Stats for 1999 Wimbledon final (Sampras-Agassi)

Score: 6-3, 6-4, 7-5

Sampras had 16 aces and 5 doubles.
Agassi had 5 aces and 6 doubles.

(The ATP site has the same count).

I counted 27 times that Sampras got a service return error from Agassi (of those, I judged 10 as service winners).

Agassi got 31 service return errors from Sampras (of those, I judged 2 as service winners).

That is remarkable, that Agassi got more of these "free" kind of points on his serve than Sampras did, considering the superiority of Sampras' serve. It just illustrates the quality of Agassi's return.

If you include the aces, then Sampras won 43 "free" points, Agassi 36.


Per the ATP site:

Sampras won 3 of 9 break points, Agassi 0 of 4.


My count of the winners:

Sampras had 30 winners: 8 FH, 4 BH, 9 FHV, 4 BHV, and 5 overheads.

Agassi had 22 winners: 10 FH, 7 BH, 0 FHV, 3 BHV, and 2 overheads.


Sampras had 4 service return winners, three from the forehand. Two of the winners came off Agassi’s first serve.

None of the return winners were passes, and he had just one passing shot, from the backhand.

Agassi returned Sampras’ second serve with four winners – two off each side. All of these were passes.

Agassi had 7 additional passing shots (four from the backhand).


Stats from NBC:

Per NBC, as of the fourth game of the second set Sampras was serving at 63%, with 26 winners and 6 unforced errors. Agassi was serving at 45%, with 10 winners and 8 unforced errors.

NBC credited Sampras with 58 winners and 22 unforced errors for the whole match.

(If NBC counted 30 non-service winners as I did, then NBC must have credited him with 11 service winners, compared to the 10 that I gave him. Note that they gave him 17 aces, rather than the 16 in the ATP count and my own; they credited him with an ace on a serve that Agassi's racquet barely grazed).


As of the fifth game of the second set, NBC had Sampras winning 18 of 25 at net, Agassi 2 of 4. As of the 10th game of the third set, Sampras was at 41 of 62, Agassi at 9 of 13.

I’ve found some new stats in the press, again contradicting the ATP stats.

This boxscore was in the Western Daily Press:

MATCH STATISTICS
PETE SAMPRAS (USA) v ANDRE AGASSI (USA)
17 Aces 5
5 Double Faults 6
60 First Serves In (%) 43
88 1st serve points won (%) 74
49 2nd serve points won (%) 50
43 Unreturned Serves 36
8 Forehand Winners 9
4 Backhand Winners 6
13 Volley Winners 6
1 Passing Winners 5
4 Return Winners 4
61 Returns In (%) 51
27/50 Net Points Won of Approaches 10/12
3/9 Break Points Won 0/4

I got the same numbers of Unreturned Serves, if I include aces along with return errors.

I got the same numbers of Return Winners.

It looks like the Passing Winners are non-return passes, and my own count agrees for Sampras: 1 such pass in the whole match. I have Agassi at 7, though, rather than 5 (it might depend on whether you count passes only as ground strokes hit near the back of the court; I count any groundies even if they're hit from near the net).

For Agassi I got 10 rather than 9 Forehand Winners, and 7 rather than 6 Backhand Winners.

I got 18 volley/overhead winners for Sampras (including 5 overheads), rather than the 13 Volley Winners in the box. And for Agassi I got 5 volley/overhead winners (including 2 overheads) rather than the 6 in the box.

The net stats are lower than NBC’s. Two games before the end, NBC already had Sampras at 41 of 62, Agassi at 9 of 13.

The break points line up with the ATP stats, but again it looks like the ATP figures for Total Points Won are wrong.

By my count Sampras made 31 return errors, Agassi 27. So if I use the % of “Returns In” provided by the boxscore, it looks like Sampras made 49 successful returns, while Agassi made 28. Add all these numbers to the aces and doubles, and you've got 167 points played in the match, compared to 223 reported by the ATP.

167 total points is more realistic, because it would yield 5.4 points per game. The ATP’s figures would yield 7.2 per game, which is just too high considering how well both players were serving -- particularly Sampras.

(The only good way of calculating the actual number of points played is to do your own count, so 167 can only be treated as an approximation. But to get to 223 the boxscore figures for % of "Returns In" would have to be seriously off).

Finally, the boxscore has Sampras serving at only 60% rather than the 66% he has at the ATP. The success rate on first serve, for both players, is close to the ATP’s. But on second serve Agassi moves up to 50% rather than 45% at the ATP, and Sampras moves up to 49% rather than the wholly unrealistic figure of 39% that the ATP gives him.

Sampras took more risk on the return, so he is bound to have more return errors. It's not about Sampras serve or Agassi return, it's just the way Sampras returns the serve.

So you're saying Sampras has a better return? Or Agassi doesn't take risks on the return? Lulz.

This BTW was the best performance I've seen, because Agassi was playing well. He just couldn't do anything.

i agree with superman1
this is probably the greatest performance ever in tennis: imo greatest match of all times even though it was too one-sided but its 100% sure that sampras from this match is unbeatable player
yes agassi did play extremely
but watching this match you could "feel" amazing power and energy coming from both sides unlike in fed-nadal matches

Well, the period from 3:3 in the first set to 3:1 in the 2nd set has to the best tennis that anyone player has put together against another man who was playing damn well, Sampras won 5 games in a row to really open up the final plus really he should have had a double break in the 2nd set because Agassi was struggling on literally every service game for most of the 2nd set.

Sampras hit a high level of grounstroke winners - both Pat Cash and Michael Stich said that Sampras matching Agassi in the baseline battles will be a key part of the match, also interesting because Sampras wasn't staying back on his second serve like he would on hardcourts, so those winners came on Agassi's service games - so he was really dominating Agassi from the net and the baseline. The grass court that year was quite firm and high bouncing, giving Agassi the platform to do well and get to the final.

In fact, Agassi's 3 set defeat of Rafter in the semifinal is some of the best tennis I've ever seen Agassi play - so really both men were in top form going into that final, although Agassi was made slight favourite because of his hot streak and the media figured (or hoped) that Sampras' run had to come to an end, they were saying the same things about Sampras in 1998.

Here's extensive clips of the match I put on youtube recently:

http://www.youtube.com/watch?v=XqmJYDZyuSo

http://www.youtube.com/watch?v=bXknRNY3woE&feature=related

http://www.youtube.com/watch?v=UUhAiChezHs

http://www.youtube.com/watch?v=f5d0LKwGxQ0

http://www.youtube.com/watch?v=OoGvfmbrxgg

By the way, looks like youtube have really improved the flash recenty - at last.

I remember, that Agassi had a low percentage of first serves in. I think i read somewhat in the 40% range. Not a big server, Agassi needed a better percentage of first serves. That was always a problem for Agassi against Sampras (or later Federer): He tended to overhit his first serve, and put too much pressure on his second serve. In the AO match of 2000, he served somewhat looser, and stayed in the match in crucial situations.

Agassi usually gets over 60% of first serves in when he's playing his A-game. So if he was only getting 40% of first serves in vs Sampras then that was probably the biggest reason for the lopsidedness. Agassi's 1999 Wimbledon semi-final straight-sets win over Rafter was extremely impressive, maybe his most polished performance ever.

what was the unfored error stat?

You never count that one.

Service percentages and total points won

I've done my own count of the service percentages and total points won. On those categories, again, the ATP stats are wrong. But as is usually the case when those are wrong, the ATP counts still look good for aces, double-faults and break points.


My count:

Sampras won 106 points overall, Agassi 85. That's a total of 191 points (compared to 223 at the ATP). So Sampras hit a clean winner or ace on 24% of all the points in the match (compared to 27% in the 2002 USO final).


Sampras won 67 of 93 points on serve, Agassi 59 of 98.


Sampras served at 59%, making 55 of 93 first serves.
Agassi served at 43%, making 42 of 98 first serves.

(NBC had Sampras at 61% early in the third set, and Agassi at 44% in his last service game. The ATP's percentage for Sampras is too high).


Sampras’ service percentages by set:
18/30 (60%)
19/27 (70.4%)
18/36 (50%)

Agassi’s service percentages by set:
12/28 (42.9%)
13/32 (40.6%)
17/38 (44.7%)


Sampras converted 3 of 9 break points, Agassi 0 of 4.

Sampras got his first serve into play on 4 of 4 break points, Agassi on 2 of 9. Each time that Agassi was broken it was on second serve.

Now the boxscore from Western Daily Press makes sense; I had read it incorrectly.

In the boxscore, Sampras got 61% of his “RETURNS IN”, Agassi 51% – meaning that Sampras failed to put 39% of his returns into play, and Agassi 49%. The boxscore gives the numbers behind the latter percentages: Sampras failed to return 36 serves successfully, Agassi 43 (I have the same numbers, if I include aces as unreturned serves). Calculating from the percentages, then, Sampras got back 56 serves, Agassi 45. If you add all those numbers – unreturned serves and successful returns – to the 11 double faults, you get 191 points played in the match, the same as my count.

So the category for % of “RETURNS IN” does not mean what it might sound like. When I first read it, I thought it referred to how many returns Agassi, for example, was able to get into play, out of the total number of returns that he hit. The category actually refers to how many he got into play, out of all the serves that Sampras got into the box – including those serves that Agassi never touched. It’s an index of how many good serves a receiver can return, not how many of his returns are good and how many are errors.

MATCH STATISTICS
PETE SAMPRAS (USA) v ANDRE AGASSI (USA)
17 Aces 5
5 Double Faults 6
60 First Serves In (%) 43
88 1st serve points won (%) 74
49 2nd serve points won (%) 50
43 Unreturned Serves 36
8 Forehand Winners 9
4 Backhand Winners 6
13 Volley Winners 6
1 Passing Winners 5
4 Return Winners 4
61 Returns In (%) 51
27/50 Net Points Won of Approaches 10/12
3/9 Break Points Won 0/4

Click to expand...

So my original estimate of 167 total points is, obviously, wrong.

Errors (forced and unforced)

Subtracting the winners and aces from the total points won, to get the total forced and unforced errors:

Sampras made 58 total errors. Of those, I counted 31 return errors and 5 double-faults.

Agassi made 60 errors. Of those, I counted 27 return errors and 6 double-faults.

Usually Agassi makes fewer total errors than Sampras, though he did make fewer return errors.

In all exchanges that had at least a successful return of serve, Sampras made 22 errors, Agassi 27.

Early on McEnroe noted how unexpectedly well Sampras was hitting his ground strokes in the rallies.

urban said:

I remember, that Agassi had a low percentage of first serves in. I think i read somewhat in the 40% range. Not a big server, Agassi needed a better percentage of first serves. That was always a problem for Agassi against Sampras (or later Federer): He tended to overhit his first serve, and put too much pressure on his second serve. In the AO match of 2000, he served somewhat looser, and stayed in the match in crucial situations.

Click to expand...

He felt pressure when he played Sampras and later Federer (although he was aging by then) because they are more athletic and overall talented players with more weapons then he had, which isnt the case at all with most people he played. He felt differently playing against guys like that then anyone else, and they got him off some of his own game and patterns. Pressing too much for sure, including on the serve as you said.

Boxscore from Wimbledon.org

Recently Moose told me about webarchive so I've been looking up some archived pages for stats.

This is the boxscore for Sampras-Agassi from Wimbledon.org, archived at http://web.archive.org/web/19991122183359/championships.wimbledon.org/scores/stats/day14/0127ms.html:



Whoever produced it made no effort to distinguish between types of errors, because the Unforced Errors leave virtually no room for forced errors: if you add up the Unforced Errors to the Winners and the double faults, you have almost exactly the Total Points Won.

And we know that NBC put Sampras at only 22 unforced errors.

The non-service winners implied in this boxscore – by subtracting the aces from the total winners – line up with my own count. Wimbledon.org, unlike NBC, didn’t have a category for service winners; or if they did, they didn’t include them in their row for “Winners.”

Service winners are, of course, a type of forced error; and in this chart the forced errors have all been collapsed into the Unforced Error category. Strange as it sounds, that's where the service winners appear in this box: as Unforced Errors.

I have Sampras at only 16 aces (what I call a service winner, they called his 17th ace). And I’ve got him slightly lower on service percentage. But all the other counts I did – the totals points won on serve, break points, and Agassi’s service percentage – all agree with this boxscore.

Note that the Winners category includes service while the Unforced Errors do not appear to include the double-faults. The double-faults need to be added on top of the Unforced Errors and the Winners in order to approximate the Total Points Won.

However, even then there is not an exact alignment, because adding up all those categories would leave Agassi a little bit short with only 83 points won, and Sampras slightly over the top with 107.

krosero said:

Recently Moose told me about webarchive so I've been looking up some archived pages for stats.

This is the boxscore for Sampras-Agassi from Wimbledon.org, archived at http://web.archive.org/web/19991122183359/championships.wimbledon.org/scores/stats/day14/0127ms.html:

Whoever produced it made no effort to distinguish between types of errors, because the Unforced Errors leave virtually no room for forced errors: if you add up the Unforced Errors to the Winners and the double faults, you have almost exactly the Total Points Won.

And we know that NBC put Sampras at only 22 unforced errors.

The non-service winners implied in this boxscore – by subtracting the aces from the total winners – line up with my own count. Wimbledon.org, unlike NBC, didn’t have a category for service winners; or if they did, they didn’t include them in their row for “Winners.”

Service winners are, of course, a type of forced error; and in this chart the forced errors have all been collapsed into the Unforced Error category. Strange as it sounds, that's where the service winners appear in this box: as Unforced Errors.

I have Sampras at only 16 aces (what I call a service winner, they called his 17th ace). And I’ve got him slightly lower on service percentage. But all the other counts I did – the totals points won on serve, break points, and Agassi’s service percentage – all agree with this boxscore.

Note that the Winners category includes service while the Unforced Errors do not appear to include the double-faults. The double-faults need to be added on top of the Unforced Errors and the Winners in order to approximate the Total Points Won.

However, even then there is not an exact alignment, because adding up all those categories would leave Agassi a little bit short with only 83 points won, and Sampras slightly over the top with 107.

Click to expand...

Outside of the unforced errors mess they have Pete at 27/50 at the net. You uoted NBC as having him at 41/62 in the 3rd set. I don't know?? I emailed NBC sports but haven't heard anything yet.

Sampras, Wimbledon final 1999, remains the greatest display of tennis I've seen on any court.

Nadal comes second with his Roland Garros 2008 crushing of Federer in the final, but a distant second!

As Agassi said, Sampras walked on water that day. But given a choice, I'd rather watch Sampras' tennis perfection of that day over anyone walking on water! Such was the level that the King of Swing took his tennis to. A match that will never be forgotten.

My stats :

Krosero, what's your count for Sampras's UE (just from the game, not bad returns or double faults)?

Nadal_Power said:

Krosero, what's your count for Sampras's UE (just from the game, not bad returns or double faults)?

Click to expand...

I didn't count them. I know that NBC gave Sampras 22 unforced errors of every kind for the whole match.

Nadal_Power said:

Code:

Time : 1 hour 54 minutes

Pete Sampras                                        Andre Agassi

87           Rallies 1-4 (157)                      70 
16           Rallies 5-8 (30)                       14 
3            Rallies 9-12 (4)                       1

Click to expand...

So in this case the rally length seems to make no difference. Sampras won 55% of the shortest points and 53% of the ones that went to 5-8 hits.

Even if you take out the 1-stroke points (aces/df's), it makes little difference: that leaves Sampras winning 52% of the rallies going to 2-4 strokes (Sampras wins 65, Agassi 60).

Sampras did win 3 of the 4 longest points, which may have increased the perception that he dominated Agassi from the baseline (those were probably dramatic points).

I just noticed that your rally counts include double-faults, as 1-stroke points (along with aces).

I'm not sure that's the way it's done in official stats but it makes sense, if errors are included as the last shot in a point (ie, unreturned serves are 2-stroke rallies). It's just that in the case of double-faults the error is the only stroke of the point!

I've done rally counts for some matches, can't recall how I did it. But counting double-faults as 1-stroke rallies does make sense.

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.

Quite extraordinary. His first serve average speed from 1993 equals his second serve average speed from 2001 loss to Federer.

Interestingly, his second serve average speed increased even more dramatically than his first serve average speed.

The question is are all these measurements comparable?

Bursztyn said:

Quite extraordinary. His first serve average speed from 1993 equals his second serve average speed from 2001 loss to Federer.

Interestingly, his second serve average speed increased even more dramatically than his first serve average speed.

The question is are all these measurements comparable?

Click to expand...

Well Pete started to SV a lot more, often on both serves, in the second half of the 90s. In order to do that he would have had to go for more on the second serve; and I think he says something like that in his book.

Of course he always came in behind both serves at Wimbledon; but if he beefed up his second serve in later years you would expect that to show up in his Wimbledon matches as well.

So that might explain why the readings for second serve went up more than those for first serve. But as for the overall increase, there's the argument that radar technology grew more sophisticated and able to capture the ball closer to the strings, when it's moving fastest. I would think that has to be factored into all this.

krosero said:

But as for the overall increase, there's the argument that radar technology grew more sophisticated and able to capture the ball closer to the strings, when it's moving fastest. I would think that has to be factored into all this.

Click to expand...

If technology made it possible to capture the ball closer to the strings I would expect the readings for the first and for the second serve to go up. Moreover the readings for the first serve should go up more than the readings for the second serve because the increase in reading should be positively correlated with the service speed (deceleration of the ball is positively linked to the service speed).

If this model is correct I would say that the increase in Pete's service speed can not be solely explained by the improvements in radar technology. Pete indeed has increased his service speed in Wimbledon during the 1990 s, however the effect of technology still might have played its role in the general increase of readings values.

Someone passed me along these links, what do you make of them?

http://sportsillustrated.cnn.com/tennis/1998/wimbledon/news/1998/07/03/wimbledon_menadv/

http://sportsillustrated.cnn.com/tennis/1998/usopen/news/1998/09/09/usopen_first/

The first has Sampras serving at 136 mph at the '98 Wimbledon.

The second has him reaching 134 mph at the '98 USO.

Could very well have been different measuring systems, particularly at different Slam venues.

Generally these articles support the notion that the readings of Pete serve went up in the second half of his career.

I agree with you that probably Pete must have served harder towards the end of his time, and that technology reasons may have facilitated the overall increase in serve speed readings as well.

Other factors that influence serve speed readings are random (weather conditions, opponents strengths and weaknesses) so they should not be responsible for any systematic increase in serve speed measurements.

I wouldn't be surprised if players generally started to serve harder in the course of 1990 s. Pete had to run to stay in the same place.

Bursztyn said:

Generally these articles support the notion that the readings of Pete serve went up in the second half of his career.

I agree with you that probably Pete must have served harder towards the end of his time, and that technology reasons may have facilitated the overall increase in serve speed readings as well.

Other factors that influence serve speed readings are random (weather conditions, opponents strengths and weaknesses) so they should not be responsible for any systematic increase in serve speed measurements.

I wouldn't be surprised if players generally started to serve harder in the course of 1990 s. Pete had to run to stay in the same place.

Click to expand...

Makes sense, I just wish we knew more about the individual readings and where they came from, and how they were done.