Stats for Federer-Sampras (2001 Wimbledon)

[TheFifthSet]

If you're going to head that route by saying one match doesn't tell the whole story then there's no argument to say Agassi is better returner than Fed on grass. In that case Del Potro isn't > Nadal at the USO or Cash isn't > Lendl at Wimbledon because it's only one match. If you feel that way, fine. For me, one match is better than nothing, at least we have some data to look into.

Of course there is. How about you compare their return stats on grass? That's a start. How about compare their return stats against big servers on grass? You take refuge in one match when there's a whole body of data, contemporary opinions and common sense at hand. To imply there is no other data is patently false. The burden of proof rests on the one making the outsanding claim.

 

[14line]

Federer looked unstoppable in the 5th set, making a superb half volley when he was down a BP at 4-4. In this sense, the match is somewhat similar to the 2012 Nadal/Rosol match, where Rosol made only 2 UEs in the deciding set.

It's also interesting to comment on Pete's low winning percentage on 2nd serve. Remember in '99 final, Pete played his greatest grass court tennis, hitting an average of 109mph 2nd serve, but won only 49% of them. Note that Pete played much better at the net and hit fewer double faults in that match. So I think it's not the return on 2nd serve that made Federer win the match. What really impressed me was he handled Pete's powerful 1st serve well in this match; cf. Safin in USO 2000 final.

 

[krosero]

So I think it's not the return on 2nd serve that made Federer win the match. What really impressed me was he handled Pete's powerful 1st serve well in this match; cf. Safin in USO 2000 final.

Peter Fleming was one of the BBC's commentators and he thought that Federer's ability to return Sampras' first serve was a significant factor, and maybe Federer's only statistical edge (in all other categories they were pretty much tied with each other).

At 2-3 in the fourth, Federer was getting 46% of first serves and 71% of second serves back. Sampras was at 41% and 64%.

So Federer really had an edge getting back both 1st and 2nd serves. But the edge in returning 1st serves was probably of greater importance, simply because there were many more first serves than second serves in the match.

 

[krosero]

Moose has Federer making his first serve on 9 of 11 break points, Sampras on 10 of 14.

By my count:

Federer made 6 of 12 first serves in the tiebreaks. By tiebreak:

4 of 8 (he won all 4 of those first serves)
2 of 4 (he won 1 of those first serves)

Sampras made 8 of 13 first serves in the tiebreaks. By tiebreak:

4 of 8 (he won all 4 of those first serves)
4 of 5 (he won all 4)

So every time that Sampras made his first serve in the tiebreaks he won the point.

In the first tiebreak (won by Federer 9-7), Sampras put in a second serve at 121 mph (11 mph faster than his average for the match.

In the next tiebreak (won by Sampras 7-2), Pete’s only second serve was 120 mph. One of his first serves was at 136 mph – his high for the match. Another was at 134 mph.

Federer came in behind most of his serves in the tiebreaks; but in the first tiebreak he stayed back twice on second serve (losing one point), and in the next tiebreak he stayed back once on second serve (losing the point).

Sampras came in behind all his serves as he did throughout the match.

 

[14line]

Peter Fleming was one of the BBC's commentators and he thought that Federer's ability to return Sampras' first serve was a significant factor, and maybe Federer's only statistical edge (in all other categories they were pretty much tied with each other).

At 2-3 in the fourth, Federer was getting 46% of first serves and 71% of second serves back. Sampras was at 41% and 64%.

So Federer really had an edge getting back both 1st and 2nd serves. But the edge in returning 1st serves was probably of greater importance, simply because there were many more first serves than second serves in the match.

Sampras' return of serve deteriorated sharply in the last few years of his career, and I think this cost him two important matches - AO 2000 against Agassi and Wimby 2001 against Federer. Sampras has often said that it’s the return of serve which wins Wimbledon. I think it also explains why he lost the match against Federer.

 

[krosero]

Sampras' return of serve deteriorated sharply in the last few years of his career, and I think this cost him two important matches - AO 2000 against Agassi and Wimby 2001 against Federer. Sampras has often said that it’s the return of serve which wins Wimbledon. I think it also explains why he lost the match against Federer.

That's true, his return stats did deteriorate in the last years of his career.

 

[krosero]

I've been able to compile these service speeds for Sampras at Wimbledon.

1993 final
Fastest Serve – 123 mph
Average 1st Serve – 110 mph

1995 final
Fastest Serve – 129 mph
Average 1st Serve – 116 mph
Fastest 2nd Serve – 106 mph
Average 2nd Serve – 95 mph

1996 loss to Krajicek
Fastest Serve – 126 mph
Average 1st Serve – 117 mph
Average 2nd Serve – 100 mph

1999 final
Fastest Serve – 131 mph
Average 1st Serve – 119 mph
Average 2nd Serve – 109 mph

2001 loss to Federer
Fastest Serve – 136 mph
Average 1st Serve – 121 mph
Average 2nd Serve – 110 mph

In every category there is a slight progression upwards, with only one exception: in '95 his fastest serve was 129, the next year it was 126. But his average 1st serve still went up that year.

So how much of this is genuine increase in speed by Sampras? Or is it due to increasingly accurate radar measurements that record the speed of the ball closer to the point that it leaves the racquet strings?

 

[scotus]

So how much of this is genuine increase in speed by Sampras? Or is it due to increasingly accurate radar measurements that record the speed of the ball closer to the point that it leaves the racquet strings?

Hard to tell, but I believe Sampras' faster serves would be clocked at 130-140 mph by today's radar guns.

Here's why:

(1) I saw Sampras' exhibition match against Courier during a WTA tournament a few years ago. His fastest serve that evening was clocked at 136 mph down the side line. I'm sure if that same serve were aimed at the T, it would have been clocked at around 140 mph. If a retired Sampras can serve that fast, the prime Sampras would have served in the 130s routinely.

(2) In The Tennis Channel's "Signature" featuring Sampras, Andre Agassi describes returning Pete's serves as having to defend 135-mph serves to the deep corners and also guard against a 140-mph serve straight to the body. I think Andre who had such a long career and experience of returning those 120-mph serves of Pete, Becker, Goran, as well as the 140-mph serves of the younger greats is uniquely qualified to comment on how Pete's serves would be clocked on radar today.

 

[corners]

I've been able to compile these service speeds for Sampras at Wimbledon.

1993 final
Fastest Serve – 123 mph
Average 1st Serve – 110 mph

1995 final
Fastest Serve – 129 mph
Average 1st Serve – 116 mph
Fastest 2nd Serve – 106 mph
Average 2nd Serve – 95 mph

1996 loss to Krajicek
Fastest Serve – 126 mph
Average 1st Serve – 117 mph
Average 2nd Serve – 100 mph

1999 final
Fastest Serve – 131 mph
Average 1st Serve – 119 mph
Average 2nd Serve – 109 mph

2001 loss to Federer
Fastest Serve – 136 mph
Average 1st Serve – 121 mph
Average 2nd Serve – 110 mph

In every category there is a slight progression upwards, with only one exception: in '95 his fastest serve was 129, the next year it was 126. But his average 1st serve still went up that year.

So how much of this is genuine increase in speed by Sampras? Or is it due to increasingly accurate radar measurements that record the speed of the ball closer to the point that it leaves the racquet strings?

I think it's definitely a measurement issue. I don't think a server of Sampras' quality could possibly increase his fastest 1st serve by a full 10 mph between the ages of 23 and 30.

I suppose it could be possible that he was spinning it more on both 1st and 2nd when younger, which would mean that his accuracy got so good as he got older that he could hit flatter and still keep his percentages up. But this seems unlikely.

So, by the numbers, it looks like the measurement method changed sometime between '96 and '99. Should we really be adding roughly 10 mph to serves clocked before '99? This would make sense. I remember seeing some serve speed stats for Becker at Wimbledon and being somewhat shocked that his numbers were very similar to Federer's typical tournament averages: Fastest 1st ~ 130 mph, Average 1st ~ 116 mph, Average 2nd ~ 95 mph. After seeing that I thought that Becker may have had the most overrated serve of all time, which is a silly notion easily disproved by watching a match like the '96 Masters final. Adding 10 mph would give Becker: Fastest 1st ~ 140 mph, Average 1st ~ 126 mph, Average 2nd ~ 105 mph. Is that more realistic or less?

But, given that Federer's numbers in 2001 are nearly identical to those he posts today, I think we can be pretty sure that Sampras' 2001 numbers can be compared directly to serve speeds of today's players. And by that standard, his 121 avg. 1st is very good and his 110 avg. 2nd is simply amazing.

And yet, despite that incredible 2nd serve, his career 2nd serve points won percentage (53%) is lower than Federer's (56%), whose avg. 2nd serve speed is nearly 15 mph slower than Sampras' was.

 

[Bursztyn]

I think it's definitely a measurement issue. I don't think a server of Sampras' quality could possibly increase his fastest 1st serve by a full 10 mph between the ages of 23 and 30.

Why is it so? If Sampras was such as good server (I guess this is what you mean) then perhaps he could increase his fastest 1st serve by a full 10 mph between the ages of 23 and 30.

Had the increase in serve speed readings been solely caused by measurements technology I would expect the readings for the first serve to go up more than the readings for the second serve because the increase in reading should be positively correlated with the service speed.

So, by the numbers, it looks like the measurement method changed sometime between '96 and '99.

We should look at serve speed readings stats for other players from that era to see if similar trends can be observed.

And yet, despite that incredible 2nd serve, his career 2nd serve points won percentage (53%) is lower than Federer's (56%), whose avg. 2nd serve speed is nearly 15 mph slower than Sampras' was.

Roger hasn't finished his career yet, so these stats are not comparable at the moment.

 

[krosero]

I think it's definitely a measurement issue. I don't think a server of Sampras' quality could possibly increase his fastest 1st serve by a full 10 mph between the ages of 23 and 30.

I suppose it could be possible that he was spinning it more on both 1st and 2nd when younger, which would mean that his accuracy got so good as he got older that he could hit flatter and still keep his percentages up. But this seems unlikely.

So, by the numbers, it looks like the measurement method changed sometime between '96 and '99. Should we really be adding roughly 10 mph to serves clocked before '99? This would make sense. I remember seeing some serve speed stats for Becker at Wimbledon and being somewhat shocked that his numbers were very similar to Federer's typical tournament averages: Fastest 1st ~ 130 mph, Average 1st ~ 116 mph, Average 2nd ~ 95 mph. After seeing that I thought that Becker may have had the most overrated serve of all time, which is a silly notion easily disproved by watching a match like the '96 Masters final. Adding 10 mph would give Becker: Fastest 1st ~ 140 mph, Average 1st ~ 126 mph, Average 2nd ~ 105 mph. Is that more realistic or less?

But, given that Federer's numbers in 2001 are nearly identical to those he posts today, I think we can be pretty sure that Sampras' 2001 numbers can be compared directly to serve speeds of today's players. And by that standard, his 121 avg. 1st is very good and his 110 avg. 2nd is simply amazing.

And yet, despite that incredible 2nd serve, his career 2nd serve points won percentage (53%) is lower than Federer's (56%), whose avg. 2nd serve speed is nearly 15 mph slower than Sampras' was.

Sampras' numbers on second serve are calculated incorrectly in the ATP database; they are too low.

Here's a basic summary of the problem: http://tt.tennis-warehouse.com/showthread.php?p=6127075#post6127075

Essentially, for several years Sampras had his double-faults counted twice in the category of 2nd serve points won.

And his aces were counted twice in 1st serve points won.

1999 Wimbledon final, Wimbledon.org had him winning 50% of second serve points, and so did I in my own count. The ATP has him at 39%.

 

[corners]

Sampras' numbers on second serve are calculated incorrectly in the ATP database; they are too low.

Here's a basic summary of the problem: http://tt.tennis-warehouse.com/showthread.php?p=6127075#post6127075

Essentially, for several years Sampras had his double-faults counted twice in the category of 2nd serve points won.

And his aces were counted twice in 1st serve points won.

1999 Wimbledon final, Wimbledon.org had him winning 50% of second serve points, and so did I in my own count. The ATP has him at 39%.

I've read you talking about this issue before in relation to other things, but thanks for linking to a post dedicated to this issue, as I was unclear on just what it was all about. Is ATPWorldTour.com aware of this issue?

 

[krosero]

I've read you talking about this issue before in relation to other things, but thanks for linking to a post dedicated to this issue, as I was unclear on just what it was all about. Is ATPWorldTour.com aware of this issue?

I have sent them the links above, but beyond that it's anyone's guess whether the stats can or will be fixed.

 

[corners]

I have sent them the links above, but beyond that it's anyone's guess whether the stats can or will be fixed.

I'll send them an email too. I think it's important, if ATP is going to keep an archive of stats, that those stats are accurate!

 

[Mitcheson]

Actually Sampras was almost 30 and Federer almost 20, both born in early August. Very little in it and it could have gone either way. Likely that Sampras was not expecting the opposition to be this good or may be he would have been more fired up. Consider also that Federer lost to Henman next, whom Sampras had dominated. Sampras lost to journeyman George Bastl the next year playing absolutely awful and obviously had lost his form and motivation to a great degree.

Because of his fantastic serve and athleticism Sampras had a tendency to just do enough and was relatively lazy. He didn't like to be patient or tolerate long points, which contributed to him not winning the French. No surprise his motivation dropped once he'd beaten Emerson's record.

Federer is a completely different character who loves playing tennis. His footwork is superb and he looks to improve everything leaving no stone unturned. He benefitted from a dearth era and clocked up a lot of slams until Nadal and Djokovic matured (they were 5 and 6 years younger) but then he had these incredibles to contend with. He also benefitted from opportunities when Nadal was taken out of his way in 2009 at the French and much later when Djokovic was out of the way due to injury, but there again he still had to take his chances and earn his victories.

Djokovic, like Sampras, has been dominant against the best since his early big deficit in h2h against Nadal and Federer whereas Federer has incredible longevity. He was once 6-13 down to Federer but has won 21 of the last 31 matches against him. Obvious age and experience benefitted Federer in the early years but worked against him in the latter ones.

It makes more sense to appreciate them rather than say one is better than the other e.g. is longevity more important than dominance? McEnroe was awesome for a few years but didn't last long at the very top.