Found an interesting article detailing hold and return statistics for the top players over recent eras. There are obviously other variables to consider, but it's a worthwhile read.
The ATP is not exactly famed for being generous with providing statistical data. Now, however, they've provided access to a statistical goldmine for the last 20 years.
The focus on this article is on the hold and return game.
How important are they? Can you be No. 1 with one, but not the other? Naturally, both are important, and you normally don’t get to be world No. 1 without having a fairly good game at both.
But as we shall see, there are exceptions to that general rule.
Taking a look down memory lane can provide interesting observations. The below provides statistical evidence to the return and hold game of the year end World No. 1-3 from 1998-2011.
If any of the readers want, they are very welcome to supplement the list with statistics from 1991-1997, found here. Pete Sampras was world No. 1 for four of these years, and he roughly had a hold game of 90-92 percent and a return game between 22 and 29 percent.
Before we start, I’ll just give you some general numbers as to what is needed in order to have the best, a good and a medium/bad hold and return game respectively.
If you win 90 percent of your own service games, you are at the top or in the top three on tour, more or less regardless of year. Winning 85 percent places you in or very close to the top 10, whereas 80 percent places you anywhere from No. 20 to No. 40 depending on year.
Players gunning for the world No. 1 shouldn’t fall much short of 85 percent, but there are exceptions as we’ll see below.
As for return game, the list below shows that you can get by with less in certain cases and still be world No. 1 (i.e. Andy Roddick in 2003).
There’s a bit more variation with regards to how much is needed to have the best return game in the world. Some years, winning 32 percent of your return games is enough. Other years, as 2011, the top of the list even goes above 40 percent.
On average, 35 percent will put you either at the very top or in the top three.
Generally speaking, though, a return game above 30 percent puts you in the top five or top 10 on tour, and anything above 25 percent will be good enough to put you in or very close to top 20 most years.
Now, here’s the year-end top three lists from 1998 onwards. I've added ranking points to show the difference between No. 1 and his closest rivals.
The first number is the hold game percentage, the second is the return game percentage and the third is ranking points earned:
1998: 1. Pete Sampras: 89/25 (3,915), 2. Marcelo Rios: 85/33 (3,670), 3. Alex Corretja: 80/28 (3,398)
1999: 1. Andre Agassi: 88/34 (5,048), 2. Yevgeny Kafelnikov: 79/28 (3,465), 3. Pete Sampras: 90/21 (3,024)
2000: 1. Gustavo Kuerten: 86/25 (4,195), 2. Marat Safin: 84/24 (4,120), 3. Pete Sampras: 91/18 (3,385) (none of them in the top 35 in return games won)
2001: 1. Lleyton Hewitt: 83/33 (4,365), 2. Gustavo Kuerten: 87/26 (3,855), 3. Andre Agassi: 84/32 (3,520) (Hewitt and Agassi leading return games won this year and the next).
2002: 1. Lleyton Hewitt: 80/33 (4,485), 2. Andre Agassi: 87/31 (3,395), 3. Marat Safin: 83/25 (2,845) (the lowest hold game percent for the No. 1 of the decade).
2003: 1. Andy Roddick: 91/21 (4,535), 2. Roger Federer: 87/29 (4,375), 3. Juan Carlos Ferrero: 83/30 (4,205) (the shortest distance in rankings points between the No. 1 and No. 3 in the decade and the worst return game by any No. 1 player in this period).
2004: 1. Roger Federer: 92/30 (6,335), 2. Lleyton Hewitt: 82/32 (3,655), 3. Andy Roddick: 91/22 (3,590)
2005: 1. Roger Federer: 89/31 (6,725), 2. Rafael Nadal: 84/38 (4.765), 3. Andy Roddick: 93/21 (3,085)
2006: 1. Roger Federer: 90/32 (8,370), 2. Rafael Nadal: 86/29 (4,470), 3. Nikolay Davydenko: 80/35 (2,825)
2007: 1. Roger Federer: 89/29 (7,180), 2. Rafael Nadal: 86/33 (5735), 3. Novak Djokovic: 84/28 (4,470)
2008: 1. Rafael Nadal: 88/33 (6,675), 2. Roger Federer: 89/27 (5,305), 3. Novak Djokovic: 87/30 (5,295)
2009: 1. Roger Federer: 90/24 (10,550), 2. Rafael Nadal: 84/34 (9,205), 3. Novak Djokovic: 85/31 (8,310)
2010: 1. Rafael Nadal: 90/29 (12,450), 2. Roger Federer:89 /27 (9,145), 3. Novak Djokovic: 82/32 (6,240)
2011 (as of now): 1. Novak Djokovic 87/41 (13,860), 2. Rafael Nadal: 84/35, (10,375) 3. Andy Murray: 80/37 (7,825) (These three also lead the tour in return games won.)
It should be clear that there is no strict correlation about percentage of games won and position in the ranking.
In 2009, both Rafa and Djokovic won more games, percentage-wise, than Federer, yet Federer must have won more games and matches that mattered, since he came out on top nonetheless. The same applies to Roddick versus Federer and Ferrero in 2003.
Nevertheless, the higher a total percentage you've got, the more games you've won and the higher the likelihood of ending the year as No. 1.
Concerning the ranking points, they changed the system in 2009. Roughly speaking, you need to multiply the old points by two to compare them to the new point system (a slam went from 1,000 to 2,000 points for example), but this doesn’t work all the way around.
For example, a slam final or a slam semifinal would be awarded 1,400 and 900 points today in the new system were we to multiply the old points by two. In fact, they only give 1,200 and 720 points. Given this, one should subtract a bit, especially if the player has a lot of runner-ups and semifinals rather than tournament wins. Nevertheless, it should give an indication.
So, what can we learn and have we learned from all the stats above?
It’s no surprise that Federer’s 2006 season is the best point-wise. After all, he played 97 matches and won 92, so while Djokovic has only lost three matches this year, he has still won almost thirty matches fewer.
We can also pinpoint the exact year from where Federer’s return game drops from being top-10 or better at 29 percent or above to more mediocre (2008 and onwards), while his hold game is still as good as ever.
We learn that Lleyton Hewitt managed to stay world No. 1. for two years, despite a hold game at 83 and 80 percent respectively, the two worst hold games for a No. 1 in the period.
He was leading return games won both years in front of Agassi.
We also learn that the No. 1 player, judging on both ranking points and percentage of games won, has been more dominant from 2004 onwards compared to the period before.
The least dominant player in the latter period is Federer 2009, who still has accumulated more ranking points than the most dominant in the period before 2004, Agassi in 1999. Yet Agassi had a bigger lead to the No. 2 than Federer had.
Finally, we note that people with great hold games can still be No. 1 without a terrific return game. Cases in point are Sampras 1998 (89/25), Roddick in 2003 (91/21) and Federer in 2009 (90/24).
The opposite, a player being No. 1 mainly by a good return game, is slightly less common, but Hewitt proved it could be done two years in a row.
Moreover, there are numerous examples of players above being second or third mainly on virtue of their return game (Kafelnikov, Davydenko, Djokovic and Nadal to name a few). Roddick and Sampras are the primary examples of players inside the top three on virtue of their hold game alone.
What is also interesting to note is that Rafa had his best hold game in 2010 at 90 percent and that it coincided with his worst (together with 2006) return game at still very respectable 29 percent.
2005, 2009 and 2011 represent Rafa’s worst hold game seasons, all with 84 percent. Against Djokovic this year, Rafa is only 60 percent in hold games won, which explains some of the drop from last year.
Andy Murray, who only made the year end top three once (strictly speaking, not yet), has the third best return game over all the years on these lists (second if we factor in how much Nadal played on clay in his best return game year in 2005, where he won eight clay tournaments. Needless to say, the percentage of return games won on clay is generally higher than on grass and hard court.
Yet, with an 80 percent hold game, Murray also has a shared second worse hold game for a top-three player from 1998 onwards.
Murray had a 85 percent hold game in 2009, and taking his height and general game into consideration, there’s really no reason, why he shouldn’t be at 85 percent or above every year.
Finally, Djokovic's 2011 numbers stand out.
It is by far the best return game on the entire list (41 percent) and combined with a top class hold game, 87 percent, he has actually won a far higher percentage of the games than anyone before him in this period.
He’s won 128 out of 200 percent, or 64 percent in total versus 122 or 61 percent for Federer in 04 and 06 and Nadal in 05 and quite a bit higher than Nadal’s 119 in 2010 as well (incidentally, Nadal has also won 119 in 2011).
Given that the similar number for the years from 2006 onwards for the World No. 1 is 122 (Fed 2006), 118 (Fed 2007), 121 (Rafa 2008), 114 (Federer 2009) and 119 (Rafa 2010), it seems that Novak with 128 has set a new standard high.
By this measure, Djokovic has indeed taken the game up a notch in 2011.
If we compare Djokovic’s 128 to world No. 1s earlier, like Sampras in 98 at 114 or Agassi in 99 at 112 or Hewitt in 2002 at 113 or Roddick in 03 at 112, we can see how just how good this is.
Is there an overall conclusion to be drawn?
Well, it seems to me that it’s become harder and harder to dominate without both having a good hold and return game. Federer’s poor return game at 24 percent, just outside the top 20, in 2009 is the aberration rather than the norm these years.
If Murray wants to play balls with the big guys, he needs to improve his hold game. If Federer wants to get anywhere near No. 1 again (unlikely), he needs a return game above 30 percent again at the very least.
Moreover, it appears that the return game has gained in importance.
With four of the top five taking the top-four positions in return games won in 2011 and Federer at a decent 27 percent at No. 15, it is abundantly clear why a guy like Isner with a terrific hold game will have a hard time getting near the top 10.
His return game is a meager 13 percent in 2011.
It also tells us why a big guy like Del Potro with the potential of an excellent hold game (88 percent this year and 84 percent in 2009) has a good chance to come back to the top, as his return game is pretty decent at 27 percent in 2009 and 26 percent in 2011.
However, he would have to improve either or both to compete for the top two spots.
The much heralded newcomer, Milos Raonic, would need to improve from his 17 percent return game to make his way into the top 10, let alone top five. His hold game is obviously there, but one trick doesn’t do the job these days.