Sunday, March 27, 2016

What if the new no early start rating rule had been used for 2016 early start leagues? Two-thirds or more of teams may have had line-ups affected!

Happy Easter!

I wrote a week and a half ago that the USTA is apparently doing away with early start ratings for 2017 early start leagues.  Each section/district must now decide how to apply the rule, specifically when players that are bumped at at year-end become ineligible.  But regardless of what each section decides, bumped up players will not be eligible for Nationals.

So I thought it would be interesting to look at some statistics on how many players played early start leagues and were bumped up, as these players and the teams these players play on would be affected by the new rule.

To start, I looked at the 18 & over and 40 & over divisions.  Specifically, I looked at how many players played in 2016 early start leagues and were bumped up from 2014 year-end to 2015 year-end.  Under the new rule, unless the players were playing up in the early start league, they would become ineligible to play on their team at some point.

The total number of players affected would be 3,536 and the total number of teams would be 3,396.  Keep in mind that not all sections/areas have early start leagues, or don't have them in the 18 & over and/or 40 & over divisions.  So what if we look at it by section?

Here Southern leads the way:

  • Southern - 1,597 players / 1,889 teams
  • Texas - 685 / 587
  • Midwest - 336 / 268
  • New England - 269 / 219
  • Missouri Valley - 203 / 151
  • Southwest - 188 / 143
  • Florida - 118 / 89
  • Northern - 88 / 66
  • Eastern - 46 / 23
  • NorCal - 10 / 11

Note that in some areas, notably in Southern, there are multiple early start seasons so that is why there can be more teams affected than there are players.  Note also that I'm not sure all of the Texas teams here play in advancing leagues, some may be Fall leagues that don't advance.

The next question is, how does Southern break out into states?
  • Georgia - 1,238 players / 1,601 teams
  • Alabama - 186 / 185
  • Tennessee - 174 / 103

Georgia is the king of early start with 2017 leagues starting soon and another set in the Summer and Fall, and so they will clearly be affected by the new rule.  And the same happened in 2015 for 2016 championship year teams.

For reference, I count the following number of teams for the above states that played 2016 early start leagues in 2015:
  • Georgia - 2,680 - 60% of teams affected
  • Alabama - 325 - 57%
  • Tennessee - 157 - 66%

Those are some significant percentages of teams affected.  Since the players affected are those being bumped up, there is even a higher chance that a team winning States or Sectionals will have one of these players, perhaps multiple, and be affected.

With early start, players on their way up might be early start bumped so they are on a team at the "right" level.  With the new rule though, they will be able to play on these early start teams at the lower level, but then if/when they get bumped up at year-end, these players are no longer eligible for Nationals and potentially not for State or Sectional playoffs.  This could be a significant impact to a team's ability to field a competitive line-up, or a line-up at all.  How Southern and/or Georgia decide to handle this will determine when/where that impact is.

I always thought early start ratings served a purpose and did a reasonable job at getting players at the right level rather than allowing players to play at a level too low based on their results.  But apparently enough of the sections of the USTA thought they were confusing or too hard to managed and decided to do away with them.

We'll see how it works.  But what do you think?  Will it be better or worse with the new rule?

Sunday, March 20, 2016

Now that early start ratings are no more, how will sections handle year-end bumps for early start players?

I wrote last week about the end of Early Start Ratings for 2017 leagues.  Without them, players will play in 2017 early start leagues using their 2015 year-end rating level rather than having early start bumps up or down determining what level they can play at in the early start league.

This then brings the interesting question of how to handle those players that are bumped up at 2016 year-end should their team advance to playoffs that are played in 2017.  Do they get to continue playing at the lower level?  Or must they play at the higher level, and thus possibly be ineligible to play in playoffs?

USTA National has established that players at 2017 Nationals must play using their 2016 year-end rating level.  But they have apparently left it up to each section to determine if they will follow this rule for District/State/Sectional playoffs, or if instead they will allow players to continue playing at their 2015 year-end rating.

There are pros and cons to each approach.  Let's look at a scenario to consider them.

Say an 2017 early start league starts in July 2016.  A 2015 year-end 3.5 will be eligible to play as a 3.5, even if their Spring league play was outstanding and their dynamic rating is well into the range for a 4.0 at this point.  Assuming the league has any local or flight playoffs prior to December 2016 when year-end ratings come out, they will be able to play in those playoffs as a 3.5 as well.

Where things depend on a section's choice of option is after this.  Say the player is bumped up to 4.0 at 2016 year-end and there are district or state playoffs in early 2017 that the player's 3.5 team advances to.  Can the they still play as a 3.5 on this team?

If the section elects to follow the National rule, the player would not be eligible to play in playoffs as a 3.5 as they are a 2016 year-end 4.0.  If several players on a team fall into this camp, it could significantly affect the make-up of the team, and potentially even their ability to field at team if too many players were bumped up.  This seems particularly punishing on the surface, a team that followed the rules and qualified for States/Districts/Sectionals now can't field a team or bring their key players.

So a section could elect to implement the other option and let the bumped up player's continue to play on their team at the lower (3.5) level.  The problem with this is, if the team wins their section, all that has been done is to defer the issue as the player's will not be able to play at Nationals as a 3.5 and now they may not be able to field a team, or if they can, it won't be a competitive team.  This means a section may be sending a team to Nationals that is somewhat handicapped and perhaps not as strong as a team that was beat at Sectionals that would have had all their players eligible.  The section isn't being represented as well and the losing team at Sectionals will feel they were gipped out of a trip to Nationals.

There is no easy answer, and the point of early start ratings was to try to address this somewhat, but I'd probably lean towards having sections follow the National rule and not letting bumped up players continue to play at a lower level.  While this is going to affect teams earlier, it keeps the play in playoffs more fair and consistent with what teams will have to abide by should they advance to Nationals.  It seems worse to have teams at Nationals completely different from what they were to get there.

What do you think?  Whatever it is, you may want to contact your League Coordinator to make your voice heard as the section may not have made their decision yet.

Note: All of the above is based on my understanding from what I seen or been told, but has not to my knowledge been documented in any 2017 regulations document.  Until that happens, treat the above as likely how things will play out in such a scenario but not for sure what will happen.

Wednesday, March 16, 2016

The USTA does away with Early Start Ratings for 2017

In the past couple of days, I have learned and confirmed that for 2017 the USTA will not be publishing/using Early Start Ratings for Early Start Leagues.  Instead, a player's current NTRP level from 2015 year-end (or current self-rating) will be used for these leagues.

But let's back up for a minute for those who are wondering what Early Start Ratings and Leagues are.  See this post I wrote detailing them, but the quick summary is as follows.

In some sections there are leagues that start early for the following year due to court availability, avoiding overlapping seasons, or just to provide more playing opportunities for players.  The challenge with this is what level should a player register at?  A player that has improved and will be bumped up from say a 3.5 to 4.0 at year-end probably shouldn't be able to still play as a 3.5 for a "next year" league when they will be a 4.0, but year-end ratings aren't out yet.

So Early Start Ratings were published to try to reflect these improved (or going down) player's ratings so they would be playing in this "next year" league at the right level.  In some districts, specifically Georgia in the Southern section, they have Early Start Leagues for 2017 starting in the next month or so, while most other sections that have early start leagues start in the Fall.

And what happened is that Georgia, because their early start leagues are starting right away, just published information about this regulation change, stating "Early Start League Ratings will no longer be published (March and/or August), which means players will always register with their most recent valid computer year-end rating or their current self-rating".  I subsequently confirmed from someone who attended the recent USTA Annual Meeting that this is in fact a change for all sections and there will not be Early Start Ratings at all for 2017 Early Start Leagues.

So what does this mean to you?

If you play in a section/district where you have an Early Start League, and there are a bunch of them, you will register for team in this league at your current (2015 year-end) rating.  No more waiting for Early Start Ratings to come out and the rapid assembling of teams that take into account who was early start bumped up or down.  This simplifies things and lets teams plan farther in advance and not have to react in just a few weeks, but it also means players in these leagues may be at an advantage or disadvantage that they weren't before.

For example, a player that is improving and is well on their way to being bumped up from 3.5 to 4.0 and would have had an Early Start Rating of 4.0 will no longer have to play as a 4.0 but can continue to play as a 3.5 in a 2017 Early Start League.  Similarly, a player that has had a bad year and is on their way down from 4.0 to 3.5 will no longer get to play as a 3.5 but must continue to play as a 4.0 in the 2017 Early Start League.

Some will argue this isn't fair in either case, but I'm sure there were complaints or concerns about Early Start Ratings.  Or perhaps players and captains were simply confused by Early Start Ratings and that is why the USTA has elected to discontinue them.

So what happens to the protection Early Start Ratings gave us against clearly out of level players that are bumped up at year-end getting to continue to play as a lower level player?  There are a few things.

First, each section is given the option of whether to allow these players to continue to play at the lower level in playoffs/Districts/States/Sectionals or if they must play at their new year-end level.  This is more or less the same as it has been in the past, but some sections may elect to change which option they choose to implement.  Sectionals that want to focus on "fair play" at the expense of disrupting rosters can elect to not allow these players to advance.

However, and this has been the existing regulation as well, if a player exceeds the "clearly above level" threshold, my understanding is that even if the section chooses to let the player continue into playoffs, reaching this threshold will preclude them from doing so.  This does keep players that are way too good for the lower level from being able to play there.

Second, I've been told that even if a section allows a player to advance through playoffs/Sectionals playing at the lower level, they will not be able to play at Nationals at the lower level.  Nationals players must play at their 2016 year-end level.

I'm sure there will be more to come as the other sections review the new regulation and options and decide how they will implement it and communicate that, and I'll try to keep everyone up to date on what I hear, but I have to say I'll kind of miss Early Start Ratings and the excitement of getting a preview of what player's year-end levels might be.

But if you are still wanting to get a preview, my Estimated Dynamic NTRP Rating Reports are a good way to get an idea of where you stand.  Contact me to learn more or request a report.

Saturday, March 5, 2016

How many games does a team winning 5-0 lose? Interesting tennis league stats

I've written about how often a winning team wins 5-0 vs 4-1 vs 3-2, but I came across a match where a team not only won 5-0, but only lost 12 games, and got curious how often that or similar scenarios occur.

So I went about looking at all matches in 18 & over and 40 & over in 2015 where the winning team won 5-0 and are matches were completed, e.g. no defaults or retirements where the score would not represent a complete match, and did it for both Men and Women, and here is the result.

First the men:
We see that losing just 12 games happened 15 times, so certainly not common, but some teams gave up even fewer, the fewest being just 6 games which happened once.  So losing just 12 is impressive and in the 98th percentile, but not the best that has occurred.

We also see that the mid-point is right around 30 with a big chunk between 25 and 35 which makes sense.  Losing a total of 30 games over 5 matches would mean on average winning 6-3,6-3.

At the other extreme we see a few at/over 50 games lost, the most being 56.  That is a pretty difficult task as that is over 11 games per court being lost without losing a court.  Note that this one was a match where 3rd sets were played out so that made it easier.  The most with 3rd set tie-breaks was 52 with 4 of the courts going to a 3rd set.

On to the women: 
This chart looks very similar, a standard bell curve as you'd expect, but there was a team that gave up 0 games in winning a match!  The match was in a 40+ 3.0 league in Texas and TennisLink shows all courts as completed so if the scores were entered right, it appears it happened.

The mid-point for the ladies seems to be centered slightly lower around 28, but on the high end there was a match where the winning team won every court but lost 60 games!  This was in Southern and yes, they did play out the 3rd sets.  The most where the 3rd set was a match tie-break was 54 and every court did go to a match tie-break.

What are you experiences with a team winning 5-0 and losing very few or a lot of games?

Friday, March 4, 2016

How do Novak, Fed, Rafa, and Murray's matches rate on bad days?

I've written lately about the highest match ratings and dynamic ratings that my Estimated Dynamic NTRP Ratings calculate for ATP pros.  But they don't always play at that level.  Just like us recreational players, they have bad days and low rated matches.

For most players that play USTA League, their high and low match rating during a year are typically about 0.4-0.5 apart.  For example a upper half 4.0 (a rating around 3.80) will likely have some match ratings just over 4.0 and others down around 3.5.  This happens naturally just due to good/bad days or good/bad match-ups or variations in how partners play etc.

But to ATP pros have the same variation?  Or do they even have more?

The charts I included in my comparison of some of Novak's and Federer's best seasons actually shows this fairly well.

In 2011, one of Novak's best years, he had a high match rating of 7.39 in the Australian Open final and a 7.31 in the US Open final, but he had some stinkers too.  He had a few 6.85's in closer than expected wins in the first half of the year, but then lost to Ferrer and Tipsarevic at the Tour Finals getting a 6.53 and 6.71.  That is a 0.86 difference from high to low!  Now, perhaps Novak just ran out of gas at the end of the year, but you can see the effect of that in the chart below.

Novak 2011

Federer had one of his best years in 2004 and had a 7.34 against Hewitt at Wimbledon and a 7.42 against him at the US Open.  But he also had a 6.56 in a loss to Costa in Rome and a 6.66 in a loss to Kuerten at the French.  These were on clay against clay court specialists so perhaps that explains the 0.86 difference, but he also had a 6.71 in a loss to Hrbaty in Cincinnati, one of his favorite tournaments.  And you can see quite a few lower rated matches in the chart below.

Federer 2004

What about Nadal?  He's had a few unexpected losses, the loss to Rosol in 2012 at Wimbledon garnered a 6.69 only weeks after a 7.49 in the French Open semi-final.  In 2013 he went from a 7.38 in the French final against Ferrer to a 6.67 losing to Darcis at Wimbledon.  He also had a 6.63 against Del Potro in Shanghai later in the year.  Then in 2015, while he did not have the highs he'd had in prior years, a 7.02 at the French against Sock the highest, he had numerous lows including a 6.33 t Berrer in Doha, a 6.46 losing to Berdych in Australia, a 6.44 losing to Dolgopolov in London, and a 6.50 losing to Dustin Brown at Wimbledon.

Nadal 2013

Andy Murray is not immune from this either.  In 2013 when he won Wimbledon with a 7.17 beating Novak, he had a 6.45 against Mahut at Queens Club, a 6.51 in Monte Carlo, and a 6.53 at the US Open, the last two against Wawrinka.

Murray 2013

A common theme in the lower rated matches is playing at tournaments or on surfaces that aren't the player's best, or when changing surfaces, but in some cases also playing a player that is a bad match-up for them or at a time when they are tired or fatigued.  This is exactly what happens with us in USTA League play too.  There are facilities or courts that we don't like, bad match-ups, being tired, etc.  The top pros are not immune from this and have bad matches too.

So take heart, if you have a bad match, you can rebound too and play well in your next one.

Attribution: Match data is courtesy Jeff Sackmann / Tennis Abstract.

Thursday, March 3, 2016

What are some of the highest rated ATP players using Estimated Dynamic NTRP Ratings? Jim Courier anyone?

I wrote yesterday about what some of the highest rated matches have been for ATP players using my Estimated Dynamic NTRP Ratings.  But individual matches can have extreme highs due to a really good day or opponent having a bad day.  So it can also be useful to look at what the highest dynamic ratings have been.

The dynamic rating is weighted very much to a player's most recent play, but it does effectively calculate an average of sorts of those most recent matches, so a solitary really good match amongst ordinary matches won't move it that much, but a sequence of good matches is required to attain the highest dynamic rating.

So, what are the highest dynamic ratings of recent years?  Here is the list, each player listed just once with their highest, and when it occurred.

  • Jim Courier - 7.40 - 1992 French Open Final
  • Bjorn Borg - 7.34 - 1980 June Davis Cup (right after French Open Final)
  • Sergi Bruguera - 7.34 - 1993 French Open Final
  • Rafael Nadal - 7.34 - 2008 French Open Final
  • Pete Sampras - 7.34 - 1997 Australian Open Final
  • Thomas Muster - 7.34 - 1995 French Open Final
  • Andre Agassi - 7.29 - 1994 US Open Final
  • Pat Rafter - 7.27 - 1997 US Open Final
  • Ivan Lendl - 7.26 - 1986 US Open Final
  • Yevgeny Kafelnikov - 7.26 - 1996 French Open Final
  • Roger Federer - 7.25 - 2004 Australian Open Semi-final
  • Mats Wilander - 7.25 - 1988 Australian Open Final
  • Guillermo Vilas - 7.25 - 1977 French Open Final
  • Stefan Edberg - 7.24 - 1993 Australian Open Semi-final
  • Juan Carlos Ferrero - 7.23 - 2001 French Open Quarter-final
  • John McEnroe - 7.21 - 1984 Philadelphia Quarter-final
  • Boris Becker - 7.21 - 1996 Australian Open Final
  • Novak Djokovic - 7.21 - 2011 Dubai R32 (right after Australian Open Final)
  • Lleyton Hewitt - 7.20 - 2001 Tokyo R32 (right after US Open Final)

It is somewhat remarkable to me that Djokovic is so low on this list.  We all perceive that he dominated in 2011 and 2015, but whether because many of his matches were closer than expected or his opponents were not that high rated, he never achieved the high dynamic ratings many before him did.  Similarly, McEnroe seems low given his dominating 82-3 1984, and Federer had many years where he was clearly above the competition.

It is also interesting to see how many of these bests were done at the end of the French Open and how many of those were pre-2000.  One could hypothesize that this was when there was a larger disparity between surfaces and clay was more unique, and so the clay court specialists were able to inflate their ratings by playing the non-clay court guys when their ratings were higher coming in to the tournament.  Perhaps I'm stretching but it is a thought.

But the above are just high points for each player, and it doesn't tell the story of if the high point was a flash in the pan or if they were actually able to sustain the level.  So here is a count of the times each player's dynamic rating was at or over 7.20.
  • Jim Courier - 20
  • Pete Sampras - 18
  • Rafael Nadal - 14
  • Bjorn Borg - 13
  • Ivan Lendl - 12
  • Sergi Bruguera - 11
  • Andre Agassi - 10
  • Pat Rafter - 7
  • Thomas Muster - 5
  • Mats Wilander - 5
  • Yevgeny Kafelnikov - 4
  • Roger Federer - 4
  • Stefan Edberg - 3
  • Juan Carlos Ferrero - 3
  • Boris Becker - 3
  • Novak Djokovic - 3
  • Guillermo Vilas - 2
  • Lleyton Hewitt - 2
  • John McEnroe - 1

Having Courier, Sampras, and Agassi all in the top-7 of this list and having them playing at the same time is amazing.  It wasn't just one of them cleaning up and raising their rating, but having to play each other drove their ratings up.

The era just before them, that also overlapped with them somewhat, fills the list too with Lendl, Borg, Rafter, Muster, Wilander, Kafelnikov, Edberg, and Becker.

The current era is not represented as much with just Nadal, Federer, Djokovic, and Hewitt, with most of those farther down on the list.

Now obviously dynamic NTRP ratings are not an ideal way to measure or rate players, so some of the above should not be taken too seriously, but it is fascinating to see certain players and eras stand out.  Were the 80's and 90's the golden era?  Has it been a weaker era since in the 2000's and 2010's?  Or it is just deeper now and harder to stand out and achieve very high ratings now?

Attribution: Match data is courtesy Jeff Sackmann / Tennis Abstract.

Wednesday, March 2, 2016

What are some of the highest rated ATP matches using Estimated Dynamic NTRP Ratings? Nadal leads the way.

I am now calculating my Estimated Dynamic NTRP ratings for ATP matches and have posted a few lists, ratings through the end of 2015 and then through the Australian Open.  And I also looked at some specific seasons for Djokovic and Federer.

The way the NTRP algorithm works is a rating is calculated for each match based on the score and the rating of the opponent, and then a new dynamic rating is calculated based on recent ratings.  Since a rating is calculated for each match, what is interesting is to look at which individual matches rate the highest.

I ran across a thread on Talk Tennis asking what various players greatest matches were at their peak.  Naturally, I thought it looking at the highest rated matches would be a good indicator of this.  Because the algorithm uses the score and strength of opponent, this should in theory be some valuable information.

So here goes using the Big-4 in their current ranking order.

Novak Djokovic has naturally had a lot of high rated matches, but here are some of his top ones:
  • 7.39 - 2011 Australian Open Final - 6-4,6-2,6-3 over Murray
  • 7.35 - 2013 Australian Open SF - 6-4,6-2,6-1 over Ferrer
  • 7.31 - 2011 US Open Final - 6-2,6-4,6-7,6-1 over Nadal
  • 7.27 - 2015 US Open SF - 6-0,6-1,6-2 over Cilic
  • 7.23 - 2008 Australian Open QF - 6-0,6-3,7-5 over Ferrer

Perhaps surprisingly, Novak's two wins at this year's Australian Open don't make the list, both rating 7.22, so certainly high rated, but not as high as some of his others.  And it is interesting that two of his best are against Ferrer.

Andy Murray does not have matches that have rated as high as those above, but still has several very good matches:
  • 7.27 - 2012 London Olympics Final - 6-2,6-1,6-4 over Federer
  • 7.17 - 2014 Australian Open SF - 6-4,6-7,6-3,6-7,6-2 over Federer
  • 7.17 - 2014 Wimbledon Final - 6-4,7-5,6-4 over Djokovic
  • 7.17 - 2012 US Open SF - 5-7,6-2,6-1,7-6 over Berdych

But he is a clear step below Novak.

Roger Federer's top is a bit higher than Novak from a top rated match perspective:
  • 7.42 - 2004 US Open Final - 6-0,7-6,6-0 over Hewitt
  • 7.34 - 2004 Wimbledon QF - 6-1,6-7,6-0,6-4 over Hewitt
  • 7.32 - 2006 Wimbledon Final - 6-0,7-6,6-7,6-3 over Nadal
  • 7.32 - 2009 Australian Open QF - 6-3,6-0,6-0 over Del Potro
  • 7.32 - 2007 Australian Open SF - 6-4,6-0,6-2 over Roddick

Federer is interestingly the only one to have his top-5 at three different venues.

Rafael Nadal has had some monster individual matches, when he is on, he is on, especially at Roland Garros:
  • 7.58 - 2008 French Open Final - 6-1,6-3,6-0 over Federer
  • 7.49 - 2012 French Open SF - 6-2,6-2,6-1 over Ferrer
  • 7.43 - 2012 French Open R16 - 6-2,6-0-6-0 over Monaco
  • 7.38 - 2013 French Open Final - 6-3,6-2,6-2 over Ferrer
  • 7.37 - 2008 French Open QF - 6-1,6-1,6-1 over Almagro
  • 7.36 - 2008 French Open R16 - 6-1,6-0,6-2 over Verdasco

Clearly Nadal is fantastic at the French Open and has the highest rated match I've calculated, but one trick pony?  He has several others high ratings that come after the above also at the French Open, and just one Davis Cup match in the mix to break things up.

Federer's best results appear to be grouped the closest, so he seems to have been able to play at his best more consistently, while Nadal's very best has been higher albeit at a single event.

Some other high rated matches from the past:
  • Jim Courier - 7.53 - 1992 French Open Final - 7-5,6-2,6-1 over Korda
  • Jim Courier - 7.52 - 1992 French Open SF - 6-3,6-2,6-2 over Agassi
  • Ivan Lendl - 7.50 - 1986 US Open Final - 6-4,6-2,6-0 over Mecir
  • Bjorn Borg - 7.49 - 1978 French Open SF - 6-0,6-1,6-0 over Barazzutti
  • Pete Sampras - 7.48 - 1997 Australian Open Final - 6-2,6-3,6-3 over Moya
  • Bjorn Borg - 7.47 - 1978 French Open Final - 6-1,6-1,6-3 vs Vilas

Courier had a nice set of back to back matches at the 1992 French Open, the highest back to back of any player, and Borg was very good at the 1978 French Open.

What do you think?  What player at his best has had the best result?

Attribution: Match data is courtesy Jeff Sackmann / Tennis Abstract.

Handicapping 2016 USTA League 40+ 4.5 Women's Nationals - What areas have the favorites to win it all?

USTA League play in the 40 & over division has started in many areas.  So it is time to start looking at what areas my Estimated Dynamic NTRP Ratings say have the strongest teams.  I started with the 3.5 Men and Women, 4.0 Women and Men3.0 Men and Women, and 4.5 Men.

Note of course that some areas have not started their 40+ leagues yet and so without rosters to look at are not included.  And some areas have multiple seasons so some teams have multiple chances to advance in playoffs or have to make a choice.  So this is preliminary and will be updated in the coming months.

Here are the top-20 teams by top-8 average for the 40+ 4.5+ Women:

NO. CALIFORNIA / Women's 4.5 - SA4.44
NO. CALIFORNIA / Women's 4.5 - DN4.43
NO. CALIFORNIA / Women's 4.5 - LP4.42
NO. CALIFORNIA / Women's 4.5 - SA4.41

Once again, we see a few times above the top of the range, but unlike the 4.5+ Men, just three teams are.

This is one of the genders/levels that has the most even distribution with Florida having several teams but from different regions, Southern also but from different states, the California's represented well, and even a few other sections like New England, Texas, and PNW thrown in.

As always, if you'd like to get more details on the specific teams in the list, or get a report on your area or sub-flight, I can generate a report for you showing both the full roster average and top-8 average by team.  Contact me for more details.

More to come!