Incorrect ATP stats (a survey)
 

Many posters here have mentioned that the ATP site has a lot of mistakes when it comes to things like titles won, or scorelines. I'm starting this thread as a resource for ATP errors on match statistics.

I'll start with a list of published boxscores that all differ with the ATP stats.

Last week I ran a search in Lexis-Nexis for boxscores from 1991-2000. I just searched for the term "Total Points Won."

I found 7 boxscores. Six had serious discrepancies with the ATP stats on service/return percentages and Total Points Won/Played.

I sent an email to the ATP with a list. I can't say that the boxscores are definitely correct, but I do know that the boxscore I had found for the 1992 W final had accurate figures while the ATP did not.

I also know that the ATP stats for the 1995 AO final are wrong because they clash with my own counts and ESPN's.

On the other hand, while the ATP only agrees with 1 of 7 boxscores I found in that particular search, from time to time I have seen other boxscores that agree with the ATP.

Sometimes I've also seen, not entire boxscores, but individual stats that agreed with the ATP (for example, just the first-serve percentages). Occasionally these stats come from television coverage.

Also, when we've done our own counts, I can't recall a time when our numbers differed from the ATP stats by a lot. In fact we've done a few matches where our own counts for the total number of points agree with the ATP. We've also come up with slight discrepancies, but with the boxscores we're talking about large differences on things like total points played in the match.

Later I’ll make some posts on ATP stats that are look okay.

For now let me just list all the boxscores I know about that clash with the ATP stats on service/return percentages and Total Points Won/Played.

A list of the matches:


Agassi-Ivanisevic, 1992 Wimbledon final
Bruguera-Courier, 1993 French Open final
Courier-Edberg, 1993 Wimbledon semifinal
Sampras-Agassi, 1995 U.S. Open final
Krajicek-Washington, 1996 Wimbledon final
Sampras-Moya, 1997 Australian Open final
Korda-Sampras, 1997 U.S. Open
Korda-Rios, 1998 Australian Open final
Rafter-Philippoussis, 1998 U.S. Open final


And here are the actual boxscores. I've been able just to cut and paste them in here; the boxes are gone but the numbers look readable.


Agassi-Ivanisevic, 1992 Wimbledon final.
From USA Today, the St. Petersburg Times (Florida), the Gazette (Montreal), and the St. Louis Dispatch:
The men's singles championship box score (Agassi wins 6-7 (8-10), 6-4, 6- 4, 1-6, 6-4):
Agassi-Ivanisevic
Aces 9-37
Double faults 1-7
1st serve percentage 58-58
Pct. 1st serve points won 77-82
Pct. 2nd serve points won 64-50
Service games held 22-22
Break points against 5-16
Service games broken 2-3
Placement winners 49-39
Pct. backcourt points won 57-34
Points won at net 22-41
Total points won 163-159
Time of match 2:50


Bruguera-Courier, 1993 French Open final.
From the St. Petersburg Times:
How Bruguera won
Here is the box score of Sergi Bruguera's 6-4, 2-6, 6-2, 3-6, 6-3 victory over Jim Courier in the men's final at the French Open:
Bruguera Courier
Aces 7 5
Double faults 2 1
First service % 56 56
Break points won 26 14
Unforced errors 66 83
Forced errors 33 16
Total points won 153 146


Courier-Edberg, 1993 Wimbledon semifinal.
From the Miami Herald and the San Jose Mercury News.
Source: -- ASSOCIATED PRESS
Statistics from Jim Courier's 4-6, 6-4, 6-2, 6-4 victory over Stefan Edberg:
Category Courier Edberg
Aces 5 4
Double Faults 5 8
1st Serve Percentage 59 66
Pct. 1st Serve Points Won 77 69
Pct. 2nd Serve Points Won 50 31
Service Games Held 15 12
Break Points Against 6 14
Service Games Broken 4 7
Pct. Backcourt Points Won 42 34
Points Won at Net 19 51
Total Points Won 131 113
Time of match: 2 hours, 40 minutes.
Published on July 3, 1993, Page 3D, San Jose Mercury News (CA)


Sampras-Agassi, 1995 U.S. Open final.
The following boxscore appeared in (at least) the St. Petersburg Times and the Salt Lake Tribune:
Sampras-Agassi Statistics
Date: September 11, 1995
NEW YORK -- Statistics from No. 2 Pete Sampras’ 6-4, 6-3, 4-6, 7-5 victory over No. 1 Andre Agassi for the men’s singles championship of the U.S. Open Sunday:
Sampras Agassi
Total Points Won 124 112
Aces 24 8
Double Faults 6 5
1st Serve Percentage 54 56
Pct. 1st Serve Points Won 86 70
Pct. 2nd Serve Points Won 54 63
Break Point Conversions 4-8 2-6
Winners (including service) 60 33
Unforced Errors 40 33
Net approaches 43-59 13-20
Time of match 2:28


Krajicek-Washington, 1996 Wimbledon final, 6-3, 6-4, 6-3.
From the St. Petersburg Times and the Globe and Mail:
By the numbers
Kra. Was.
1st serve pct. 58 65
Aces 14 5
Double faults 3 2
1st srv. pts. won+ 88 61
2nd srv. pts. won+ 60 45
Brk. pts. converted 5-11 1-2
Net points won 25-36 21-44
Baseline pts. won 32-60 18-52
Total points won 93 66
Time of match 1:34
+-percentages


Sampras-Moya, 1997 Australian Open final, 6-2, 6-3, 6-3.
From The Age (Melbourne):
Pete Sampras (USA) 6 6 6
Carlos Moya (Spain) 2 3 3
THE MATCH
Sampras Moya
1st serve % 72% 63%
Aces 12 2
Double faults 2 3
% won on 1st serve 82% 58%
% won on 2nd serve 53% 41%
Winners 38 13
Unforced errors 22 18
Break-point conversions 100% 50%
Net approaches 80% 20%
Total points won 89 59
Match time: 1hr 27mins (sets 23, 28, 36)


Korda-Sampras, 1997 U.S. Open, 6-7 (4), 7-5, 7-6 (7), 3-6, 7-6 (3).
From the New York Daily News:
MATCH BOX SCORE
Korda Sampras
First serve 66% 58%
Aces 15 24
Double faults 6 5
First serve points won 78% 75%
Second serve points won 60% 50%
Winners 67% 58%
Unforced errors 45 66
Break points 3-12 4-8
Net approaches 16-23 74-124
Total points won 189 170


Korda-Rios, 1998 Australian Open final, 6-2, 6-2, 6-2.
From The Age (Melbourne):
THE MATCH
Korda Rios
1st serve % 52 58
Aces 4 2
Double faults 4 4
% won on 1st serve 78% 59%
% won on 2nd serve 57% 30%
Winners 32 7
Unforced errors 26 31
Break point conversions 58% 17%
Points won at net 53% 47%
Total points won 90 58


Rafter-Philippoussis, 1998 U.S. Open final, 6-3, 3-6, 6-2, 6-0.
From the Sydney Morning Herald:
US OPEN FINAL STATS
Rafter Philippoussis
Age 25 21
Seeding 3 unseeded
First serve % 72 56
Aces 6 5
Double faults 1 13
Winners 39 34
Unforced errors 5 29
Break pt conversions 6 of 20 1 of 3
Net approaches won 62 of 82 55 of 93
Total points won 118 85
Time of match: Two hours, 4 minutes


In every one of these cases, the ATP's aces, doubles, break points and number of service games look okay, or at least they resemble the boxscore figures. The remaining ATP stats are sometimes off by so much that they don't appear to belong to the match in question.


Some other discrepancies I've seen:

- For Agassi-Sampras, 1995 Australian Open, my own counts contradict the ATP stats on points won/played on service and overall.

- For Sampras-Hrbaty, 1997 Australian Open, 4th round (6-7 (7-4) 6-3 6-4 3-6 6-4), The Australian reported that Sampras led Hrbaty in total points by 157-154 (the ATP has 189-184); that Hrbaty led in aces by 20-17 (the ATP has 21-17); and that Hrbaty converted 3 of 10 break points (in agreement with the ATP).

Maybe if we get enough people e-mailing the ATP we'll get a response.

Should contact Q&M... Nice to try at least.

I assume you mean 67 & 58 winners in that last line?

I was watching this recently & late in the 5th, I think both players only had around 52 winners.

That one for some reason has percentage signs in the article I accessed. As with any boxscore it could be a typo, or incorrect data. Or just a different count from the network stats.

Borg now appaers with 63 titles... Connors with 107... Q&M? Who else write to ATP?

Other discrepancies
 

Here are a few more matches in which the ATP stats are different from stats in the print media or TV coverage:

- For Sampras-Ivanisevic, 1998 W final, NBC had counted 306 points as of 1-all, 30-15 in the fifth. I counted 27 more points, for a total of only 333 points in the match, compared to 404 reported at the ATP.

- For Sampras-Ivanisevic, 1995 W SF, the AP and Irish Times report that Sampras led in total points won by 146-145; the ATP has him trailing 175-185.

- For Becker-Agassi, 1995 W SF, the Washington Times had Agassi winning 21 of 53 second serves, while the ATP has 21 of 59.

- For Sampras-Becker, 1995 W final, the ATP stats differ from the Washington Post on the number of points that Becker won in return games (31 instead of the Post's 20), his aces (17 instead of 16) and his double-faults (16 instead of 15). And the ATP’s Total Points Won (309) would yield 7.9 points per game, higher than any confirmed average I know.

- For Sampras-Corretja, 1996 USO QF, the ATP has a total of 431 points played, while other sources have 375 (Washington Times, Atlanta Journal-Constitution, Orange County Register).

- For Sampras-Agassi, 1993 W QF, the ATP has 318 points played, but HBO's count with eight games left has only 224. I counted 48 points remaining, for a total of only 272. And the ATP lists the match as 8 minutes long.

I'm sending this excellent data to those guys, we'll see!

Who knows, it might help, thanks!

Aces and df's counted twice
 

I couldn't figure out what was wrong with the ATP stats until the discussion in the other thread prompted me to take another look today. I started adding up how many extra points Sampras was being given by the ATP, and that was the key.

It turns out that the ATP, in a lot of its matches, counted aces and double-faults twice in their figures for Total Points Won. The aces and df's themselves are listed correctly -- but they were included twice in the other rows (e.g., service percentages, success on first and second serve, Total Points Won, etc.)

Counting them only once resolves all the major discrepancies that I've listed in the posts above.

These are the ATP stats for the 1995 AO final:



They've got Agassi winning 151 points overall, and Sampras 154, even though Agassi won more games (and won the match).

Sampras was given 154 points because his aces, and Andre's double-faults, were counted twice.

28 aces + 4 df = 32 extra points

154 - 32 = 122 points won (in line with my own count)


And Agassi got his aces, and Pete's doubles, twice:

10 aces + 6 df = 16 extra points

151 - 16 = 135 points won (in line with my own count)


And this explains why Agassi, per the ATP, won fewer points than Ivanisevic despite beating him in the '92 W final: because he had only 9 aces while Goran had 37.


In fact all the boxscores I posted in the original post now line up with the ATP stats, when the aces and doubles are subtracted from the Total Points Won:

Almost all of them line up exactly with the ATP. With the '92 W final, subtracting the aces and df's actually leaves Goran with 157 points rather than the 159 that he has in my count and in the newspaper boxscore. It leaves Bruguera-Courier with a point margin of 152-145 instead of the 153-146 in the print media (partly because the ATP has blank boxes where the double-faults should go). And Sampras-Agassi at '95 USO is left with a point margin of 123-111 rather than what's in the boxscore and my count (124-112).

So the ATP's counting is not perfect; but otherwise it lines up exactly with published figures, once the aces and df's are subtracted.

The problem is that there's no way of knowing which matches in the ATP database have mistakes, other than to check against another source. A lot of ATP stats, esp. in the '90s, are incorrect. But even in that decade there are at least a few matches where the ATP does agree with published boxscores, or with other stats (including our own counts).

The only way to confirm an ATP boxscore is to check it against stats in a newspaper, tournament website, TV network, or your own count.

The total points won line up exactly if aces and df's are subtracted once.

Subtracting the 71 aces and double-faults results in NBC's 333 points.

Same agreement here.

Since most aces are first serves, it makes sense that Becker's first-serve percentage would be inflated at the ATP, since those first serves are being counted twice.

I've actually found stats for Becker's return games in this final that differ from one newspaper to another; a lot of disagreements. I don't know the correct figure but it does look like the ATP's figures for Total Points Won are inflated a lot, with these two men serving so many aces (and Becker serving so many double faults). If the aces and doubles are subtracted, it would leave Sampras winning 138 points overall, Becker 108. That's a total of 246, for a more believable average of 6.3 points per game.

The ATP has 375 too, once aces and df's are subtracted. And the points break down as they do in the print media: 188 for Alex and 187 for Pete.

I did this match recently and my total was actually 275, with Sampras leading 146-129 (as he does in the ATP stats if aces and df's are subtracted).

Incidentally I have Agassi facing 15 break points, while the ATP has only 13.

So the bias in the ATP stats, when they're wrong, is toward big servers like Sampras, Becker, Ivanisevic.

Not just their total points won, but their service percentages and success on first and second serve, are going to appear higher than they actually were.

I think I've read on this board a few debates about whether Agassi or Hewitt had the better return game. That's a good question, but if you use their stats against Sampras, it's more likely that Agassi's return stats are going to appear lower than they really were. Hewitt's stats are probably more accurate because I don't think I've found any ATP stats that are incorrect since around the year 2000. I do sometimes see the ATP differ with other sources by tiny amounts, but that's all. It's in the '90s that, so far, I've found all the big mistakes.

Thanks for info kevin!

More discrepancies
 

- For Edberg-Chang, 1992 USO SF, the Minneapolis Star-Tribune has Edberg leading 209-175 in overall points, for a total of 384. The Washington Post and SI have 404. The ATP has 436, but that drops down to 404 if the aces and doubles are subtracted once (with Edberg leading 209-195). So I'm inclined to think that 404 is the correct number; only the Star-Tribune disagrees. I'm not surprised that anyone made a mistake given the record length of the match (5:26 at the ATP and in the press reports). The Post, SI, and the ATP agree that there were 23 service breaks. The LA Times has Chang converting 11 of 34 break points (same as the ATP) and facing 20 himself; the ATP has him facing 21.

- For Becker-Pioline, 1995 W QF, Becker was broken early in the fifth set according to several sources but the ATP has him unbroken. Each man served 28 times but the ATP has each serving 20. So those stats are incomplete in a way I haven't seen anywhere else; and I don't know if anything is going on with the aces and doubles.

- For Sampras-Agassi, 1999 W final, I have Sampras leading 106-85 in overall points, and so does the ATP if aces and doubles faults are subtracted once.

- For Agassi-Sampras, 1992 RG QF, I have Agassi leading 102-77. The ATP's numbers drop down to 103-78 if aces and df's are subtracted. And I counted 8 aces by Sampras; the ATP has 6.

Finally, I did my own stats for Sampras-Agassi, 1995 USO, and they agree with the boxscore posted in the OP. The ATP has the total points wrong as well as the break points: they have Sampras converting 4 of 7 but he was at 4 of 8 in my count and in the boxscore.

Next I'll post matches for which the ATP stats look good.

Actually success on second serve is going to skew low in the ATP stats, if double-faults are being counted twice. And that would often affect the big servers, or anyone who had a bad day double-faulting.

Doing a quick review of the boxscores above from the print media, they do give players greater success on second serve than the ATP does, but less success on first serve.

Two discrepancies
 

I’d missed these two boxscores in my notes. They’re the same as the boxscores above: they disagree with the ATP because the latter has counted aces and double-faults twice.

Sampras-Ivanisevic, 1994 Wimbledon final.
From the St. Petersburg Times (Florida) and the Washington Times:
How Sampras won
PS GI
Total Points Won 118 88
Aces 17 25
Service winners 54 44
Double Faults 5 6
1st Serve Percentage 50 62
Pct. 1st Serve Won 90 69
Pct. 2nd Serve Won 60 45
Service Games Held 15 12
Service Games Broken 0 3
Break Points Against 2 11
Placement Winners 35 33
Pct. Backcourt Pts Won 63 34
Advances to Net 34 47
Points Won at Net 22 18
Time of Match - 1:55
(The ATP lists the running time as 6 minutes).


Kuerten-Bruguera, 1997 French Open final.
From the Miami Herald:
Source: Herald Sports Staff
The boxscore of the men's singles championship of the French Open:
Kuerten Bruguera
Aces 4 1
Service Winners 11 9
Double Faults 0 1
1st Serve % 48 58
% 1st Serve Pts Won 82 58
Service Points Won 58 59
Service Break Pts 18 9
Break Points Saved 8 12
Return Points Won 53 25
Net Points Won 16 6
Unforced Errors 47 40
Total Points Won 111 84
Time of Match 1:50

ATP stats that look good
 

I've put together a list of matches in which the ATP figures for Total Points Won are supported by at least one source.

By support I don't always mean exact alignment. For some of these matches there are still minor discrepancies between the ATP and other sources, and I've described those below as well as I can pick them out.

But in all these cases it appears, at least, that the ATP has not counted aces and double-faults twice.

The list runs through 2003.


- For Courier-Agassi, 1991 RG final, the Associated Press reports that Courier won 145 points, the same number as the ATP.

- For Stich-Edberg, 1991 W semifinal, the Sun Herald (Sydney) confirms that Stich trailed 143-149 in Total Points Won.

- For Connors-Krickstein, 1991 USO, R16, I completed a CBS stat late in the fifth set for total points played. I ended within one point of the ATP figure of 374.

- For Sampras-Becker, 1996 TMC final, the New York Times and the ATP both have Sampras trailing by 166-178 in Total Points Won.

- For Federer-Sampras, 2001 W, R16, the BBC’s service percentages support the ATP’s.

- For Safin-Hewitt, 2002 Paris final, I completed a late-match stat by ATP Masters Series TV and got 192 points played in total, with Safin leading 108-84, in agreement with the ATP.


I also have some boxscores.

Stich-Becker, 1991 Wimbledon final
6-4, 7-6 (4), 6-4
From the St. Petersburg Times (Florida) and The Gazette (Montreal):
Stich vs. Becker
Aces 15 10
Service winners 51 40
Double faults 3 4
First serve pct. 46 58
Return winners 26 20
Total points won 111 103
Break points won 4-16 2-3
No disagreements with the ATP.


Agassi-Medvedev, 1999 French Open final
1-6, 2-6, 6-4, 6-3, 6-4
From the Globe and Mail:
Agassi Medvedev
Total points won 144 152
Aces 2 23
Winners incl. srv 46 65
Double faults 6 8
1st serve pct 62 63
Pct. 1st serve pts 69 74
Pct. 2nd serve pts 46 41
Break pt. conv 4-13 5-12
Unforced errors 56 65
Net points 22-33 17-30
Time of match 2:51
The ATP actually has Agassi at 145 points won. The other stats line up, though.


Agassi-Martin, 1999 U.S. Open final
6-4, 6-7 (5), 6-7 (2), 6-3, 6-2
From the Miami Herald:
Source: Herald Sports Staff
Agassi Category Martin
64 1st serve percentage 60
7 Aces 23
4 Double faults 8
75 Percent 1st serve points won 71
66 Percent 2nd serve points won 59
39 Winners (inc. service) 74
23 Unforced errors 60
5-10 Break points won 0-8
16-19 Net points won 62-97
165 Total points won 147
* Time of match: 3 hours 23 minutes.
The Sydney Morning Herald had the same unforced errors but it had Agassi trailing in winners by 45-77; other than that there are no discrepancies between these two sources or between either one and the ATP.


Agassi-Sampras, 2000 Australian Open semifinal
Boxscore from CNN.si and The Age (Melbourne):
Following are key statistics from Andre Agassi's
6-4, 3-6, 6-7 (0-7), 7-6 (7-5), 6-1 semifinal win
over Pete Sampras (Reuters)
Agassi Sampras
1st serve (percentage) 68 63
Aces 13 37
Double faults 3 5
Winning % 1st serve 74 81
Winning % 2nd serve 68 49
Winners 52 86
Unforced errors 19 56
Break points converted 3 of 13 , 1 of 9
Net approaches won 19 of 26 , 71 of 122
Total points won 155 149
Source: Reuters
The figures are the same as the ATP’s.


Agassi-Kafelnikov, 2000 Australian Open final
From CNN.SI:
Following are key stats from Andre Agassi's
3-6, 6-3, 6-2, 6-4 win over Yevgeny Kafelnikov.
Agassi Kafelnikov
1st serve (percentage) 66 53
Aces 9 14
Double faults 5 9
Winning % 1st serve 68 65
Winning % 2nd serve 59 53
Winners 38 53
Unforced errors 35 61
Break points converted 6 of 17 3 of 7
Net approaches won 9 of 12 32 of 52
Total points won 125 113 Source: Reuters
The ATP stats agree with these.


Sampras-Rafter, 2000 Wimbledon final
From the Washington Post:
The box score of No. 1 Pete Sampras's
6-7 (12-10), 7-6 (7-5), 6-4, 6-2 victory
over No. 12 Patrick Rafter for the men's singles
championship at Wimbledon:
Sampras Rafter
First Serve Pct. 62 57
Aces 27 12
Double Faults 12 8
Unforced Errors 79 55
First Serve Winning Pct. 89 75
Second Serve Winning Pct. 56 50
Winners (Inc. Service) 40 29 [these do not include service]
Break Points 3-14 0-2
Net Approaches 28-49 39-93
Total Points 160 138
Time of Match: 2:52
The ATP has Sampras at 24 aces, and Rafter at 10 aces and 7 doubles. They have Sampras winning 161 points. And they disagree on first-serve percentages and success on first and second serve (but always disagreeing by two percentage points or less).



Safin-Sampras, 2000 U.S. Open final
Xinhua General News Service:
Following are key statistics from Sunday's men's final at the U.S. Open in which sixth-seeded Russian Marat Safin beat fourth-seeded American Pete Sampras 6-4, 6-3, 6-3
SAMPRAS SAFIN
Age 29 20
Seeding 4 6
First serve points won 38 of 50 34 of 40
Second serve points won 12 of 29 26 of 42
Aces 8 12
Double faults 4 2
Winners (including service) 32 37
Unforced Errors 25 12
Break point conversions 0-of-2 4-of-9
Net approaches won 43-of-82 5-of-12
Total points won 72 89
Time of match: 1:38
This lines up with the ATP in all categories.



Ivanisevic-Rafter, 2001 Wimbledon final
6-3, 3-6, 6-3, 2-6, 9-7
This boxscore is from USA Today:
Ivanisevic Rafter
First-serve % 55 63
Aces 27 13
Double faults 16 4
Unforced errors 30 11
First serve winning % 81 77
Second serve winning % 51 56
Winners (including service) 42 49
Break points 3-6 3-6
Net points 38-70 49-70
Total points won 154 150
Time of match: 3:01
The ATP has Ivanisevic making 1 less double-fault and winning 1 more point overall.


Sampras-Agassi, 2001 U.S. Open QF
The Advertiser:
PETE SAMPRAS d ANDRE AGASSI
6-7 (7-9), 7-6 (7-2), 7-6 (7-2), 7-6 (7-5)
Match time: 3 hrs, 33 min.
Sampras Agassi
62 First serve pc 65
25 Aces 18
12 Double faults 4
10 Forehand winners 20
12 Backhand winners 13
40 Unforced errors 19
0 of 6 Break point conversions 0 of 3
96 of 137 Pts won at net 12 of 21
176 Total points won 162
This lines up with the ATP in all categories.


Hewitt-Sampras, 2001 U.S. Open final,
7-6 (4), 6-1, 6-1
New York Daily News:
HEWITT SAMPRAS
1st Serve percentage 64 60
Aces 7 11
Double Faults 5 6
Unforced Errors 13 38
1st Serve Winning Pct. 84 63
2nd Serve Winning Pct. 50 43
Winners (inc. service) 36 35
Break Points 6-13 1-2
Net Points 8-8 49-98
Total Points Won 100 76
Time of Match 1:54
These figures match the ATP’s exactly, except that the ATP rounded all its figures down.


Sampras-Agassi, 2002 U.S. Open final
6-3, 6-4, 5-7, 6-4
The New York Daily News:
GRAPHIC: FINAL BREAKDOWN
1st Serve percentage 66 57
Aces 7 33
Double Faults 4 13
Unforced Errors 21 46
1st Serve Winning Pct. 67 80
2nd Serve Winning Pct. 53 52
Winners (inc. service) 27 84
Break Points 2-12 4-8
Net Points 10-13 60-105
Total Points Won 126 151
Time of Match 2:54
This lines up with the ATP in all categories.



Agassi-Schuettler, 2003 Australian Open final
USAToday.com:
Agassi-Schuettler box score
Andre Agassi (2) def. Rainer Schuettler (31), 6-2, 6-2, 6-1
Schuettler Agassi
1st Serve percentage 51 67
Aces 4 7
Double Faults 4 1
Unforced Errors 25 13
1st Serve Winning Pct. 58 74
2nd Serve Winning Pct. 28 74
Winners (including service) 16 30
Break Points 1-1 8-13
Net Points 12-16 11-14
Total Points Won 43 80
Time of Match 1:16
The ATP stats are no different.


I haven't listed any more recent matches because I often see the ATP stats corresponding to stats on tournament websites.

Edberg-Courier
 

Finally there’s this unique case. Here’s a match in which the aces and doubles were not counted twice, and there’s only a minor discrepancy in Total Points Won between the ATP and another source, but there are nevertheless significant discrepancies (first-serve percentage, success on first and second serve).


Edberg-Courier, 1991 U.S. Open final,
6-2, 6-4, 6-0
USA Today's boxscore:
Edberg Courier
1st serve percentage 56 55
Service winners 1 0
Aces 3 3
Double faults 1 4
Pct. 1st serve points won 84 59
Pct. 2nd serve points won 69 38
Service games held 12 6
Winners 33 16
Unforced errors 18 20
Pct. break point conversions 43 0
Pct. advances to net 72 55
Points won at net 68 12
Total points won 96 61
Time of match 2:02

The ATP has Courier serving 10 percentage points higher than this, and Edberg is up by 4 points. In success on first serve, Courier is down 4 points, Edberg down 2. Less significantly, they have Edberg at 2 aces and Courier winning 59 points.

The St. Petersburg Times and the Associated Press agree with USA Today that Edberg's success on first serve was 84%.

So this match (see also Becker-Pioline above) suggests that there may be significant mistakes in the ATP database that have nothing to do with counting aces and df's twice.

Ivanisevic vs. Krajicek and Norman
 

Ivanisevic-Krajicek, 1998 Wimbledon sf, 6-3, 6-4, 5-7, 6-7, 15-13.

That's 71 service games altogether, plus one tiebreak. But the ATP -- despite having the correct score -- reports only 64 service games.

And they have the break points wrong, too. They've got Ivanisevic 3/3 on breakers, Krajicek 2/4. But the New York Times has Goran at 5 of 6: he broke three times in the first four sets (serving for the match in the fourth), and twice more in the fifth.

In fact the ATP's break points look like they might be confined to the first four sets.

Tennis Magazine (“Where Are the Big Guns?” by Jon Levey, August 2008 ):

Those stats may be taken from the ATP boxscore, which also has 290 total points. The 91 points from aces and double-faults may be fine (the aces are confirmed in the Times), but I doubt that only 290 points were played over the full 72 games. It would mean an average of 4.03 points per game, almost numerically impossible.

Norman-Ivanisevic, 1997 Wimbledon, 2nd round,
6-3, 2-6, 7-6, 4-6, 14-12.

That's 65 service games, plus one tiebreak. The ATP boxscore reports only 39 -- as if the 26 games of the fifth set had been dropped.

The Dallas Morning News had a boxscore:

That's 398 total points, compared to 302 at the ATP.

Rusedski-Stark, 1997 Wimbledon, R64 (same day as Norman-Ivanisevic), 4-6, 6-7, 6-4, 6-3, 11-9.

The ATP, for some reason, lists the fifth set as 11-9(5).

The ATP reports only 53 service games, but in the scoreline (confirmed at The Independent) there are 61.

Sampras-Korda, 1997 Wimbledon R16, 6-4, 6-3, 6-7, 6-7, 6-4

Dallas Morning News:

The ATP has Sampras leading in total points 222-178 (and hitting 27 aces), but if you subtract the aces and df's one time, you get the numbers in the boxscore.

The line for the Winners appeared with percentage signs, but the numbers look okay. Per the Cincinatti Post, “Sampras fired 58 winners, including 24 aces.” Per The Nation, Korda "crushed an astounding 67 winners and drew 66 unforced errors from Sampras."