Category Archives: Roger Federer

A Quarterfinal on Federer’s Racquet

The Roger Federer-Andy Murray head-to-head is a bit of a baffling one. In twenty career meetings–18 of them on hard courts–Murray has won 11, including four of the last five.

Yet for a superficially tight one-on-one record, Fed and Murray haven’t played many tight matches against each other, especially lately. When they went five sets in last year’s Australian Open semifinal, it was the first time they had gone the distance in ten matches. The outcome of a match between them is up for grabs, but whoever wins it tends to do so by a handy margin.

Even that five-set semifinal last year wasn’t as close as it looked. Murray won 54.0% of total points and racked up a Dominance Ratio (DR) of 1.32, meaning that he won far more return points than Roger did. Five setters are usually much closer to 50% and 1.0, respectively. While Murray won far more points, Federer displayed his historically-great tiebreak skill to keep himself in the match.

DR is a convenient measure of the closeness of a match, where 1.0 is a dead heat. Only two Fed-Murray matches–both before 2009–fell in the range between 0.85 and 1.15. By contrast, Novak Djokovic and Rafael Nadal have played seven matches (including two Grand Slam finals) in that range, and Djokovic and Murray have played five.

Tactical nonsense

To traffic in conventional wisdom for a moment, Federer is the most aggressive of the Big Four, while Murray is the most passive. To the extent Andy is likely to hurt Roger, it has more to do with his ability to force Fed into trying to do too much, particularly on the backhand side. If Federer plays patiently and picks his spots, he can crush Murray. If he plays too passively or hits bunches of unforced errors, it can be a rough day at the office.

However, there may not be much Murray can do to determine which Roger shows up.  Simply forcing Fed to hit backhands certainly isn’t enough. The Match Charting Project has amassed shot-by-shot data, including the number of groundstrokes hit from either side, for 23 Federer matches so far. Nadal is particularly good at directing the ball to Federer’s backhand, forcing Roger to hit 56% to 58% of groundstrokes from the backhand side in both a win (last year’s World Tour Finals) and a bad loss (the 2011 Tour Finals).

Taking the average of these 23 matches (most of which are Federer wins, as the Match Charting Project seems to have drawn lots of Fed fans), Roger hits 52.5% of his groundstrokes from the forehand side. This reflects the balance of two factors: Federer wanting to hit his forehand, and opponents trying to keep the ball away from it.

Surprisingly, hitting lots of balls to Fed’s backhand side seems to have few benefits. There is no meaningful correlation between DR and the percentage of groundstrokes Fed hit on the backhand side.

Based on the limited data available, it appears that Murray has tried a variety of tactics.

In the two Fed-Murray matches for which we have shot-by-shot data–the 2010 Australian Open final and the 2012 Dubai final–Murray took opposite approaches to the problem. In the Melbourne final, he managed to direct 57% of balls to Fed’s backhand, which is as good as anyone but Nadal has managed. In the Dubai match, Roger hit 64% of his groundstrokes from the forehand side, the second-highest rate of any of the 23 Federer matches in the database.

In both cases, Murray lost. To take another example, Juan Martin del Potro has beaten Fed while letting him hit 57% forehands and lost to him while forcing him to hit 57% backhands.

The database–limited in matches and biased as it is toward Fed’s victories–probably can’t take us any farther. But from here, we can speculate that Federer has it in his power to win or lose regardless of the tactics thrown his way. Murray, like Nadal, has always forced him to hit one extra ball. The sort of aggression that takes a player far out of position to hit, for instance, an inside-out forehand can backfire against such a talented defensive player.

In four matches at the Australian Open so far, Federer has offered us plenty of glimpses of his glory days. Murray will likely prove to be his biggest test of the tournament, but Fed’s fate still hangs on his own racquet.

4 Comments

Filed under Andy Murray, Australian Open, Match charting, Roger Federer

Roger Federer and the Missing Tiebreaks (+Updated WTForecast)

For most of his career, Roger Federer has been one of the very few players to play better in tiebreaks than in standard deuce games.  His career record, winning breakers at a 65% clip, illustrates his success at the business end of tight sets.  But there’s more to the story.  Even a player a good as Federer has been should not have won that many tiebreaks.

As I wrote in a pair of posts a year ago, there is very little evidence for any kind of tiebreak-specific skill.  Some players do well in tiebreaks, of course, but their success is almost always due to being good in general–better players win more points, and that translates into tiebreaks.  Plenty of big servers, such as Ivo Karlovic and Milos Raonic, don’t win any more tiebreaks that you would expect simply by looking at the rate at which they win points.

However, a tiny fraction of players defy this regression to the tiebreak mean. Playing a ton of tiebreaks seems to help a bit–John Isner always wins more than expected–and a few other cases might be explained by extreme confidence or intimidations.  These include Pete Sampras and–you guessed it–King Roger.

In the eight seasons from 2004 to 2011, Federer won almost 10% more tiebreaks than his stats say he should have.  In 2006, his outrageous 37-14 tiebreak record was a big part of his equally outrageous overall success.  But even a player as good as Roger was that year “should” have only gone 31-20.  That would still have been an impressive win rate, and let’s not forget, many of his tiebreaks were against excellent players who had already pushed him that far.

As with so much else, that tiebreak magic has eluded Fed in the past two seasons.  Last year was the first season since 2003 when he failed to win more tiebreaks than expected.  He has been neutral this year and last.

It’s tempting to wonder, then, how big a part the disappearance of Roger’s tiebreak magic has played in his overall decline.  If he had won tiebreaks at the “extra” rate he did throughout his peak, he would have claimed two, or possibly three more than he actually did, flipping his pedestrian 13-10 tiebreak record to a more Fed-like 15-8 or even 16-7.  (This post was written before Fed’s tiebreak win over Djokovic in London on Tuesday.  In any event, improving his record to 14-10 doesn’t drastically change anything.)

How much of an impact would those bonus tiebreaks have had?  With a bit of guesswork and a handful of counterfactuals, we can put a number on it.  We’re looking at “flipping” two or three of Roger’s ten lost tiebreaks.  Of those ten, three didn’t end up mattering, as he won the match anyway.   The remaining seven occurred in five matches:

The final match in this list provides the simplest illustration of the math involved here.  Flip the lost tiebreak in the Delpo match, and Federer wins the title, earning 200 additional ranking points.  Since we’re only switching the outcome in two or three tiebreaks, that’s either a 20% or 30% chance of that particular tiebreak counting among those switched, for either 40 or 60 additional points.

It gets much more involved with something like the Stakhovsky loss.  Not only do we need to consider the different outcomes of flipping both tiebreaks (and Roger winning) and flipping just one (and Roger maybe winning), we also need to estimate Fed’s chances of progressing through the draw.  Despite the very early loss, Wimbledon was almost double the lost opportunity of any of the other matches, as his path to the semifinal would’ve gone through Jurgen Melzer, Jerzy Janowicz, and Lukasz Kubot.  To quantify the effect of flipping the Wimbledon outcome, we must consider the probability of his reaching those later rounds and the number of points he would have collected had he gotten that far.

Crunch all the numbers, and if you flip two tiebreaks, Federer gains about 380 ranking points.  Flip three, and it’s about 560.  Either of those numbers would move him in front of Berdych in this week’s rankings and given him a lot more breathing room on the road to London.  These bonus points would still have left a huge gap between him and the top five.

Perhaps more important than a few hundred ranking points, how different would the 2013 Federer storyline look if you flipped just a small number of those results?  Give him the 4th set against Stakhovsky and the 2nd with Delbonis, watch him win the deciders, and there’s a different Fed narrative for the summer.  Whether it’s bad luck, decreased confidence, less intimidation, or something else entirely, it’s crucial that we remember that tiebreaks are often decided by a single bad service point or great return point.  If a narrative can’t hold up against a couple of points going the other way, it probably isn’t telling us very much about a player’s actual performance level.

Yet, if Federer has turned a corner this fall, it would be a mistake to expect improved results to come from a resurgence of his tiebreak mojo.  Whatever mysterious factors cause a tiny minority of players to exceed tiebreak expectations, it seems less likely that fading 30-something Fed has them.  He certainly hasn’t benefited from them for the last two years.  But most of all, unless he gets back into more very high-profile matches–as he may this week–the few hundred points he could gain from tiebreak magic just won’t make much of a difference.

London forecast: Today, the results went as expected, with Nadal beating Ferrer and Novak defeating Federer.  Nadal was such a heavy favorite that his win doesn’t affect his chances much, but Djokovic enjoys a bigger bump. The top two seeds are now almost equal, while Federer faces increasingly long odds.

Player     3-0  2-1  1-2  0-3     SF      F      W  
Nadal      50%  42%   9%   0%  91.2%  52.5%  31.5%  
Djokovic   43%  46%  11%   0%  88.5%  54.4%  31.0%  
Ferrer      0%  29%  50%  21%  31.9%  12.2%   4.5%  
Del Potro  22%  50%  28%   0%  71.3%  36.6%  16.7%  
Federer     0%  30%  51%  20%  30.2%  14.1%   6.3%  
Berdych     0%  14%  48%  38%  16.4%   5.7%   2.0%  
Wawrinka   13%  48%  38%   0%  60.5%  21.2%   6.9%  
Gasquet     0%  10%  44%  45%  10.0%   3.3%   1.1%

Click here for the pre-tournament forecast.

Leave a comment

Filed under Roger Federer, Tiebreaks, World Tour Finals

Number One Bagels and Clutch Break Points

The big story from yesterday’s action at the US Open was the dominance of the world #1s.  Both Novak Djokovic and Serena Williams dished out two 6-0 sets, making one wonder if we’d been transported back in time to the first Tuesday, when top players are more likely to face opponents who don’t challenge them.

Djokovic’s drubbing of Marcel Granollers was only the 146th men’s Grand Slam match of the Open era in which one player won two bagel sets.  That’s a little less than once per Slam for that time period.

Only 15 of those double-bagels have come in the fourth round or later, and such final-16 drubbings have gotten more rare over time–only 5 of the 15 have taken place since 1983.  The most recent was Rafael Nadal‘s defeat of Juan Monaco at last year’s French Open, 6-2 6-0 6-0.  Roger Federer shows up on the list as well, twice: His quarterfinal win over Juan Martin del Potro at the 2009 Australian, 6-3 6-0 6-0, and the final in his 2004 US Open title over Lleyton Hewitt, 6-0 7-6 6-0.

Double bagels are a bit more common in the women’s game, though not as frequent for Serena at Slams as you might expect.  While there have been over 180 in the Open era, yesterday’s defeat of Carla Suarez Navarro was only her fourth.  Several of the game’s greats tallied more than that, notably Chris Evert with 13, Margaret Court with 8, and Steffi Graf with 7.

Where Serena stacks up more impressively is in her record of 6-0 sets this year.  She has now served a bagel in ten different Grand Slam matches in 2013, including two double bagels.  Only Court in 1969 and Graf in 1988 won a 6-0 set in more Slam matches in a single year, and only Graf won more 6-0 sets at Slams in a single year.

Of course, Serena isn’t done yet.  However, in nine career matches against her semifinal opponent, Na Li, she has only won a single set 6-0.  She might not want to do it again: After serving a bagel set to open their 2008 in Stuttgart, Serena lost the next two sets for her only career loss against Li.

As we all mulled over Roger Federer’s future yesterday, Carl Bialik outlined a useful way of thinking about break point conversions.  As I noted yesterday, while Federer has played horribly on such key points in his last several slam losses, it’s not clear how much we should read into those numbers.  Yes, he probably would’ve won the match had he converted more break points, but does a dreadful 2-for-16 showing (or several) mean he is a fundamentally different player than he used to be?

Carl’s algorithm involves comparing performance on break points to performance on all other points.  If tennis players were robots, we would expect them to perform exactly as well at 30-40 as they do at 30-0.  The only slight difference is that most break points take place in the ad court, and lefties have an advantage there.  For now, let’s ignore that.

Thus, a player who wins 44% of break point opportunities against only 40% of other return points is playing 10% better in those pressure situations.  We might even say he is performing well in the clutch.

I ran these numbers for every member of the top 50 in 2013.  As is so often the case, the results don’t offer a lot of confidence in the connection between break point results and clutch skills.

The four players who have performed the best this year on break points, relative to other points in the same matches, are Jo-Wilfried Tsonga (+14%), Martin Klizan (+12%), Nicolas Almagro (+10%), and Ernests Gulbis (+10%).  Of the big four (or five, or seven), tops is Rafael Nadal, at +5%.

At the other end of the spectrum are Tommy Robredo (-5%), Sam Querrey (-6%), Kei Nishikori (-6%), Michael Llodra (-7%), and David Ferrer (-7%).

(These numbers don’t include the US Open.  If they did, presumably Robredo would move up a few spots.)

Federer ranks 38th among the top 50, winning 2.6% fewer break points than non-break points.  That’s certainly nothing to be proud of, but it’s only two spots behind Novak Djokovic, at -1.7%.

Another approach that matches our intuition a little better is to look only at break point opportunities–that is, clutch return points.  Here, Federer is -7.8%, worse than 40 members of the top 50.  Djokovic and Andy Murray are still in the bottom half, but a full 10 spots ahead of Roger, at -3.2% and -3.7%, respectively.  Nadal is +2.1%.

If nothing else, these numbers show us how thin the margins are in top-level men’s tennis.  A few percentage points differentiate the very best from a fading player having a disappointing season.

The presence of Djokovic so far down these lists serves as another reminder.  Converting break points is a numbers game.  Look through Novak’s season and you’ll find a couple 3-for-11s, a 2-for-12, and a 4-for-18 (against Bobby Reynolds!).  You only need to convert a few to win a match, and the best way to convert a few is to earn as many as possible.

In other words, break point conversion rates represent only a small part of a player’s performance on any given day.  Earning those break opportunities can be every bit as important, and that’s one category in which Federer remains strong.

If you missed it last night, check out my recap and detailed stats for Murray vs. Istomin.

Here’s another interesting graph from Betting Market Analytics, showing win probability throughout yesterday’s Ivanovic-Azarenka match.  Because Vika was so heavily favored yesterday, she retained a better than 50/50 chance of winning the match even after Ana took the first set.

Leave a comment

Filed under Bagels and breadsticks, Novak Djokovic, Roger Federer, Serena Williams, U.S. Open

Unexpected Quarterfinalists: Gasquet, Hantuchova, and Not Fed

Yesterday, Richard Gasquet won a fourth-round match at a Grand Slam.

If that doesn’t surprise you, you haven’t been paying much attention to Gasquet for, say, the last eight years.  The Frenchman with the stunning backhand has advanced to the fourth round at a Slam 17 times now, making him only the 35th man in the Open era to do so.  The problem is what happens next.

Entering yesterday’s match, Gasquet was 1-15 in round-of-16 matches at majors, his one victory coming at 2007 Wimbledon over Jo-Wilfried Tsonga.  Since then, he’s lost his last eleven tries, including one to Tsonga and two to David Ferrer, his quarterfinal opponent this week.  No player has lost more than 15 fourth-round Slam matches; only Wayne Ferreira reached the same plateau, and Lleyton Hewitt will match it if he loses today.

One thing that has held him back is an inability to beat higher-ranked players, as Carl Bialik noted earlier this year.  At slams, he has played 28 matches against players with superior ATP rankings, and won only four of them.  Against lower-ranked players, he is 62-11.  Since Gasquet’s ranking has rarely reached the top eight, that mark hasn’t helped him the fourth round, where players outside of the top eight generally meet a higher-ranked opponent.

Now that Gasquet has broken through with his second Grand Slam quarterfinal appearance, history suggests he’ll go no further.  He has beaten Ferrer only one time in nine tries, and that was five years ago.  And Ferrer’s ranking puts him firmly in the category of guys Gasquet doesn’t beat at majors.

There’s one reason for hope, though.  Despite all the disappointment in the fourth round, he has never lost a Grand Slam quarterfinal.

Daniela Hantuchova‘s appearance in the quarterfinals of this year’s US Open is surprising for a different reason.  When the tournament began, her spot was pegged for Petra Kvitova, before an ailing Kvitova was upset by Alison Riske.  For all my talk recently about easy bracket on the men’s side, no one in either single’s draw has faced such lowly-ranked competition.

Hantuchova’s four opponents thus far include two qualifiers and two wild cards.  Among them, only Riske is ranked inside the top 100, and she’s #81.  By contrast, Hantuchova’s presumptive quarterfinal opponent, Victoria Azarenka, will have faced #13 and #28.

Of over 850 women’s Slam quarterfinalists since 1987, only six have reached the quarters without playing someone in the top 80.  The luckiest path was that of Claudia Kohde Kilsch, who reached the 1989 Wimbledon quarterfinals by beating #126, #246, #247, and #131.  Then her luck ran out: Steffi Graf ended her run in the quarters.  Steffi herself is one of the six, having won her first four rounds at 1993 Wimbledon without playing anyone ranked better than #87.

These lucky draws have become less common in recent years.  Of the six, only one has occurred since Steffi’s run in 1993.  Nadia Petrova reached the quarterfinals at the 2006 Australian Open without having to beat anyone ranked better than #100.

After four easy matches, there’s little pattern to how these players fare in the quarters.  As we might expect, the success rate in their fifth matches has much more to do with their quarterfinal opponents than the women they faced to get there.

And perhaps you’ve heard: Tommy Robredo defeated Roger Federer in straight sets.

It was the first time in twelve meetings that Robredo beat Fed.  It’s the Spaniard’s first quarterfinal appearance in New York, despite seven previous fourth-round showings (including one against Roger, in 2009).  Even Gasquet hasn’t been that bad, losing in the US Open round of 16 a mere four times.  And Robredo pulled off the upset while winning fewer return points than his opponent did–something that happens in only one of 15 US Open men’s matches.

When oddities like this occur–Gasquet’s match is another, as he won only 48.5% of total points–it is almost always because the winner played much better on high-leverage points.  In many matches, those important moments are at the back end of tiebreaks, when two points can make or break a set.  In Federer’s loss, the finger-pointing is directed at break points.  Roger barely converted any of them. It’s been a problem for Fed for years, particularly in his last several Slam losses.

It’s difficult to know how to evaluate poor break point performances.  In one sense, it’s obvious: If Fed was going to win the match, he needed to win more.  A failure to convert break points is a good explanation for any loss.

But what does it say about Fed’s current level, or about what we can expect from him going forward?  Is he suddenly weak on break points?  When I ran the numbers a couple of years ago, he was winning slightly fewer return points in the ad court, but the difference isn’t nearly extreme enough to explain a 2-for-16 performance on break points.

What’s particularly frustrating about squandering so many break points is that he earned them with good play on other return points. And, of course, there’s no difference between a typical ad-court point and a break point except for the pressure.

So, if Federer is still generating all those break-point opportunities, is he simply suffering through a run of bad luck?  Has he lost his clutch superpowers?  Have other players ceased to fear him in big moments?  Judging from the growing number of surprising defeats in Roger’s record, it certainly seems to be something more than bad luck.

Finally, a couple of notes.

Don’t miss this win probability graph of the Raonic-Gasquet match.  Mike says it’s “almost too interesting.”

In the New York Times Straight Sets blog (known for its coverage of the United States Open), Clayton Chin gives a brief overview of a forecasting method.  He emphasizes his reliance on the Monte Carlo method–a technique that utilizes thousands or even millions of simulations–which isn’t necessary here.

If you estimate each player’s serve and return points won, it’s straightforward to calculate each player’s chances of winning a game, set, or match.  Generally speaking, Monte Carlo techniques are useful when such closed-form solutions aren’t available.

The most important part of Chin’s approach is one he doesn’t shed any light on.  If Serena is holding serve at a certain rate and breaking serve at a certain rate over the course of the year, how do you generate hold and break rates for an individual match?  It can be done, and many have tried, but that’s much more challenging that simulating outcomes at the match or tournament level.  Without that glimpse under the hood, it’s tough to know how much weight to give his results.

Leave a comment

Filed under Daniela Hantuchova, Richard Gasquet, Roger Federer, U.S. Open

Guaranteed Five-Setters, Exhausting Routes to R4, and Master Bakers

With so many of the world’s top players in action yesterday, it’s only fitting to lead with Denis Istomin and Andreas Seppi.

Istomin and Seppi have now met four times in the last 15 months, all at Grand Slams.  And thanks to yesterday’s effort, they’ve now gone five sets in all four of those matches.

Cue the chorus: “That’s got to be some kind of record, right?”

Yep, it is.  While their US Open third-rounder was Seppi and Istomin’s seventh meeting overall, it was only their fourth at a major, meaning that each time they’ve met in the best-of-five format, they’ve gone the distance.  Two pairs of players (Thomas Muster and Albert Costa, and Guillermo Canas and Gaston Gaudio) have met three times in a best-of-five and reached a decider each time, but no two players had ever gone four-for-four.

In fact, Seppi and Istomin are only the eighth pairing in the Open era to record four or more five-setters.  Petr Korda and Pete Sampras played four five-setters in five matchups, but their first such meeting, a 1992 Davis Cup match, only went four sets.  Radek Stepanek and David Ferrer are also close, having played four five-setters in five best-of-five meetings.

Most of the pairs that have played so many five-setters required many more meetings to do so.  You might be familiar with some of the guys who make up the three head-to-heads that have played five five-setters: Jimmy Connors-John McEnroe, Stefan Edberg-Ivan Lendl, and Roger Federer-Rafael Nadal.  But all of those pairs met more than 10 times in best-of-five situations.  In this context, all those three- and four-setters seem rather weak.

When the Australian Open draw comes out, while everyone else figures out whose quarter Federer landed in, I’ll be checking Seppi’s proximity to Istomin.

When Marcel Granollers edged by Tim Smyczek in five sets yesterday, the big story was the futility of American men’s tennis.  (Thankfully, for the depressed patriots among us, Sloane Stephens was putting up a spirited challenge against Serena Williams on another court.)

However, the Spaniard was making a bit of history of his own.  In beating Jurgen Zopp, Rajeev Ram, and Smyczek, he’s won three five-setters in his first three rounds, becoming only the 15th man to do so in the Open era, the first since Janko Tipsarevic did so at Wimbledon in 2007.  It’s only the third time someone has done it at the US Open.  The last man to do so in New York was Wayne Ferreira, in 1993.

Amazingly, three players have gone five sets in each of their first four matches in a slam.  The last such occurrence was when Dominik Hrbaty reached the fourth round at the Australian, in 2006.  He fell to Nikolay Davydenko in the fourth round.

This is one bit of history that Granollers surely won’t be making.  As remarkable as it is to reach the fourth round of the back of all those five-setters, it isn’t a good sign when you lose two sets apiece to three players ranked outside the top 100.

It certainly doesn’t bode well when your next opponent is Novak Djokovic.

As Federer, Nadal and Djokovic plow their way through the early rounds this year, none is wasting any time.  All three players have posted a 6-0 set in their second- or third-round matches, exclamation points amidst broader displays of dominance.

A quick check of the database reveals yet another category in which Federer is charging toward the top.  The Open era record for bagel sets won at Grand Slams is held by Andre Agassi, who retired with 49.  Fed’s bagel of Adrian Mannarino on Saturday was the 43rd of his career.

Here is the all-time list:

Player           Slam bagels  
Andre Agassi              49  
Roger Federer             43  
Ivan Lendl                42  
Jimmy Connors             41  
Bjorn Borg                35  
Guillermo Vilas           29  
John Mcenroe              29  
Stefan Edberg             25  
Boris Becker              23  
Rafael Nadal              22  
Novak Djokovic            21

Andy Murray is tied for 19th, with 16.

This is one category which highlights the extreme dominance of some of the greatest female players in history.  Chris Evert puts Agassi, Federer, and everyone else to shame, with a record 104 Grand Slam bagels.  Serena Williams’s first-round defeat of Francesca Schiavone moved her past Arantxa Sanchez Vicario into fifth place on the all-time list:

Player                   Slam bagels  
Chris Evert                      104  
Steffi Graf                       74  
Martina Navratilova               70  
Monica Seles                      51 
Serena Williams                   49 
Arantxa Sanchez Vicario           47  
Margaret Court                    44  
Gabriela Sabatini                 44  
Lindsay Davenport                 43  
Maria Sharapova                   41

With a quarterfinal matchup against Carla Suarez Navarro, it’s possible Serena isn’t done for the year.  Each of the two previous times the two women have played, Williams has won a 6-0 set.

Leave a comment

Filed under Bagels and breadsticks, Five-setters, Records, Roger Federer

Federer vs Zemlja Serve Profile: In Extreme Detail

As I wrote last week, tennis needs more detailed statistics.  Most of all, we need them in an open format so that researchers can utilize all the data stored for every match.  No use in have Hawkeye cameras on every court if the data stays locked up.

I’m working on a system for charting matches and storing extremely detailed serve and shot information.  It will have to stay under wraps until I get a few more kinks worked out, but in the meantime, I want to show off some of what it can do.

Click here for more exhaustive serve data than you’ve probably ever seen before.

Today’s match wasn’t the most gripping that Roger Federer (or Grega Zemlja) ever played, but there’s still plenty of interesting stuff:

  • Roger won 85% of first-serve points. No surprised there.  More impressively, he won 60% of his first-serve points on or before his second shot.  (That’s “<=3W” in the tables.)
  • Fed went down the T with just under half his first serves (47%), but up-the-middle offerings accounted for 11 of his 12 aces.
  • Zemlja hit a shocking 27 serves into the net–almost half of his faults, and just over 20% of all of the serves he hit today.  (Watching the match, it felt like even more.)
  • Roger’s first serves were somewhat more dominant in the deuce court, as he lost only three first-serve points in that half, and won two-thirds of his first-serve points in the deuce court by his second shot.  In the small amount of data on offer today, he was noticeably weaker with his deuce court second serve, losing 5 of 12 second-serve points in that direction, compared to only 3 of 18 second-serve points to the ad court.
  • Zemlja fared better serving to the ad court today (64% of service points won to 56% in the deuce court), and was particularly deadly when he landed a serve wide in the ad court.  He won seven of the eight points that started that way, five of them with or before his second shot.

(If you didn’t click on the link the first time you saw it, now would be a good time.)

You get the idea, I hope.  With this much data, the sifting is as important as the collecting.  There are hundreds of data points we can generate just from tracking each player’s serve performance, and we can expect that most of them won’t have much to tell us.

And, of course, one match is just that–a small sample, fewer than 100 service points for each player.  While we can look at these tables and gain some insight into exactly how Roger was dominant today, it would be a mistake to draw much in the way of broader conclusions.

For that, we’ll need more matches, more data.  We’ll get there.

Leave a comment

Filed under Match charting, Roger Federer

Federer, Nadal, and Semifinal-or-Later Streaks

The Indian Wells men’s draw has been released, and a big question has been answered.  Rafael Nadal, about as dangerous a floater as can be imagined with a #5 seed, landed in Roger Federer‘s quarter.  (Sorry Roger, it had to happen to someone, and David Ferrer has suffered enough lately.)

If Fed and Rafa both win three matches, they’ll face each other in a quarterfinal match.  That’s something that’s never happened before.  The pair has met 28 times, 26 of them in a semifinal or final.  The only exceptions are their first match in 2004, when Nadal was seeded 32nd in Miami, and a round-robin pairing at the 2011 tour finals.  Ignoring the round-robin, that’s 26 matches in a row in one of the last two rounds of an event.

That’s a historically great streak, but it’s not the record.  In fact, one player is a part of two streaks–the only two streaks–that are better.

Jimmy Connors is 1st, with 28 consecutive semis or finals against Ivan Lendl, and 2nd, with 27 consecutive semis or finals against (who else?) John McEnroe.  He’s also eighth (21 straight against Bjorn Borg) and 12th (14 with Ilie Nastase).

Until the threat of this week’s draw, Federer and Nadal were right on Connors’s tail.  If Roger and Rafa meet in the quarters, the heir presumptive pair will have to include Novak Djokovic.

Here’s the all-time top ten:

Streak  Player1          Player2           
28      Jimmy Connors    Ivan Lendl        
27      Jimmy Connors    John McEnroe      
26      Rafael Nadal     Roger Federer     
23      Rafael Nadal     Novak Djokovic    
22      Stefan Edberg    Boris Becker      
22      Roger Federer    Novak Djokovic    
22      John McEnroe     Ivan Lendl        
21      Bjorn Borg       Jimmy Connors     
19      Stefan Edberg    Ivan Lendl        
17      Ivan Lendl       Boris Becker

If Nadal stays #5 for long (unlikely as that seems), both the all-time #3 and #4 streaks could be halted.  But as long as Federer stays within the top four, the current #6 streak will climb the rankings.

Of course, there are a couple of other combinations with the potential to crack this list, even reach the top:

Streak  Player1         Player2        
11      Andy Murray     Roger Federer  
10      Novak Djokovic  Andy Murray

But we shouldn’t get ahead of ourselves.  It took five years for Fed and Nadal to get from 11 up to 26.  As the top of the list shows, it takes two consistently great players to put together a streak like this.

All is not lost, though.  If they play in the quarters, they’ll just have to shift their focus to a new record: consecutive meetings in quarterfinals or later.  27 straight would put them behind Connors-McEnroe (32), Connors-Lendl (29), and one pair they’re unlikely to chase down: Nadal-Djokovic (29).

Leave a comment

Filed under Head-to-Heads, Jimmy Connors, Novak Djokovic, Rafael Nadal, Records, Roger Federer

The Most Familiar Faces

In last week’s Basel final, Roger Federer and Juan Martin Del Potro faced off for the seventh time this year, and the 16th time overall.  Seven times in one year is an awful lot, about 10% of Delpo’s matches.  It’s even more remarkable because only two of those contests have been finals — in order to meet so many times, the draws of several tournaments had to complement their consistently strong play.

Making matters even more extreme is that there is a better-than-50% chance that Federer and Del Potro will meet in London next week, bringing the total to 8.  And there’s a slim chance–if they are drawn in the same group, then play again in the final–that the sum will reach 9.

So, what’s the record?  Seven is already pretty good, right?

Single year head-to-heads

In fact, as with so many other records, Federer is #1 in the last 30 years.  He holds the record with Jo Wilfried Tsonga, against whom he played eight times last year.  (In the entire professional era, the mark belongs to Ilie Nastase and Tom Gorman, who played at least nine times in 1972.  I’ve excluded years before 1980 because a variety of factors caused the top players to meet much more frequently than they do these days.)

As long as Fed and Delpo are at seven, they will be tied with four other pairs: John McEnroe and Ivan Lendl in 1984, Jim Courier and Michael Chang in 1995, Novak Djokovic and Rafael Nadal in 2007, and Novak/Rafa again in 2009.  Another 11 pairs met six times in a single year, including Nadal and Djokovic in 2008 and 2011.  (Along with, weirdly, Rajeev Ram and Donald Young in 2007.  Must be the wild cards.)

All-time head-to-heads

Since Djokovic and Nadal show up at the top of the single-year list no more than four times, it stands to reason that they must be near the top of the all-time list, as well.  Indeed, they are.

In fact, assuming Nadal returns to health in anywhere near his historical form, this current pair of stars will almost undoubtedly take over the all-time lead next year.  They could hold it for a very long time.

Player 1       Player 2        H2Hs    W-L  
Ivan Lendl     John McEnroe      35  20-15  
Ivan Lendl     Jimmy Connors     34  22-12  
Pete Sampras   Andre Agassi      34  20-14  
John McEnroe   Jimmy Connors     34  20-14  
Rafael Nadal   Novak Djokovic    33  19-14  
Boris Becker   Stefan Edberg     32  22-10  
Roger Federer  Novak Djokovic    28  16-12  
Rafael Nadal   Roger Federer     28  18-10  
Stefan Edberg  Ivan Lendl        26  14-12  
Roger Federer  Lleyton Hewitt    26   18-8

This is one record that, for all of his dominance, Federer will probably never co-hold.  To find yourself on this list, you not only need to rank among the all-time greats, you need a very-near-contemporary who ranks just as high.

(If you’re interested in head-to-head records, I hope you’re already using the Head-to-Head Matrix on TennisAbstract.com.  It’s updated every week, and shows the career H2H records of every matchup within the current top 15.  Each H2H record is linked directly to a listing of the relevant matches.)

2 Comments

Filed under Head-to-Heads, Juan Martin Del Potro, Records, Roger Federer

Withdrawal Effects

Yesterday, Mardy Fish withdrew from his fourth-round match against Roger Federer.  As we saw earlier today, Federer may gain some benefit from the extra rest, but with the additional rest days built into the grand slam schedule, Roger runs the risk of getting too little time on court.

What’s the true effect, then?  Will the extra rest make Federer an even bigger favorite in his quarterfinal match against Tomas Berdych?  Or will match-court rust hold him back?

As it turns out, there is virtually no effect.  Players handed a walkover win almost exactly half of their next matches, and a closer look at those matches reveals that 50% is about what we would’ve expected from them, walkover or not.

To hunt for a potential relationship, I found 139 ATP main draw walkovers since 2001 where the winner went on to play another match at the same tournament–in other words, excluding finals.  While it may seem that players tend to withdraw when they’re least likely to win a match (as with Fish this week, or like the other two players to withdraw before facing Federer this year), there’s nothing to that theory, either. The average pre-match odds of the withdrawing player are about 51%.

Thus, we can work on the assumption that there’s little bias in the pool of 139 men who received a free pass to the next round.  For every Federer, there’s a Donald Young advancing uncontested over Richard Gasquet.  Balancing the withdrawals of players without a chance may be higher-ranked players who are quicker to withdraw because their success allows them to play it safe and make longer-term decisions.

In the 139 follow-up matches, our players went 67-72, winning 48.2% of the time.  Prematch predictions (generated by Jrank) would have projected a winning percentage of 48.9%.

If we narrow the search to slams, we get a nearly-meaningless pool of only 12 matches.  The player coming off the walkover went 6-6; prematch numbers would’ve predicted 7-5.  Perhaps rust does play a small part; considerably more likely is that the walkover simply doesn’t affect the beneficiary.

For Federer fans, though, there’s little reason for concern.  This is the ninth time in his career he’s advanced via walkover, and he’s only lost the next match twice.  One of those was in 2002.  The other was in Indian Wells in 2008.  The man who beat Fed?  Mardy Fish.

Leave a comment

Filed under Roger Federer, U.S. Open, Withdrawals and Retirements

The Unbreakable and Record-Setting Cincinnati Finalists

When Roger Federer and Novak Djokovic met in the Cincinnati final on Sunday, they represented a unique event in tennis history: Neither one had been broken.  Four matches each, no breaks of serve.

That’s not just a Masters-level record, it’s a first for the ATP tour, at least since 1991, the time span for which point-level stats are available.    That’s over 1500 tournaments, including nearly 200 Masters events.

It’s very rare to even come close.  Of the 195 Masters tournaments for which data is available, only four pairs of finalists entered the title match with three or fewer breaks.  Djokovic leads the pack: When he met Rafael Nadal in the 2011 Miami final, Nadal had been broken once, Djokovic not at all.  When Djokovic and Federer met in the 2007 Montreal final, each player had only been broken once.  The Miami achievement is particularly notable because each player had won five pre-final matches, compared to only four each in Cincinnati and Montreal.

Federer set some records on his own, as well.  By holding his serve against Djokovic, he made it through an entire Masters tournament without suffering a break.  That’s the first time it has ever happened at this level.  Eight other times the winner has only been broken once–twice that winner was Federer, including Cincinnati two years ago.  Ten additional times, the winner was only broken twice–and Roger is responsible for three of those.

At lower level tournaments, it’s somewhat more common–the winner of a non-Masters event has made it through without losing serve a total of 17 times.  Surprise, surprise: Two of those are Federer, at Doha in 2005 and Halle in 2008.  Four other men have done it twice: Andy Roddick, Joachim Johanssen, Richard Krajicek, and Ivan Ljubicic.  Milos Raonic did it earlier this year in Chennai.

Federer set at least one more record last week, and it might be the most impressive of all.  He only faced three break points all week–the lowest known total at a Masters tournament.  The previous record was four, set by Andre Agassi at the 2002 Madrid Masters.  Fed’s total in Cinci was only the 10th ever in single digits–and Roger is now responsible for four of those top ten results.

At lower-level events, Fed’s mark has been bettered a couple of times.  At the 2007 Memphis tournament, Tommy Haas claimed the trophy without facing a single break point.   At San Jose this year, Raonic faced only two break points, though Tobias Kamke converted one of them.  Two other players–Andy Murray at 2009 Queen’s Club and Roddick at Lyon in 2005–got through an event facing only three break points.

No breaks, and record-settingly few break points. If hard courts are truly becoming slower, it seems that someone forgot to tell Roger.

1 Comment

Filed under Cincinnati, Novak Djokovic, Records, Roger Federer