Djokovic also very good. Dimi likewise. Alcaraz and Shelton with plenty of room to improve. Rune and Sinner as well, albeit to a lesser extent (especially Sinner; quite good, but will of course be even tougher to beat if he can improve even more). De Minaur shockingly inaccurate on average. But see the next picture in the IG post for specific tournament averages. Both Sinner (winner) and De Minaur (runner-up) serving very well in Rotterdam.
R
Djokovic also very good. Dimi likewise. Alcaraz and Shelton with plenty of room to improve. Rune and Sinner as well, albeit to a lesser extent (especially Sinner; quite good, but will of course be even tougher to beat if he can improve even more). De Minaur shockingly inaccurate on average. But see the next picture in the IG post for specific tournament averages. Both Sinner (winner) and De Minaur (runner-up) serving very well in Rotterdam.
Data on 1st serve speed and accuracy of the ATP top 20 players (Hurkacz unrivaled)
Really nice graphic, thanks for posting. When assessing serves, it's all too easy to get caught up in serve speeds and not pay enough attention to the accuracy.
The GOAT's serve pace/accuracy combo at that height 
Alcaraz seriously needs to work on his serve. Even the pace is not good there. 
Only in WTF it looked somewhat good. Dimitrov is so good. Meddy's serve is seriously lacking given his height. 
I would rather have Djokovic's serve over his any day of the week.
Is there a similar graphic for 2nd serves among the top 20? Obviously, some have notoriously erratic 2nd serves, but it would be nice to compare the averages.
I'd also be very interested to know what happens on serve when players are facing break points. I'd assume that the 'clutch' players serve better but it would be good to have confirmation.
I'd also be very interested to know what happens on serve when players are facing break points. I'd assume that the 'clutch' players serve better but it would be good to have confirmation.
Dimitrov is the biggest surprise here. I would have not thought he exceeds both Tsitsipas and Fritz in both speed and placement. If you factor all things including first serve percentage, Zverev is the second best after Hubi, closely followed by Dimitrov. Had Kyrgios not been injured, I think he would have topped Hubi in both speed and placement.
No surprise at all with Shelton. I criticized his poor serve placement before, but I would not have thought he would compare that badly with other top 20.
Also Rublev is a better server than Medvedev in both categories?! Medvedev does not have a very high first serve percentage to compensate for this, and both of them have equally bad second serves. And I still thought of Medvedev as the better server of the two. I think part of this is that his serve regressed in 2022-2023. But still, this is the one is my perception deceived me the most.
Speaking of Medvedev, when he talked about why Djokovic S & V was more effective against him compared to others like Alcaraz, he mentioned his serve placement made his serve more difficult to return. In a separate interview, he said Hubi's serve bothered him the most because of the effective combo of speed and placement. Medvedev interviews are true gems that I make a point of always listening to.
Hubi is stupid not to serve and volley with that serve, especially given his volleying skills. If Cressy can have a 90+ hold percentage by S & V, Hubi can definitely do better.
Khachanov serve speed compared with his height is really bad especially given his movement. He gets none of the benefits of that height but still moves poorly (relative to Zverev and Medvedev).
Lastly, the one element of serve that can not be charted is disguise. One example is how Djokovic had a much easier time returning the serve of Raonic compared to some other big servers, because he was able to read it better. Kyrgios also springs to mind when it comes to disguise, his toss is the same for every serve, and it is a very low toss, which makes it very hard to read. Federer of-course was also one of the best when it came to this.
No surprise at all with Shelton. I criticized his poor serve placement before, but I would not have thought he would compare that badly with other top 20.
Also Rublev is a better server than Medvedev in both categories?! Medvedev does not have a very high first serve percentage to compensate for this, and both of them have equally bad second serves. And I still thought of Medvedev as the better server of the two. I think part of this is that his serve regressed in 2022-2023. But still, this is the one is my perception deceived me the most.
Speaking of Medvedev, when he talked about why Djokovic S & V was more effective against him compared to others like Alcaraz, he mentioned his serve placement made his serve more difficult to return. In a separate interview, he said Hubi's serve bothered him the most because of the effective combo of speed and placement. Medvedev interviews are true gems that I make a point of always listening to.
Hubi is stupid not to serve and volley with that serve, especially given his volleying skills. If Cressy can have a 90+ hold percentage by S & V, Hubi can definitely do better.
Khachanov serve speed compared with his height is really bad especially given his movement. He gets none of the benefits of that height but still moves poorly (relative to Zverev and Medvedev).
Lastly, the one element of serve that can not be charted is disguise. One example is how Djokovic had a much easier time returning the serve of Raonic compared to some other big servers, because he was able to read it better. Kyrgios also springs to mind when it comes to disguise, his toss is the same for every serve, and it is a very low toss, which makes it very hard to read. Federer of-course was also one of the best when it came to this.
Another thing that surprised me is that, although most of the top 20 have faster 1st serves than the tour average, about half of them have 1st serves that are either no more accurate as or considerably less accurate than the tour average.
Medvedev's serve is just flat and fairly predictable to me. Sinner figured out how to get himself into his serves eventually and didn't look back. Plus a serious lack of heaviness. 1st% is nothing special and speed is nothing special. Accuracy is somewhat good but that's it. If his serve was close to Zverev's level, he would have won 3 slams by now probably. Seeing Rublev above him there in both categories is probably shocking but not THAT shocking imo. His serve really is somewhat average given his height. Didn't bother to improve it.
Accuracy does not seem to be strongly correlated with success. There might be several explanations for this. One might be that high accuracy often comes at the cost of a less effective disguise. If you often hit close to the lines, it might be hard to also take the trouble to make your service motion difficult to read. Also, some servers might be good to see when the returner "cheats" (taking steps left or right before the racket has hit the target), and can change direction very late. A serve executed like this can be both good and effective, without being accurate.
Really interesting. But combined with 2nd serve, the results would be completely different.
Zverev for example is day and night with 1st/2nd serve.
Zverev for example is day and night with 1st/2nd serve.
Zverev serve is overrated.as i said in other thread Zverev,Carlos,Danill have terrible placement.
Hitting huge serve in the middle of the box not doing enough demage against great returners like Novak and Jannik.
Hitting huge serve in the middle of the box not doing enough demage against great returners like Novak and Jannik.
I actually expected Shelton to be higher on the average serve speed list. I wonder where would prime Roddick would fall on that graph.
Mannarino has almost a WTA serve.
I suspected Shelton's accuracy wasn't good but I thought his avg. 1st. serve speed was faster.
I suspected Shelton's accuracy wasn't good but I thought his avg. 1st. serve speed was faster.
Wow, i knew it Grigor's serve was good, but not that good. 3rd fastest and 3rd most accurate.
While the data is interesting, I also think it is easy to overestimate it.
Serve effectiveness is not measured by speed and being close to the line alone, which is one reason some players deliberately vary their service speed, lowering the average speed. As for placement, how can we be sure we knew where the server wanted the serve to go? For example there was a goal to rule out body serves, but it is not quite so binary. A body serve is not the only way reason for a serve to be not intending to paint the lines. This tries to measure ability to painting the lines with fast serves, but it is one one tactic. What about weight of serve, sliders targeting more curve, and disguise? To some extent this graphic measure how often a player focuses on the one specific serve tactic. For example, Djokovic may keep his fastest serves for when he needs them the most, lowering the stat, but be able to serve consistently faster than a player with a faster stat if he wanted to.
In the end a stat trying to track serve effectiveness, which may reveal for some players very different results, would also be interesting.
Serve effectiveness is not measured by speed and being close to the line alone, which is one reason some players deliberately vary their service speed, lowering the average speed. As for placement, how can we be sure we knew where the server wanted the serve to go? For example there was a goal to rule out body serves, but it is not quite so binary. A body serve is not the only way reason for a serve to be not intending to paint the lines. This tries to measure ability to painting the lines with fast serves, but it is one one tactic. What about weight of serve, sliders targeting more curve, and disguise? To some extent this graphic measure how often a player focuses on the one specific serve tactic. For example, Djokovic may keep his fastest serves for when he needs them the most, lowering the stat, but be able to serve consistently faster than a player with a faster stat if he wanted to.
In the end a stat trying to track serve effectiveness, which may reveal for some players very different results, would also be interesting.
Some interesting things I paid attention to in addition to the data in OP:
Fritz improved his disguise over the last few months (after last year's Wimbledon) behind both his 1st and 2nd serves. That's one of the main reasons he beat Tsitsipas in 4 set (as even poor form Tsitsipas had a slight advantage from the baseline in the first 2 and a half sets) and the main reason he managed to save all break points he faced during the 1st set vs Djokovic (and some of the break points in the 2nd set as well).
What Medvedev lacks in speed and accuracy on his serve nowadays he partially makes up for with his IQ and delivery under pressure. Sometimes he can see to which side his opponent leans and serve properly. When he's under pressure (crucial break point for example) he can go bigger with the 2nd serve or increase his accuracy level on his 1st serve. This is exactly what he managed to pull off vs Alcaraz at the US Open.
Zverev still has issues on his 2nd serve but it looked better at this year's Australian Open. His 1st serve percentage is so high that it doesn't matter much when he's not under pressure.
Tsitsipas switching his technique looks unreasonable to me in the long term and I think it will cost him in terms of some accuracy and effectiveness of the disguise, which were excellent with his previous technique (especially on the 2nd serve).
Considering Rune's height I think his average 1st serve speed is as good as it gets for him. He needs to work a lot on improving his serve accuracy to be at least at Rublev's level in order to become a permanent Big Titles contender (along with developing an effective Plan A for his game of course).
Sinner made a massive improvement over the last 6 months when it comes to the accuracy on his serve. If he can bring it to be at least as good as Djokovic's 2013 level he will be virtually unbreakable on grass and hard courts against players without an elite return game (at least 88 of the top 100).
Fritz improved his disguise over the last few months (after last year's Wimbledon) behind both his 1st and 2nd serves. That's one of the main reasons he beat Tsitsipas in 4 set (as even poor form Tsitsipas had a slight advantage from the baseline in the first 2 and a half sets) and the main reason he managed to save all break points he faced during the 1st set vs Djokovic (and some of the break points in the 2nd set as well).
What Medvedev lacks in speed and accuracy on his serve nowadays he partially makes up for with his IQ and delivery under pressure. Sometimes he can see to which side his opponent leans and serve properly. When he's under pressure (crucial break point for example) he can go bigger with the 2nd serve or increase his accuracy level on his 1st serve. This is exactly what he managed to pull off vs Alcaraz at the US Open.
Zverev still has issues on his 2nd serve but it looked better at this year's Australian Open. His 1st serve percentage is so high that it doesn't matter much when he's not under pressure.
Tsitsipas switching his technique looks unreasonable to me in the long term and I think it will cost him in terms of some accuracy and effectiveness of the disguise, which were excellent with his previous technique (especially on the 2nd serve).
Considering Rune's height I think his average 1st serve speed is as good as it gets for him. He needs to work a lot on improving his serve accuracy to be at least at Rublev's level in order to become a permanent Big Titles contender (along with developing an effective Plan A for his game of course).
Sinner made a massive improvement over the last 6 months when it comes to the accuracy on his serve. If he can bring it to be at least as good as Djokovic's 2013 level he will be virtually unbreakable on grass and hard courts against players without an elite return game (at least 88 of the top 100).
From ATP's stats, sorted based on %1st serve points won:
2nd serve points won:
1st serve percentage:
Overall:
Hurkacz:
Eubanks:
Zverev:
Tsitsipas:
Djokovic:
Sinner:
Alcaraz:
--
Indeed different ways how players get ranked highly based on their serving.
Hurkacz (overall #1) serves fast balls close to the lines. This means his 1st serve % is only #25 but it works when the rockets go in. Not top tier behind his 2nd serve (#28).
Zverev (overall #3) also serves fast balls but with more margin. His 1st serve % is #2 overall.
Tsitsipas (overall #4) seems similar to Hurkacz. Tries to go closer to the lines, which reduces the 1st serve % but is nonetheless successful when he finds the spots. Also decent behind his 2nd serve (#7; Hurkacz #28).
Djokovic (overall #5) doesn't have the size of the guys before him. Less pace, but he goes close to the lines. This impacts his 1st serve %. 2nd serve is where Djokovic truly shines, being ranked #1 in % of 2nd serve points won.
Sinner (overall #7) is interesting in that he is listed at same height as Djokovic. He goes for a little more than Djokovic (roughly 2 mph more) but is around 3 to 5 mph behind the taller guys. 1st serve percentage is subpar (#61), but overall this works for him fairly well. #6 in % 1st serve points won. Likewise solid behind his 2nd serve at #4.
Alcaraz (overall #9) puts a good amount of his serves in (#13), but this comes at a cost of both speed and margins in that his serves land rather far from the edges of the box. Currently he is only #32 in % of 1st serve points won. The 2nd serve is where he shines, being ranked #2 behind Djokovic.
--
What would further be interesting to see is how much action/movement/spin the players put on the ball. Sth that you can also use to level up your serve.
--
Also, check this one out. From Rotterdam where Sinner was serving much more closer to the edges of the box than on average. See the 1st picture (serve and return quality) and then the 2nd one (forehand and backhand quality):
Serving well is important, but returning well is even more important? ...when you have solid groundstrokes. Just need to be able to neutralize the serve to then do damage.
- Overall #1
- % 1st serve points won #1
- % 2nd serve points won #28
- 1st serve % #25
- 1st serve speed #1 (among the top 20; from the Tennis Insights graphic)
- 1st serve accuracy #1 (same as above)
Eubanks:
- Overall #2
- % 1st serve points won #21
- % 2nd serve points won #29
- 1st serve % #3
- 1st serve speed N/A (not inside the top 20)
- 1st serve accuracy N/A (not inside the top 20)
Zverev:
- Overall #3
- % 1st serve points won #24
- % 2nd serve points won #34
- 1st serve % #2
- 1st serve speed #2
- 1st serve accuracy #12
Tsitsipas:
- Overall #4
- % 1st serve points won #2
- % 2nd serve points won #7
- 1st serve % #33
- 1st serve speed #4
- 1st serve accuracy #6
Djokovic:
- Overall #5
- % 1st serve points won #15
- % 2nd serve points won #1
- 1st serve % #23
- 1st serve speed #13
- 1st serve accuracy #2
Sinner:
- Overall #7
- % 1st serve points won #6
- % 2nd serve points won #4
- 1st serve % #61
- 1st serve speed #9
- 1st serve accuracy #10
Alcaraz:
- Overall #9
- % 1st serve points won #32
- % 2nd serve points won #2
- 1st serve % #13
- 1st serve speed #17
- 1st serve accuracy #18
--
Indeed different ways how players get ranked highly based on their serving.
Hurkacz (overall #1) serves fast balls close to the lines. This means his 1st serve % is only #25 but it works when the rockets go in. Not top tier behind his 2nd serve (#28).
Zverev (overall #3) also serves fast balls but with more margin. His 1st serve % is #2 overall.
Tsitsipas (overall #4) seems similar to Hurkacz. Tries to go closer to the lines, which reduces the 1st serve % but is nonetheless successful when he finds the spots. Also decent behind his 2nd serve (#7; Hurkacz #28).
Djokovic (overall #5) doesn't have the size of the guys before him. Less pace, but he goes close to the lines. This impacts his 1st serve %. 2nd serve is where Djokovic truly shines, being ranked #1 in % of 2nd serve points won.
Sinner (overall #7) is interesting in that he is listed at same height as Djokovic. He goes for a little more than Djokovic (roughly 2 mph more) but is around 3 to 5 mph behind the taller guys. 1st serve percentage is subpar (#61), but overall this works for him fairly well. #6 in % 1st serve points won. Likewise solid behind his 2nd serve at #4.
Alcaraz (overall #9) puts a good amount of his serves in (#13), but this comes at a cost of both speed and margins in that his serves land rather far from the edges of the box. Currently he is only #32 in % of 1st serve points won. The 2nd serve is where he shines, being ranked #2 behind Djokovic.
--
What would further be interesting to see is how much action/movement/spin the players put on the ball. Sth that you can also use to level up your serve.
--
Also, check this one out. From Rotterdam where Sinner was serving much more closer to the edges of the box than on average. See the 1st picture (serve and return quality) and then the 2nd one (forehand and backhand quality):
Serving well is important, but returning well is even more important? ...when you have solid groundstrokes. Just need to be able to neutralize the serve to then do damage.
The all-time overall serve rating leader list on the ATP site is kinda weird, to say the least. Nadal above Djokovic on this list? How?!
Only players shorter than 190cm on that list are Roddick (makes sense), Federer (makes sense as well) and that random Aussie who is apparently 185cm yet serves 15.8 aces per match!
Nadal above Muller, Ljubicic and Querrey 
You got to laugh about it. I don't think this can be taken that seriously.
You got to laugh about it. I don't think this can be taken that seriously.
Maybe he mixes in some kick serves for the first serve variety? I’m not sure, I need to watch for that in his matches more closely. It is surprising.
Bit late with responding, but great thread! In case of the extreme cases here (Zverev, Djokovic, Hurkacz), this graphic is of those occasions where the stats totally meet the eye test.
Zverev in my mind was always the hardest server on tour and combines that with an insane FS%. One could also say his accuracy is "bad", but I consider those statements equivalent. Going closer to the lines will inevitable make your FS% decline, it is rather a tactical choice than an actual weakness (a choice that makes sense given his lack of a great second serve).
Djokovic on the other hand always stroke me as one of the most accurate servers, where accuracy was the main asset of his serving strength given his rather average serving speed (even more so Federer ofc., insanely precise server).
Then Hurcacz just has everything, speed and accuracy, a true servebot.
Zverev in my mind was always the hardest server on tour and combines that with an insane FS%. One could also say his accuracy is "bad", but I consider those statements equivalent. Going closer to the lines will inevitable make your FS% decline, it is rather a tactical choice than an actual weakness (a choice that makes sense given his lack of a great second serve).
Djokovic on the other hand always stroke me as one of the most accurate servers, where accuracy was the main asset of his serving strength given his rather average serving speed (even more so Federer ofc., insanely precise server).
Then Hurcacz just has everything, speed and accuracy, a true servebot.
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
