make Nationals more fun and help captains prepare.  I have a Simulation Report that has all of the details of the simulation including the average ratings for each team, each team's schedule strength, the most likely record for each team, and the chance of each possible record for each team.  I also offer reports to help teams scout opponents in more detail, both a Flight Report with full roster averages, top-8 averages and played by court averages for each team, as well as full Team Reports with detailed ratings for each rostered player and stats who who plays with who and on which court and how they do together.  Contact me if interested in any of these reports.

USTA PNW weighs in on the new 40 & Over 1 singles / 3 doubles format - National exhibits bad math/statistics

I just received a communication from my League Coordinator regarding the format change for the 40 & Over league to use a 4-court 1 singles and 3 doubles format.

The gist of the communication is that the 4-court format is confirmed as being used in the Pacific Northwest section for 2020 leagues.  But it also provided a link to an FAQ that was prepared answering a number of questions.

Please do read the full FAQ if you are interested, but here are a few highlights.

First, a statement from National on why the change was provided.  Here that is:
Background: The regulation proposal 1.04A from the FL Section for the 40 & Over League “team match” championship format of one singles match and four doubles matches was tabled to gather further data. For the past 5 years, players have been surveyed as to the preferred 40 & Over League “team match” format. In the 2018 player survey, players were asked if they preferred 2S/3D or 1S/4D for this league and it revealed that 60% preferred 1S/4D. This survey did not offer the 1S/3D option. In April 2019, a survey was sent to players age 39 and over, asking which “team match” championship format they preferred, and results showed 25% for 1S/4D; 26% for 1S/3D; 39% for 2S/3D; ie. greater than 50% preferred only one singles match.  In May 2019, League program providers and volunteers were surveyed with 58% preferring 1S/3D. Section League Coordinators were asked their choice at the May 2019 Joint Staff meeting. Two SLCs preferred 2S/3D, while all others preferred 1S/3D.

Summary: With convincing input received from players, providers, and Section League Coordinators in favor of one singles match, we’ve learned that it’s time to consider a change in the 40 & Over League “team match” championship format. This will address the majority of our players’ preference for a one singles “team match” championship format and the challenges with limited court availability for many facilities, the magnitude of league offerings, and the complaints that teams have difficulty in recruiting singles players. The 40 & Over League program is the second largest program and represents just over 25% of the total participation this year. Improving the player experience for this population is critical to the continued success and growth of this program.
So there we have the official explanation of why the change was made.

I do find it interesting that of the three formats considered, the one that had by far the highest percentage, 2S/3D, at 39%, is the one not chosen.  Instead, they elected to effectively add together the percentages of the two other options since they were both offering one singles court and make the conclusion that most people wanted just one singles court.

The flaw in this logic is that while yes, the 1S/3D and 1S/4D both offer just one singles court and together had 51% preferring them, that isn't necessarily sound use of statistics.  It is simply not sound statistics to add together the results of two separate survey questions that themselves are composite questions, and conclude the common piece of the two questions is the reason everyone chose those options.  It is entirely possible that players chose one of those options not because of 1S, but for another reason.

For example:
  • Someone may have voted for 1S/4D because it would get 9 players in a match and explicitly didn't vote for 1S/3D because it would get just 7 in a match, so combining the options is not what they would have wanted.  Their second choice would have been 2S/3D as it gets 8 players in a match.
  • Someone may have voted for 1S/3D just because it was 4 courts, not because of 1S, and may have actually preferred a 2S/2D option for 4 courts, but that wasn't an option.  Had it been offered as an option, the survey would have been more equitable/sound rounding out the options, and the responses for 2S/2D and 2S/3D could be combined as both wanting two singles courts and very well been higher than the combined one singles options.
  • The majority of 40+ players already play doubles primarily, just 25% of players in a 2D/3D format are playing singles, so any survey of this player base is naturally going to be biased towards more doubles courts.  Then offer multiple options that have just one singles court and only one option that is has two singles and there is no way the two singles courts was going to have more support than the sum of the one singles options.  Not taking this into account in interpreting the results is a major shortcoming.

So saying that over 50% prefer one singles court is not a conclusion one can or should make based on the survey they did and its results.  For the two reasons I highlight above and perhaps others, it is entirely possible and likely that the number that actually prefer one singles court is well below 50%.

[added after initial post] If one were to apply the same bad logic the USTA did though, we should add together the 25% and 39% to show that 64% of players prefer five courts, and thus one of 2S/3D or 1S/4D should have been chosen.  Yet this wasn't done and we get the option, 1S/3D, that arguably has the least support.

In my opinion, the USTA has exhibited a pretty bad grasp of statistics and surveys in how they elected to perform and interpret the results of this survey.  I don't know that this is the case, but it almost feels like they did the survey and then worked to find a way to come to the conclusion they wanted.

Note that with nearly 200 responses, the survey I did simply asking if people prefer the old 2S/3D format or the new 1S/3D format overwhelming said the old 2S/3D format is preferred with 83% in support of it as of my writing this.

The other odd thing is that the three options together add up to 90% and not 100%.  Sure, some people may have abstained from answering this question, but the numbers not adding up makes one go hmmm.

Getting back to the PNW FAQ, it goes on to discuss a few other things that I raised as issues in my analysis of the rules changes:
  • Team match winners - The 4-court format can end in 2-2 ties, and while there are tie-breakers, PNW admits they won't break all ties and it says they are working on filling the gaps.  Kudos to PNW for working to address what National did not.
  • Singles opportunities - PNW acknowledges that opportunities for those over 40 to play singles is cut in half, and says they will work to find and promote other opportunities such as Flex leagues and tournaments.
  • Plus leagues - PNW also acknowledges the impact on plus leagues, especially with the change to only allow one plus player in 4-court matches.  They agree it is an issue on the opportunities for 4.5s to play singles and say they will work to do away with plus leagues in the future and offer an Open level for 5.0s and above in the future.

What do you think?  What are you hearing from your section regarding the change?  Will your local leagues follow this format and are steps being taken to mitigate the negatives that National did not address?

Wednesday, October 23, 2019

Simulating 2019 USTA League Nationals - 55 & Over 7.0 Women

As I did last year, I will be doing simulations of each Nationals to predict who the most likely four teams are to make the semi-finals, and also look at other interesting things that may occur.  Last year, the simulations did a pretty good job predicting who would advance to the semis including having all four teams predicted correctly several times.

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.

Nationals are approaching the fourth weekend of competition, the 55 & Over 7.0 women's event being held in Arizona.  Here is what the simulation says is likely to happen.

As a reminder, my simulations are done using my Estimated Dynamic NTRP Ratings and looking at the average rating for the top group of players on each team, the actual schedule each team will play, and then doing a million simulations of the matches with some random variation in each team's expected result.

First, the schedule with 17 teams will always allow for five teams to be undefeated and one sent home (2% chance), but while a schedule can be done that precludes it, it is possible there could be six undefeated, although very unlikely.  Four undefeated would be a nice and tidy result with no drama, but there is just a 13% chance of that so we are more than likely to have some tie-break scenarios with a 34% chance of three undefeated.

That leaves a very good chance, 85%, that there will be a tie for the last spot and it come down to tie-breakers.  That tie is likely at a 3-1 record.  The most likely size of the tie is four at 29%, but five is nearly as likely at 25%, three is still 21%, and a 6-way tie just 13%.  The chances of larger ties diminish pretty quickly, but an 11-way tie is theoretically possible, but it appears a multi-way tie is a very high probability and it could be reasonably large.

Should there be a tie on team record, it comes down first to who has the best court record, then head-to-head (if applicable), then to who lost the fewest sets, then who lost the fewest games, and finally percent of games won.  There is a 39% chance it comes down to the sets lost tie-breaker, and there is a 52% chance that is between two teams and a healthy 35% chance it is between three teams,  If the controversial tie-breakers are going to come into play, this is the likely spot.

The schedule strengths do vary a lot, the team with the easiest schedule having an opponent average of 3.51 while the team with the toughest schedule having an opponent average of 3.68.  This leads to an easier road for some teams than others, one of the predicted semi-finalists getting there due to an easier schedule.

So who is most likely to come out in the top-4?  Mid-Atlantic, EasternPNW, and Hawaii are solid choices, Mid-Atlantic the most likely.  Southwest, IntermoutainSouthern, and Texas will step in if someone slips up.

For those interested, I offer a variety of reports to make Nationals more fun and help captains prepare.  I have a Simulation Report that has all of the details of the simulation including the average ratings for each team, each team's schedule strength, the most likely record for each team, and the chance of each possible record for each team.  I also offer reports to help teams scout opponents in more detail, both a Flight Report with full roster averages, top-8 averages and played by court averages for each team, as well as full Team Reports with detailed ratings for each rostered player and stats who who plays with who and on which court and how they do together.  Contact me if interested in any of these reports.

Monday, October 21, 2019

Simulating 2019 USTA League Nationals - 55 & Over 7.0 Men

As I did last year, I will be doing simulations of each Nationals to predict who the most likely four teams are to make the semi-finals, and also look at other interesting things that may occur.  Last year, the simulations did a pretty good job predicting who would advance to the semis including having all four teams predicted correctly several times.

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.

Nationals are approaching the fourth weekend of competition, the 55 & Over 7.0 men's event being held in Arizona.  Here is what the simulation says is likely to happen.

As a reminder, my simulations are done using my Estimated Dynamic NTRP Ratings and looking at the average rating for the top group of players on each team, the actual schedule each team will play, and then doing a million simulations of the matches with some random variation in each team's expected result.

First, the schedule with 17 teams will always allow for five teams to be undefeated and one sent home (1% chance), but it is possible there could be six undefeated, although very unlikely.  Four undefeated would be a nice and tidy result with no drama, but there is just a 13% chance of that so we are more than likely to have some tie-break scenarios with a 41% chance of three undefeated.

That leaves a very good chance, 90%, that there will be a tie for the last spot and it come down to tie-breakers.  That tie is likely at a 3-1 record.  The most likely size of the tie is five at 32%, but four is nearly as likely at 27%, three and six is still 24%, and a 2-way tie just 1%.  The chances of larger ties diminish pretty quickly, but an 11-way tie is theoretically possible, but it appears a multi-way tie is a very high probability and it could be reasonably large.

Should there be a tie on team record, it comes down first to who has the best court record, then head-to-head (if applicable), then to who lost the fewest sets, then who lost the fewest games, and finally percent of games won.  There is a 50% chance it comes down to the sets lost tie-breaker, and there is a 40% chance that is between two teams and a healthy 38% chance it is between three teams,  If the controversial tie-breakers are going to come into play, this is the likely spot.

The schedule strengths do vary a lot, the team with the easiest schedule having an opponent average of 3.51 while the team with the toughest schedule having an opponent average of 3.68.  This leads to an easier road for some teams than others, one of the predicted semi-finalists getting there due to an easier schedule.

So who is most likely to come out in the top-4?  SoCal, Northern, and Texas are solid choices, then it is nearly a toss-up for the last spot between Florida, NorCalPNWMiddle States. and MoValley.

For those interested, I offer a variety of reports to make Nationals more fun and help captains prepare.  I have a Simulation Report that has all of the details of the simulation including the average ratings for each team, each team's schedule strength, the most likely record for each team, and the chance of each possible record for each team.  I also offer reports to help teams scout opponents in more detail, both a Flight Report with full roster averages, top-8 averages and played by court averages for each team, as well as full Team Reports with detailed ratings for each rostered player and stats who who plays with who and on which court and how they do together.  Contact me if interested in any of these reports.

Simulating 2019 USTA League Nationals - 40 & Over 4.5+ Men

As I did last year, I will be doing simulations of each Nationals to predict who the most likely four teams are to make the semi-finals, and also look at other interesting things that may occur.  Last year, the simulations did a pretty good job predicting who would advance to the semis including having all four teams predicted correctly several times.

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.

Nationals are approaching the third weekend of competition, the 40 & Over 4.5+ men's event being held in Orlando.  Here is what the simulation says is likely to happen.

As a reminder, my simulations are done using my Estimated Dynamic NTRP Ratings and looking at the average rating for the top group of players on each team, the actual schedule each team will play, and then doing a million simulations of the matches with some random variation in each team's expected result.

First, Hawaii is not present so there are 16 teams playing a random four opponents and there are still scenarios where five teams are undefeated, but less than 1% of them.  Four undefeated would be a nice and tidy result with no drama, but there is just a 6% chance of that so we are more than likely to have some tie-break scenarios with a 30% chance of three undefeated.

That leaves a very good chance, 86%, that there will be a tie for the last spot and it come down to tie-breakers.  That tie is likely at a 3-1 record.  The most likely size of the tie is a three or four at 27%, but five has a 19% chance, two 14%, and six 9%.  The chances of larger ties diminish pretty quickly, but an 11-way tie is theoretically possible, but it appears a multi-way tie is a very high probability and it could be reasonably large.

Should there be a tie on team record, it comes down first to who has the best court record, then head-to-head (if applicable), then to who lost the fewest sets, then who lost the fewest games, and finally percent of games won.  There is a 24% chance it comes down to the sets lost tie-breaker, and there is a 70% chance that is between two teams and a 25% chance it is between three teams,  If the controversial tie-breakers are going to come into play, this is the likely spot.

The schedule strengths do vary a bit, the team with the easiest schedule having an opponent average of 4.51 while the team with the toughest schedule having an opponent average of 4.65.  This leads to an easier road for some teams than others, one of the predicted semi-finalists getting there due to an easier schedule.

So who is most likely to come out in the top-4?  Southern, NorCalMid-Atlantic, and PNW, Southern the most likely.  Several others are in a group waiting for one of those to slip up including SoCal, Northern, Eastern, MoValley, Middle States, and Florida.

For those interested, I offer a variety of reports to make Nationals more fun and help captains prepare.  I have a Simulation Report that has all of the details of the simulation including the average ratings for each team, each team's schedule strength, the most likely record for each team, and the chance of each possible record for each team.  I also offer reports to help teams scout opponents in more detail, both a Flight Report with full roster averages, top-8 averages and played by court averages for each team, as well as full Team Reports with detailed ratings for each rostered player and stats who who plays with who and on which court and how they do together.  Contact me if interested in any of these reports.

Sunday, October 20, 2019

Weather Report - 2019 USTA League Nationals Week 4 - Arizona and Orlando

Week 3 of 2019 USTA League Nationals is complete and so we look forward to week 4.

There are just two events this coming weekend, the 55 & Over 7.0 in Surprise, AZ, and The 40 & Over 4.5+ in Orlando.

As usual, we need to do a weather check, especially given the rain, delays, and resulting short format that occurred for the 40 & Over 3.0 men and women in Orlando this weekend.

The weather has been near perfect in Vegas and Arizona the first three weeks, and it continues this weekend in Surprise.


Sunny, no rain, 80s, little wind, should be good.

Orlando on the other hand is not so good.  It is tropical storm season which we saw from this weekend, and the forecast for this coming weekend still calls for rain.


It appears likely there will be rain and some delays, matches potentially moved to the few indoor courts, and short format matches.  Forecast can change, so let's hope it improves, but why keep playing events where it is likely to rain?

Stay tuned for simulations.

More 2019 USTA League National Champions Crowned - Week 3 - A 9-way tie!

The third weekend of 2019 USTA League Nationals is complete and after the busiest weekend on the schedule, eight more champions have been crowned.

This weekend found action in both 18 & Over finishing that division up, and 40 & Over in the middle week of their three weeks of play.

The 18 & Over 4.0 women played in Las Vegas and Pacific Northwest and Southwest faced off in the final as the top-2 seeds with PNW taking the win 4-1.

The 18 & Over 4.0 men were also in Vegas and part way through day 2, it looked like there may be an undefeated team sent home, but in the end four teams were 4-0 and the final had the simulation favorites Intermountain and Texas in it and Intermountain taking it 4-1.

The 18 & Over 4.5 women played in Arizona with Southern Cal and Middle States in the final, SoCal winning 3-2.

The 18 & Over 4.5 men were also in Arizona, and this event looked like an undefeated Intermountain may be sent home, but they lost their last match 3-2.  They needed to win 5-0 to have a chance, I almost wonder if as soon as they lost a court they gave up and didn't want to be 4-0 and be sent home.  In any case, Mid-Atlantic and Southern were in the final and Mid-Atlantic won 3-2.

The 40 & Over division had the 3.0 men and women played in Orlando, and despite some rain, rescheduling, and short format matches, the finished up on time.

The women saw Midwest and Southern in the final, Southern taking the title 3-2.

The men had Caribbean and Southern in the final, here Caribbean took the title 4-1.

The 40 & Over 3.5 women were in Oklahoma City and there was a 9-way tie for first at 3-1!  The final had Southern and Mid-Atlantic, both predicted to make the semis, and Mid-Atlantic took the title 4-1.

The 40 & Over 3.5 men also in OKC had Texas and Pacific Northwest in the final, both predicted to make the semis, with Texas taking the win 3-2.

Stay tuned for previews of week 4 soon.

Saturday, October 19, 2019

9-way tie for first at 40 & Over 3.5 women USTA League Nationals!

When I do my simulations, I always include the chances of various sizes of the ties that may occur, and mention how large the tie theoretically could be.  Ties of 2-5 generally have high probabilities, 6 or more, even as many as 10 or 11, show up as possible but not likely.

Well, for the 40 & Over 3.5 women being played in Oklahoma City this weekend, one of those really large ties finally happened!  And it wasn't just for an advancing spot, but for first!

The standings show a whopping 9 teams tied at 3-1.  Here is how they all lined up:

  • Southern - 15-5 on courts, 31-10 on sets, lost to Texas
  • Mid-Atlantic - 15-5 on courts, 32-12 on sets, lost to Middle States
  • Midwest - 13-7 on courts, 27-18 on sets, 191-169 on games, lost to Mid-Atlantic
  • Florida - 13-7 on courts, 27-18 on sets, 192-174 on games, lost to Southern
  • Texas - 12-8 on courts, 27-18 on sets, lost to Midwest
  • MoValley - 12-8 on courts, 26-22 on sets, 184-175 on games, lost to Southwest
  • Intermountain - 12-8 on courts, 25-22 on sets, 194-195 on games, lost to Texas
  • Southwest - 10-10 on courts, 23-23 on sets, 180-189 on games, lost to Southern
  • Eastern - 10-10 on courts, 25-25 on sets, 201-190 on games, lost to Intermountain

Whew, that is a lot of teams!

Luckily for the USTA, the top-4 were decided just looking at court records.   No controversy in who the top-4 are.  And the order after that is largely inconsequential, but there are a few things to comment on.

The top-2 were decided on Southern losing fewer sets, which can be problematic as it ignores sets won, but had my preferred set differential or percent of sets won been used, the order would have been the same.

Midwest and Florida were tied on courts and sets and it was decided on Midwest losing fewer games.  This too can be problematic but my preferred games differential or percent of games won would have been the same in this case.

Texas, MoValley, and Intermountain were tied on courts at 12-8 and Texas got the spot on fewest sets lost.  MoValley and Intermountain both lost 22 sets, and using my rules MoValley should have gotten the spot due to winning more sets, but the USTA doesn't do that and instead goes to games lost and MoValley still got it there.

Southwest and Eastern were both just 10-10, and Southwest gets the spot losing 23 sets, but they also lost as many sets 23, and Eastern lost 25 but won 25.  It is unclear to me why Southwest's record is better based on this, it seems tied to me.  If the USTA did consider them tied it would go to games lost and Southwest would get that too as the USTA ignores games won at this point where Eastern won 11 more games than the list, while Southwest lost 9 more than they won.  I would give the spot to Eastern, but the USTA gives it to Southwest.

So only one example of the USTA's broken tie-breakers getting the order wrong, and luckily it was for 8th vs 9th so no one cares, but at some point, like happened last year, the flawed tie-breakers will get the wrong team advancing to the semis again.

Note, that head-to-head isn't mentioned above as it only comes into play if the teams all played each other.  In this case, court records between teams that played were different, except for Texas and Intermountain, but MoValley didn't also play both teams so I don't think it was used.  But the team that won, Texas, ended up ahead of the team that lost, Intermountain.

But you could make the argument that that Texas and Southern, the only two teams to give two of the tied teams their losses, were the most deserving given their two wins each over 3-1 teams.  And given Texas' loss was to a good Midwest team and beat the top-seed Southern, Texas had a tough schedule and perhaps got a raw deal finishing 5th.

Weather update: Week 3 of USTA League Nationals in Orlando

The rain arrived in Orlando, and the alternate plans had to be implemented.

The forecast was for rain much of the day and so matches were moved into the handful of indoor courts at the National Campus, and the format was changed to an abbreviated Fast Four format.

As I understand it, sets were played to 4, playing a tie-break at 3-3, and if sets were split, a super tie-break was played.  This is probably the least evil of the alternate formats, others being pro-sets to 8, 6, or even 4.  At least this format still has three sets so a slow start isn't necessarily a nail in the coffin so to speak.

Now, even with this, since there are limited indoor courts, I understand some round-robin matches are still planned to be played Sunday morning.  The current forecast seems to say the rain has moved out of the Orlando area so I'd think there is a chance play could commence outside, but I've gotten no indication that is in the cards.

Kudos to the USTA for having a plan to accelerate matches on Friday and play at least a semi normal format today, but I still contend there are sites where there is less chance of disruption like this that could have been selected.

I also heard that Oklahoma City has been going find.  A bit windy as it was last weekend, but still playable.

Part 2 of will an undefeated team be sent home from 2019 USTA League Nationals? 18 & Over 4.5 men are as likely as ever, but 4.0 won't happen this year

I wrote earlier that the 18 & Over 4.0 men had a decent chance of having five teams undefeated and one being sent home.  Well, Middle States went and lost their first match today 3-2 to New England, so that is out the door.

But, upon reviewing the 18 & Over 4.5 men, there is nearly as good a chance of it happening there, and this is even with some matches played from day 2 already.

Right now the simulation says an 11% chance, and it would happen with Mid-Atlantic, Midwest, Intermountain, Middle States, and Southern.  Four of those teams are already 3-0, and Mid-Atlantic has two weaker opponents in Northern and Hawaii so they are likely to be 4-0.

What are the potential slip-ups?

  • Midwest plays a good Missouri Valley team but should still win, especially if MoValley at 1-2 mails it in
  • Intermountain plays a good Southwest team but should still win, especially if Southwest at 1-2 mails it in
  • Middle States plays a very good (on paper) Florida team, but they are 0-3 and haven't played three of their top-11 players so are perhaps short staffed, Middle States could win this
  • Southern plays Texas, a very good team that is 2-1 and just happened to have to play Midwest earlier.  Texas will still feel in it so this is perhaps the biggest risk.
  • Mid-Atlantic should win both of theirs, Northern is the tougher match.

So look for the Middle States vs Florida, and Southern vs Texas matches.  That will likely determine if we get five undefeated.

Will an undefeated team be sent home from 2019 USTA League Nationals? 18 & Over 4.0 men are as likely as ever

With the new format for Nationals started in 2018 and continued this year, it has always been possible for five teams to finish undefeated, and then one of them would not make the semi-finals and be sent home.  In fact, because the USTA does create an optimal schedule, there has sometimes been the theoretical chance of six undefeated and two being sent home!

Now, the chances have always been quite slim, typically less than 1% and even in the more probably cases, low single digit chances.  And it has not happened to date.  But when there is even a 1% chance of something happening, take enough samples and it will happen.

After day 1 of the 18 & Over 4.0 men's event this weekend, it looks like we are seeing the best chance yet of this controversial finish.

As I write this, six teams are all 2-0: Eastern, Southern Cal, Intermountain, Texas, Middle States, and Midwest.  Looking at the remaining schedule, only Eastern and Midwest play each other, so it is entirely possible, and perhaps probable, that five finish undefeated!

Now, my simulation said going into the event that there was a 4% chance of there being five undefeated, about as high as we've ever seen.  As we enter day 2, the simulation says it is now a 12% chance.

Why isn't it higher?

First, there are some strong teams outside of the six 2-0 teams that must still be played, and tennis is a sport where match-ups matter, and who actually plays matters too, they are not simply won by the team with the higher top-11 average.  "Anything can happen", and for it to happen here, five teams would have to all go 2-0, just one slip up and it doesn't happen.

Second, the teams may not all even be favored to win.  I already noted Eastern and Midwest play so one will lose, but Eastern also has to play Florida who is a strong team and is only 0-2 because they had a brutal schedule of playing Middle States and Texas and lost both 3-2.

However, it appears Florida has not played two of their stronger players, perhaps they were unavailable and couldn't make the trip, and this would mean they aren't quite as strong and Eastern very well may be favored.  If this is the case, the chances of five undefeated may go up significantly.

If it happens, then it comes down to tie-breakers, and as I've written before, they are somewhat broken and may not identify the most deserving team.  The irony of the USTA not adopting my suggested regulations change to fix the tie-breakers and then having them send the wrong undefeated team home would be classic.

Will it happen?  We'll know in a few hours!

Note: Middle States was upset 3-2 so this did not happen, but the 18 & Over 4.5 men may still finish with five undefeated.

Friday, October 18, 2019

Simulating 2019 USTA League Nationals - 40 & Over 3.0 Women

As I did last year, I will be doing simulations of each Nationals to predict who the most likely four teams are to make the semi-finals, and also look at other interesting things that may occur.  Last year, the simulations did a pretty good job predicting who would advance to the semis including having all four teams predicted correctly several times.

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.

Nationals are approaching the third weekend of competition, the 40 & Over 3.0 women's event being held in Orlando.  Here is what the simulation says is likely to happen.

As a reminder, my simulations are done using my Estimated Dynamic NTRP Ratings and looking at the average rating for the top group of players on each team, the actual schedule each team will play, and then doing a million simulations of the matches with some random variation in each team's expected result.

First, with 17 teams playing a random four opponents, it will always be possible for there to be five (or more) undefeated teams.  It is possible to limit it to five with the right schedule, and the USTA did that this time around and there is less than a 1% chance of even five undefeated.

Four undefeated would be a nice and tidy result with no drama, and there is a healthy 32% chance of that.

That still leaves a good chance, 67%, that there will be a tie for the last spot and it come down to tie-breakers.  That tie is likely at a 3-1 record.  The most likely size of the tie is a three or four, both at 32%, but five has a 19% chance and two 9%.  The chances of larger ties diminish pretty quickly, but a 10-way tie is theoretically possible, but it appears a multi-way tie is a very high probability and it could be reasonably large.

Should there be a tie on team record, it comes down first to who has the best court record, then head-to-head (if applicable), then to who lost the fewest sets, then who lost the fewest games, and finally percent of games won.  There is a 15% chance it comes down to the sets lost tie-breaker, and there is a 81% chance that is between two teams and a 17% chance it is between three teams,  If the controversial tie-breakers are going to come into play, this is the likely spot.

The schedule strengths do vary a lot, the team with the easiest schedule having an opponent average of 2.97 while the team with the toughest schedule having an opponent average of 3.16.  This leads to an easier road for some teams than others, one of the predicted semi-finalists getting there due to an easier schedule.

So who is most likely to come out in the top-4?  Southern, NorCalMiddle States, and SoCal, very close between the first three he most likely.  Midwest is the main spoiler, but Mid-Atlantic, MoValley, and Eastern are lurking..

For those interested, I offer a variety of reports to make Nationals more fun and help captains prepare.  I have a Simulation Report that has all of the details of the simulation including the average ratings for each team, each team's schedule strength, the most likely record for each team, and the chance of each possible record for each team.  I also offer reports to help teams scout opponents in more detail, both a Flight Report with full roster averages, top-8 averages and played by court averages for each team, as well as full Team Reports with detailed ratings for each rostered player and stats who who plays with who and on which court and how they do together.  Contact me if interested in any of these reports.

Thursday, October 17, 2019

First weekend in Orlando, and we have rain ... - Format changes already

USTA League Nationals enters its third weekend tomorrow, and the National Campus in Orlando is being used for the first time.  I wrote earlier in the week there was a chance of some Thunderstorms and rain over the weekend, and it appears that forecast has gotten worse.

With a tropical storm in the gulf, the forecast is now for rain on Saturday, perhaps quite a bit, and the rain appears to be coming in on Friday and perhaps hanging around Sunday morning.


I am hearing that the USTA is being proactive to try to get ahead of things by modifying the schedule to play more matches on Friday, and to use no-ad scoring to shorten matches a bit.  That will help, but will it be enough to avoid more drastic measures?

I bet all the players are wishing they were in Arizona or California or Las Vegas ...

Simulating 2019 USTA League Nationals - 40 & Over 3.5 Women

As I did last year, I will be doing simulations of each Nationals to predict who the most likely four teams are to make the semi-finals, and also look at other interesting things that may occur.  Last year, the simulations did a pretty good job predicting who would advance to the semis including having all four teams predicted correctly several times.

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.

Nationals are approaching the third weekend of competition, the 40 & Over 3.5 women's event being held in Oklahoma City.  Here is what the simulation says is likely to happen.

As a reminder, my simulations are done using my Estimated Dynamic NTRP Ratings and looking at the average rating for the top group of players on each team, the actual schedule each team will play, and then doing a million simulations of the matches with some random variation in each team's expected result.

First, with 17 teams playing a random four opponents, it will always be possible for there to be five (or more) undefeated teams.  It is possible to limit it to five with the right schedule, and the USTA did that this time around and there is less than a 1% chance of even five undefeated.

Four undefeated would be a nice and tidy result with no drama, but there is just a 16% chance of that.

That leaves a very good chance, 83%, that there will be a tie for the last spot and it come down to tie-breakers.  That tie is likely at a 3-1 record.  The most likely size of the tie is a good sized four at 29%, but three has a 28% chance, five 20%, and two 13%, and six 9%.  The chances of larger ties diminish pretty quickly, but an 11-way tie is theoretically possible, but it appears a multi-way tie is a very high probability and it could be reasonably large.

Should there be a tie on team record, it comes down first to who has the best court record, then head-to-head (if applicable), then to who lost the fewest sets, then who lost the fewest games, and finally percent of games won.  There is a 23% chance it comes down to the sets lost tie-breaker, and there is a 70% chance that is between two teams and a 25% chance it is between three teams,  If the controversial tie-breakers are going to come into play, this is the likely spot.

The schedule strengths do vary a lot, the team with the easiest schedule having an opponent average of 3.46 while the team with the toughest schedule having an opponent average of 3.65.  This leads to an easier road for some teams than others, one of the predicted semi-finalists getting there due to an easier schedule.

So who is most likely to come out in the top-4?  Southern, Missouri ValleyMid-Atlantic, and Eastern, Southern the most likely.  Intermountain is the main spoiler, but Midwest and SoCal have a shot.

For those interested, I offer a variety of reports to make Nationals more fun and help captains prepare.  I have a Simulation Report that has all of the details of the simulation including the average ratings for each team, each team's schedule strength, the most likely record for each team, and the chance of each possible record for each team.  I also offer reports to help teams scout opponents in more detail, both a Flight Report with full roster averages, top-8 averages and played by court averages for each team, as well as full Team Reports with detailed ratings for each rostered player and stats who who plays with who and on which court and how they do together.  Contact me if interested in any of these reports.

Tuesday, October 15, 2019

Simulating 2019 USTA League Nationals - 40 & Over 3.5 Men

As I did last year, I will be doing simulations of each Nationals to predict who the most likely four teams are to make the semi-finals, and also look at other interesting things that may occur.  Last year, the simulations did a pretty good job predicting who would advance to the semis including having all four teams predicted correctly several times.

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.

Nationals are approaching the third weekend of competition, the 40 & Over 3.5 men's event being held in Oklahoma City.  Here is what the simulation says is likely to happen.

As a reminder, my simulations are done using my Estimated Dynamic NTRP Ratings and looking at the average rating for the top group of players on each team, the actual schedule each team will play, and then doing a million simulations of the matches with some random variation in each team's expected result.

First, with 17 teams playing a random four opponents, it will always be possible for there to be five (or more) undefeated teams.  It is possible to limit it to five with the right schedule, but the USTA did not do that this time around and over 127 of the million simulations had six teams undefeated.  The chances of five is not much higher at just under 1% and four is only 7%.

Four undefeated would be a nice and tidy result with no drama, but there is a 28% chance of three so we are more than likely to have some tie-break scenarios.

That leaves a very good chance, 91%, that there will be a tie for the last spot and it come down to tie-breakers.  That tie is likely at a 3-1 record.  The most likely size of the tie is a good sized four at 28%, but five has a 25% chance, three 21%, and six 13%.  The chances of larger ties diminish pretty quickly, but an 11-way tie is theoretically possible, but it appears a multi-way tie is a very high probability and it could be reasonably large.

Should there be a tie on team record, it comes down first to who has the best court record, then head-to-head (if applicable), then to who lost the fewest sets, then who lost the fewest games, and finally percent of games won.  There is a 28% chance it comes down to the sets lost tie-breaker, and there is a 67% chance that is between two teams and a 28% chance it is between three teams,  If the controversial tie-breakers are going to come into play, this is the likely spot.

The schedule strengths do vary a bit, the team with the easiest schedule having an opponent average of 3.48 while the team with the toughest schedule having an opponent average of 3.60.  This leads to an easier road for some teams than others, one of the predicted semi-finalists getting there due to an easier schedule.

So who is most likely to come out in the top-4?  Texas, FloridaPNW, and Eastern, Texas the most likely.  There is a big group close behind including NorCal, New England, SoCal, Mid-Atlantic, and Missouri Valley.

For those interested, I offer a variety of reports to make Nationals more fun and help captains prepare.  I have a Simulation Report that has all of the details of the simulation including the average ratings for each team, each team's schedule strength, the most likely record for each team, and the chance of each possible record for each team.  I also offer reports to help teams scout opponents in more detail, both a Flight Report with full roster averages, top-8 averages and played by court averages for each team, as well as full Team Reports with detailed ratings for each rostered player and stats who who plays with who and on which court and how they do together.  Contact me if interested in any of these reports.