Tuesday, January 31, 2012

Talent ID, USTA Player Development Folly

Identify the talented kids.  Bring them to a regional or national training center.  Get them the best coaching.  Put them through the best drills.  Train them over time to be champions.

That's the basic idea of talent identification within the USTA Player Development program.  It seems obviously the right thing to do.  But is the first step possible?

In this post I'll start off with some analogies that pretty clearly show the difficulty of projecting talent or skill into the future.  In another post I'll go into some of the science behind such human judgments, citing some relevant research and explaining the theory behind human judgment.

Imagine you're a National Football League general manager.  It's your job to acquire the most skilled football players for your team. You must look at players around the league who may become available as free agents or via trade.  Based upon their demonstrated performance at the NFL level you have to project how well they will perform on your team in the future, in this case probably from one to five years in the future.  This task is challenging, but not impossible.  I don't know all the criteria that GMs use to project future skill and I don't know their success rate in trades and free agent acquisitions.  I do know that they aren't either perfectly good or perfectly bad.  They have some rate of success between zero and one hundred percent.  If they're to keep their jobs the success rate probably has to be well over fifty percent or you could replace them with a coin.  Since GMs are frequently fired, perhaps replacing GMs with a coin would make sense for many NFL teams.

Let's increase the difficulty of the NFL GM's judgment problem a bit.  Each year NFL teams draft college players.  In this case, the GMs must predict how college level skills will translate to NFL environments.  They also must project over a longer time horizon, spanning from one to possible ten or even fifteen or more years in some cases.  Once again I don't know all the cues that the NFL talent evaluators use, but I know that they watch players in person and on tape.  I know that they frequently interview the players and the player's coaches.  I know that they use a variety of quantitative measures of speed, strength, size,  and even mental abilities.  Despite all these cues, the success rate for the NFL draft is considerably below one hundred percent.  No team goes even one draft year without making several mistakes.  To succeed in placing even half of a draft class on an NFL roster is probably considered pretty good, given how few roster spots are available each year.  So NFL player personnel departments spend tremendous amounts of time and money in making judgments about the future skill of grown men who have already mostly mastered the skill under consideration.  They still miss on a great many of these judgments.  They pick players who fail to pan out and they fail to pick players who go on to have outstanding careers.  This happens every year to every team.

Let's leave the NFL now and go back one level to college coaches.  They too must select players for their teams.  They do this by evaluating high school football players.  These are mostly sixteen, seventeen, and eighteen year-old young men.  They have played football for probably ten years already.  They have all gone through puberty.  They are not yet at their full size and strength, but they are within a few years of reaching their full physical maturity.  College coaches must judge how well these players high school performance will translate to the college level.  They must judge how a player will develop in the year or so before he comes to college and how much he will improve in the four or five years he spends on campus.   The younger the player, the tougher this judgment becomes.  Coaches who commit to juniors in high school have a higher error rate than those who commit to seniors in high school.  But the race to get the best players often induces coaches to commit to a "can't miss" player in that player's junior year.  Unfortunately they can and do miss.

Now let's jump to the talent identification problem facing the USTA Player Development coaches.  The goal of USTA PD is to produce world-class players.  The assumption is that the US Open, the USTA's big money maker, will be more popular with American players in the championship hunt.  The USTA's goal is to grow the game and the organization thinks the game will be more popular with Americans at the top of the professional game.  To develop these champions, the USTA PD tries to identify young kids with the potential to become world-champion players.  In light of the above discussions, how likely is the USTA to succeed in identifying children under age ten who will dominate professional tennis from age twenty to age thirty?

Projecting tennis skill, like football skill, a few years into the future is not easy.  The list of national junior or international junior champions does not completely overlap with future grand slam winners.  So skill in the late teens does not perfectly predict adult skill.  It's pretty good, certainly, but far from perfect just like in the football case.

The judgment of which ten-year olds will be great players at twenty is much, much tougher.  These kids are all yet to go through puberty.  They are a decade away from achieving their full size, strength and maturity.   Skill at age ten is not perfectly uncorrelated with skill later on.  I don't think there is a case in the last several decades where a player who subsequently became a world champion was not a very good player by age ten or twelve.  But the universe of all skilled players at age ten does not overlap very much with the set of world champions, that's for sure.

The probability that coaches can accurately judge the skill at age twenty of a child at age ten is vanishingly low.  The cues that coaches use simply are not valid, they are not stable over the time frame necessary to make accurate judgments.  Unless we discovered that skill was like eye-color or sex, that is completely genetically determined, the judgment task is essentially impossible.

By selecting small groups of kids and bringing them to regional and national training centers, the USTA Player Development is assured of making numerous errors.  They will invite plenty of kids who fail to become great and they will fail to invite kids who do become great.  By great, I don't mean world champions, unfortunately.  I really only mean great juniors or possible professionals.  I think world champions are Black Swans as Nassim Taleb would say, rare and unpredictable.

So rather than wasting resources on dozens or even hundreds of kids who show promise at a young age, the USTA Player Development system should spread its dollars far and wide to make sure the net captures as many players as possible.  By bringing more kids into the game they will by definition be growing the game.  If by chance some future champion resides in the US, then the wide net will have captured that player, too.

Monday, January 30, 2012

Server's Partner

How important is it for the server's partner to take the third shot in doubles?  In one match at the very highest level the answer is extremely.

I charted some aspects of the Bryan/Bryan def Lindstedt/Tecau 2012 Australian Open men's doubles semi-final match.  The Tennis Channel replay caught 182 of the 189 points played.  The Bryans won the match 7-5 in a third set breaker.  The match really was that close, with a winning margin of just one point, 95-94 for the Bryans.

Initially I wanted to know the winning percentage for the serving team when the server played the third shot.  Aggregating the data for both teams, the serving team won just under 45% of the points when the server played the third shot.

Next I looked at what percentage of third shots the server's partner played.  Both teams put 117 returns in play and the server's partner played 57 of those balls.  The serving team won 45 of those 57 points, or 79%.

Overall for the match, the serving team won 67% of the points.  Clearly the servers are at an advantage at this level of tennis.  However, if the returners can get their returns 1) in play, and 2) away from the net player, they swing the odds in their favor.  If the net man plays the ball, the serving team wins 79% of the points.  If the server plays the ball, the serving team wins 45%.  That's an enormous difference.

I've always preached "Get your returns in play and get them away from the net player if possible."  I finally have some pretty good data to back up that advice at the elite level.

Obviously I can't guarantee this analysis applies to any other levels of tennis.

Sunday, January 29, 2012

We Have a Winner

Novak Djokovic beat Rafael Nadal in an instant classic 2012 Australian Open Men's Final.  It took nearly six hours for Djokovic to prevail 7-5 in the fifth.  And Novak won.  Rafa didn't lose.

Let's go to the stat sheet.

Novak committed fewer unforced errors than Rafa did, 71 to 75.  So we might be tempted to say Rafa lost.  But that margin is small given that Djokovic won 193 points to 176 for Rafa.  The unforced error margin was only four out of seventeen.  Evidence that Rafa lost, but not convincing.

How about "littering the stat sheet"?  Which player dictated play by hitting more winners and committing more unforced errors?  Novak Djokovic wins the littering contest 118 to 105.

What if we add in forcing your opponent into the winning shots column?  Again, Djokovic appears to have won the match rather than waiting for Nadal to lose.  Djokovic hit 118 winning shots to 105 for Nadal.

Both players hit far more winning shots than they hit losing shots.  Djokers winning ratio was 118 to 71 (1.7) and Nadal's winning ratio was 105/75 (1.4).  Both guys were playing to win.

So congratulations to the winner, Novak Djokovic.  Well played.

Tennis Career Grand Slam

These athletes today sure are good.  Armstrong wins a hundred times in a row at the Tour de France.  The records of Babe Ruth and Roger Maris are crushed many times over.  Tennis players now keep ripping balls, sprinting, and changing directions for five and six hours at a stretch.  Must be the training.  That Rafa eats McDonalds and Skittles and guzzles sodas and has never been seen in the weight room means nothing.

During the open era of tennis which began in 1968, only one man had won all four of the tennis majors, completing the so-called career grand slam.  That man was Rod Laver.  Laver managed to win all four majors in the same year back in 1969.  An amazing feat. Laver did in one year what nobody else in the open era was able to do in their whole career.  That is until recently.

Borg failed. Connors failed. McEnroe failed. Lendl failed. Wilander failed. Becker failed. Edberg failed. Courier failed. Sampras failed.

Not until Andre Agassi teamed up with Gil Reyes did a man win a career slam in the open era.  Unlike when Laver did it, since the late 1970s the four majors have been contested on three different surfaces - clay, grass, and hard court.  So tough was it to beat all your foes on those different surfaces, in the different environments of the four majors, that nobody, none of those great champions listed above, was able to win at all four venues in their careers.

Since Agassi accomplished the feat, Roger Federer and Rafael Nadal have both done it.  Novak Djokovic is within a French title of joining them.  Three have done it.  A fourth is favored to join them.  All since the late 1990s.  Hmm.  What is it about those late 1990s that has turned tennis players into all-time greats?  Probably just the physios they all have traveling with them.  That's it.  The physios and those elastic bands.  And Skittles.

I'm sure that's what it is.

McGwire Defeats Sosa in Epic Aussie Open

It took nearly six hours for Mark McGwire to down Sammy Sosa in the longest grand slam tennis final in history.  McGwire's feat is even more impressive considering it came just two days after his nearly five hour semi-final victory.

The Spanish cycling and soccer teams sent along their congratulations.

Saturday, January 28, 2012

Doubles First Volley

I've been wondering about something for a couple of years.  I finally collected a decent amount of data on one match related to my wondering.

What percentage of points does the serving team win in doubles when the server plays the third shot?

I just charted the Bryan/Bryan vs Lindstedt/Tecau 2012 Australian Open men's doubles semi-final match won by the Bryans 4-6, 6-3, 7-6 (7-5).  The Bryans won the match 95 points to 94 points, so it was a very close, competitive match.

The serving Bryan brother played the third shot twenty-six times in the match (out of 99 service points - just over a quarter of their serving points).  The Bryans won ten of those points or roughly 38.5%.

When Tecau or Lindstedt served the server played the third shot twenty-one times (out of 97 service points - just over a fifth of their serving points).  Tecau and Lindstedt won eleven of those points or roughly 52.4%.

The cumulative totals for both teams for the match show that the server played the third shot in forty-seven of 189 points, almost exactly 25% of the points.  The serving team won twenty-one of those forty-seven points or roughly 44.7% of the points.

My assumption was that the serving team is at a disadvantage if the server plays the third ball.  In aggregate this was true.  It was true for the winning team, the Bryans, but not true for the losing team, Lindstedt and Tecau.

I guess my hypothesis needs more data than this one match.  It is interesting to me that first volleys (or half volleys or a couple of mid-court groundies) happen relatively infrequently, about a quarter of all points played.

I think it makes sense in doubles practice to prioritize 1) serves, 2) returns, 3) server's partner's first hit, and finally 4) first volleys.  I bet most players and coaches spend (waste) too much time on first volleys because they were a lot more common and important in the 1970s.

Incidentally I asked Dominic Inglot, ATP Tour doubles player, what he thought the winning percentage was for servers playing the first ball.  His answer was well over 50%.  He said great doubles players have awesome first volleys.  I'm sure they do.  I just don't think they're good enough to win with.  But to answer this question one way or the other, I'll have to chart more matches.

Win or Lose?

I think at the top of the game tennis is fundamentally a winners game.  The outcome is determined by the actions of the winning player.  To win means to hit winning shots rather than waiting for your opponent to make unforced errors that hand the match to you.  At lower levels tennis is a losers game.  The losing player will have lost the match to the winning player via unforced errors.  The level where the game switches is fuzzy.  That's my theory.  Well, it wasn't originally mine, but I've come to believe it.  The idea of winners and losers games started in finance and was applied to tennis by a businessman and engineer.

So what happened over the last couple of days in the men's Australian Open?   Rafael Nadal beat Roger Federer.  Or did he?  Novak Djokovic beat Andy Murray.  Or did he?

Very, very crudely Federer lost to Nadal.  Rafa won 146 points to 130 for Roger.  Rafa made only 35 unforced errors to 68 for Roger. 

How about being a bit less crude?  If you look at which player "littered the stat sheet" the most, it was Roger.  He had more winners (57 to 40) and unforced errors (68 to 35) than Rafa.  So Roger dictated the action, racking up a total of 125 winners and errors to just 75 for Rafa.  So Roger littered the stat sheet with fifty more entries than Rafa had.

What about total winning shots?  If we credit Roger with hitting a winning shot when Rafa is forced into an error, Roger hit 95 winning shots to just 78 for Rafa.  Again, Roger looks like he was playing winning tennis while Rafa was waiting for Roger to lose.  Roger lost.  Despite way more winning shots, Roger's unforced errors differential cost him the match. 

Minimizing errors and letting the opponent litter the stat sheet with both winners and errors was the winning strategy in the first men's semifinal.  Looks like Rafa won a loser's game.  The actions of the loser, Roger, dictated the result.

Let's move to the second semifinal, the epic battle between Djokovic and Murray.  Both guys are phenomenal defenders.  Djokovic credits his more aggressive play for his outstanding results in 2011.  Andy Murray says he must be more aggressive to get over the hump and win a major title.

Very, very crudely, Murray lost to Djokovic.  Djokovic won 184 points to 161 for Murray. Murray committed 96 unforced errors to 75 for Djokovic.  Murray lost, Novak didn't win.

Let's again look at who littered the stat sheet.  Djokovic had 135 winners plus unforced errors while Murray had 162.  By that measure, Murray dictated the action and lost.

How about factoring in forcing the opponent into errors?  On this score, Djokovic won the match, narrowly.  Djokovic hit 88 winning shots 86 for Murray.  So by this metric, we could say that Djokovic did indeed win the match.  Since the final margin was 23 points, though, this 2-point gap isn't much.

Despite all the talk of needing to be aggressive, it looks to me like semi final number two also was lost, not won.  Djokovic won by letting his opponent dictate more of the action and ultimately lose the match.

Does this mean that you can win on the ATP Tour or win a major tennis title by pushing, by simply hitting the ball down the middle and letting a great player lose to you?  Hardly.  But it does say that even at the very top of the game, matches are still lost, rather than won.

Sunday, January 22, 2012

Man of System

I'm skeptical of systems for producing tennis champions and of the ability of national federations to administer systems that produce tennis champions.  Empirically I'm going to run into a problem pretty soon, I think.  When every country has a system, then all champions will come from systems.

I guess some players could still shun the federations and systems and future champions could come from outside these systems.  With the enormous resources these federations have to throw at players, though, I don't see many young players turning down so much money.

Saturday, January 21, 2012

USTA's Huge Thumb on the Scale

Quick question.  Does it make sense that the USTA plays favorites among its players?  I know it is common practice among national tennis federations to give help to promising young players, to bring them to national or regional training centers, to pay for their coaching, to help them with travel expenses, etc.  But should they?

At the margin it must suck to see your opponent getting resources that you aren't from an organization to which you both pay annual dues.  Can you be sure that a favored player isn't also getting a sweet draw or more favorable match times?  Obviously the "sweet draw" part is true if getting a wildcard into the event in the first place can be considered sweet.

I'm sure lots of people have given this a lot of thought and come to the conclusion that national tennis federations putting their thumbs on the scale for certain players benefits the organization.

Clearly it benefits some players, parents, coaches and administrators.  But is it good for the membership generally?  I doubt it.

So Long Second Serve?

Traditionally when playing tennis, players will hit their first serve harder and flatter and closer to the lines than their second serves.  This causes a lower percentage of first serves to go in the box, but a higher winning percentage when they do.  Second serves, with less speed, more spin and a higher margin of safety go in more often, but the server wins a lower percentage of those points.  So far so good.

I was watching Ivo Karlovic play Roger Federer in the 2012 Australian Open the other night.  Ivo was struggling to win points on his second serve, but was serving a high percentage of first serves.  Since Ivo is 6'10" tall, his serve is awesome and he was winning a high percentage of points when his first serve was in the box.  Both players held serve until a first set tie-break.  Ivo got himself a set point, serving at 6-5 in the tie-breaker.  He missed his first serve.  Uh oh.  I'd been telling my wife for half an hour that he should be bombing second serves, and at set point I was yelling at my TV.  "Bring it, Ivo!!!  Pound the second serve!".  He spun it in, lost a flukey point.  Lost the set and the match.

Odds are Ivo would have lost the match no matter what he did on that point.  Federer is better than Ivo Karlovic.  However, Ivo would have given himself a better chance to win on this occasion if he had hit two "first serves" rather than a first and a second serve.  The data for this match are obvious, but what's interesting is the data for the tournament so far indicate that the strategy of hitting two first serves may become the norm in men's tennis in the not-too-distant future.

Here are the stats for Dr. Ivo against Federer.  Ivo made 71% of his first serves (ave speed 123 mph) for a winning percentage of 76% on first serve.  Ivo made 71% of his second serves (ave speed of 105 mph) for a winning percentage of 34%.  It should be obvious from those stats why hitting two bombs was a better idea for Ivo than spinning in a second serve, if you can call hitting it 105 mph spinning it in!  Ivo missed way too many of his second serves.  Typically ATP tour pros make 95% or more of their second serves.  Perhaps under pressure from Federer, Ivo made a woeful 71% of his second serves, the same as his much harder first serve.  So by abandoning the losing slower serve, Ivo would have increased his winning percentage on serve from 62% to 74%.  That's a dramatic increase in his chances of holding serve.

As I said, that case was easy because of Dr. Ivo's tragic second serve percentage and woeful winning percentage on second serve.  But what about for the field in general?

Here are the first round stats (the second two rounds are similar, but with fewer data points).  The men made 61% of their first serves and won 69% of those points.  They made 96% of their second serves and won 49% of those points.  The result of all that is that servers won 61% of their points using the traditional "hard first, spin second" serving strategy.

What would the numbers look like if the men had hit "hard" serves for both first and second serves?  They would have won 58.5% of their service points.  So in aggregate, for now, the traditional serving strategy is the better one by 61% to 58.5%.

My guess is that as players get taller and returners continue to get better, that gap will vanish for everyone like it vanished for Dr. Ivo.

If that gap does vanish and men hit two bombs, if necessary, on each point, the enjoyment level of men's tennis will fall dramatically.  We'll be back to the days of Sampras vs Ivanisevic at Wimbledon.  Yawn.  Let's hope that doesn't happen.

Sunday, January 15, 2012


"An expert is a man who has made all the mistakes which can be made, in a narrow field."
                                                      –Niels Bohr

I've made so many mistakes of the "three-across forehands" and "low to high" variety that I cringe to relive my teaching past.

Unfortunately, I think I'm a long way from "making all the mistakes that can be made" in tennis coaching.  It's not a narrow enough field.  And tennis isn't stable either.  It's a fertile environment for growing new mistakes!

Tuesday, January 10, 2012

Time to Add a Sport

So, Mike Bohn says CU is going to add a women's sport.  What sport will he add?  Time for a BRG poll.