What would the top 20 look like if players only had only one attempt on each serve? Here are some estimates using second serve/return statistics.

[DSH]

I vote for a third serve.

 

[Chanwan]

Great work and I like your method of assuming players would serve similar to their second serve if they only had one serve. I think players would probably train themselves to hit serves around 100ish mph with heavy kick and topspin would become the weapon of choice on the serve, resulting in many more rallies, so perhaps factor rally lengths into your methodology. On a tour where long rallies are the key to winning matches, you better expect Novak to be at the top. It would he interesting to see how Federer does via this metric in one of the years he competed at a high level while all the big 2 and Murray also were at a high level. Would he do better than Murray?

Murray's 3 biggest weaknesses were his 2nd serve, his first serve percentage and his forehand. While Fedalovic dominated the points won behind their 2nd serve, Murray often could be found around no. 30-50 in that category. (he's 63 here while Rafa and Fed top the list): https://www.atptour.com/en/stats/2nd-serve-points-won/all/all/all/


Federer, however, did not do enough damage on the 2nd serve return. He was often too passive, contend to just get into the rally. That was a weakness of his, for sure. He's 110 while Murray is 11th

Second Serve Return Points Won | ATP Tour | Tennis

Second serve return points won leaders and stats by career, year, surface and country from men's professional tennis on the ATP Tour.

www.atptour.com

If we take 2012 - a fairly good/great year for both - Andy's top of the pack in 2nd serve points won, while Fed is 14th:

Second Serve Return Points Won | ATP Tour | Tennis

Second serve return points won leaders and stats by career, year, surface and country from men's professional tennis on the ATP Tour.

www.atptour.com

Behind their 2nd serve, Fed's winning an outrageous 60 %, but Andy did well this year too, being no. 11th overall.

Second Serve Points Won | ATP Tour | Tennis

Second serve points won leaders and stats by career, year, surface and country from men's professional tennis on the ATP Tour.

www.atptour.com

In short, I'd give the edge to Fed for a variety of reasons (including him tending to win the big ones against Muzz), but it sure would be close.

 

[Chanwan]

@AlecG - isn't it easier to just look at these 2? Or do you intend to judge it based on the last 12 months exclusively, rather than say 2022?

Second Serve Return Points Won | ATP Tour | Tennis

Second serve return points won leaders and stats by career, year, surface and country from men's professional tennis on the ATP Tour.

www.atptour.com

Second Serve Points Won | ATP Tour | Tennis

Second serve points won leaders and stats by career, year, surface and country from men's professional tennis on the ATP Tour.

www.atptour.com

PCB - very surprisingly to me - is leading the 2nd serves won in 2022. Novak is 2nd and 1st in 2nd returns won

 

[Chanwan]

That's like saying two serves greatly benefits those with crap groundstrokes, crap returns and crap movement. It'd reduce the disadvantage of being average height or shorter. & yeah I love those players. I've seen both Nishoka & Cressy live, and although I like Cressy, Nishioka is roughly 10x more fun to watch. Nishioka's rallying skills are insane & it's mostly his height & first serve return holding him back.

Every pro on tour has a good serve for their height. The differences are almost entirely due to height. Isner & Opelka win more points on first serve than Kyrgios because they're taller. It's free points for being tall. There's at least one exception (Jenson Brooksby) who is tall & has a weaker serve for his height, but every top 100 player shorter than 6'2" has a good serve for their height or they wouldn't survive. But they don't get many free points just because they aren't extremely tall. Technique & muscle mass index are also factors in number of aces per match but they are both much smaller factors than height among pros. So in practice two serves at the pro level is free points for being tall. Yet I'm not even saying to get rid of it all together! I'd just like to see at least one 250 where they allow only one attempt on serve, as would most tennis fans, if only for the novelty at first.

Come on now, you don't really believe that? Look at Roddick. He's roughly the same height as Rafa (officially an inch difference). Look at Kyrgios, Ivanisevic and Raonic - all ATG servers between 1,93-1,96 (6'4-6'5). Delpo, Berdych, Medvedev and Zverev - all 6'5-6'6 - do not even enter into the same conversation.

Federer is 6'1 and has a vastly better serve than Rafa and Murray (who's taller). Djokovic' added so much to his serve over his career and is now an elite server.
Pete was also 6'1... Wilander was 2 cm shorter than Sampras and his serve was sh****t compared to Pete's.

Connors was 2 cm shorter than McEnroe and his serve was also far, far worse.

Is the difference really almost entirely due to height, as you claim? Of course it isn't. Just like some have a better forehand, backhand, volley etc., some have a better serve based on better technique. But yes, there is a strong correlation with being tall and having a good serve. But there's plenty of difference between players the same height.

Part of the beauty of tennis to me is that you can succeed with very different bodies. Sure, it's tough when you're below 6'0 or 5'11, but it's still doable with the right amount of talent. And then you see Diego vs. Isner - where else do you get that contrast?

Wouldn't mind to see it once or twice a year, but I like that the serve means something.


"Isner & Opelka win more points on first serve than Kyrgios because they're taller."
Yes they do. But keep in mind the difference Kyrgios and Isner (15 cm) is similar to the difference between Kyrgios and Fognini/Nishikori.
Kyrgios' and Isner's serve are in the same league. Fognigi and Nishikori... not so much

 

[Chanwan]

Fair question, I wasn’t too clear. What I mean is that you’d likely see less variance and more top-dominance if the second serve (or, functionally the first serve really) is neutered because even average serving generally allows for more randomness in outcomes and more upset potential. Even a player with a mediocre service game should hold 4 out of every 5 times on a medium-to-quick court. That’s at least somewhat of a bulwark against the “never-in-doubt” results we’d be likelier to see otherwise. As an example, picture prime Nadal in this scenario. Take the serve-return complex out of the equation and I genuinely don’t see him partaking in more than a handful of competitive matches on clay from, let’s say, 2005-2013 lol.

I think by de-prioritizing offensive serving you’d (ironically) make tennis more of a specialists sport by removing one of its key components. Less variety also just intuitively seems like a recipe for less parity. Another flawed analogy: boxing is known to be disproportionately top-heavy, whereas in MMA there’s frequent turnover at the top. I believe that’s because there are more ways to lose and more (totally disparate) forms of combat to master. If you strip MMA of grappling/kicks/clinching and turn it into boxing with 4oz gloves, you can safely bet that it becomes about as top-heavy as the sweet science within a decade. Seeing as tennis is already a very top-heavy sport, even more of that scares me.

Also, I grant you that there’s basically a straight-line correlation between height and serving prowess, but I never felt that to be particularly unfair to anyone. What taller players gain on the serve they often lose on the movement side of things. The elite movers tend to be 6’3 and below. You can see how height-related poor movement and suboptimal centre of gravity affects someone like (for instance) Isner, whose stand-still groundstrokes really aren’t that bad but he can’t hit a running forehand to save his life.

Everyone’s mileage varies but I personally think optimal height range in tennis is 6’0-6’4. Anything shorter or taller and you start getting those aforementioned serve-for-movement trade-offs. There’s a reason there have only been a few all-time great servers shorter than 6’0 (Mac, at 5’11…who else? Tanner was 6’ flat), and likewise very few all-time great movers taller than 6’4 (drawing blanks here to be honest; are there any? Admittedly the pool is much smaller).

Excellent post. Connors was 5'10, Agassi 5'11. And in the olden days, Laver and Rosewall were both (officially at least) around Diego's height (173 and 170)

 

[Chanwan]

Oh I don’t doubt this, after all I do maintain 6’0-6’4 is the optimal height range, so tennis today definitely skews taller no matter what. I just think you start to see diminishing returns after 6’4ish. But you’re correct that this could be due to the smaller share of <6’4 men amongst the general population.

While your proposed ruleset would indeed make tennis more egalitarian in that regard, it does bear mentioning that there’d still be an uneven distribution because >5’10’ers would still likely be at a disadvantage, albeit less of one…optimal height helps with hitting penetrating groundstrokes and overheads, as well as increasing returning reach.

All in all, it’s a tough one to get behind because the product would suffer too much for my liking. I think banning poly may possibly achieve many of the desired effects you’re aiming for here, but without turning the game on its head.

There's also another factor at play. The most athletically gifted 2+ meter guys tend to chose other sports (basketball).
In theory, we could have a 2 meter guy, who could move great (Medvedev and Zverev comes pretty darn close already).

 

[Chanwan]

One comment that I think you’re getting at but don’t quite state: as well as neutralizing first serve, having only one serve makes return of serve a much less important skill.

Yes. Nishioka, @AlecG's ideal no. 1 (or 2), is very poor in first serve return. Also compared to other short players. Should we reward that? If so, why?

 

[Chanwan]

Probably so, though I’d expect young Djokovic to retain his overall advantage on HC, where he’s still better in neutral rally situations than Nadal IMO. But the clay court season would just be a total formality, even more so than it is now. I shudder to think about how dominant Nadal would be on clay if his ordinary serve + ordinary pure return (relatively speaking, of course) were no longer exploitable. Would probably bag an extra 3ish Wimby’s too.

2011 stats here, Djokovic outperforming Rafa on 2nd serve serve points: https://www.atptour.com/en/stats/2nd-serve-points-won/2011/clay/all/, while Rafa is ahead in the return department. Difference is small enough to make for close matches though https://www.atptour.com/en/stats/2nd-serve-return-points-won/2011/clay/all/
@helterskelter

 

[socallefty]

I feel like a sport is more world class when at the highest level, it pushes the limits of the human body to excel. In American football and basketball, every decade demands pro players who are taller, faster and stronger and as a result, the level of play keeps going up. To me, it is a sign that pro tennis is a world class sport for top athletes extending outside of country clubs when only taller, stronger players can excel at the highest level. The sport is faster with more powerful shots/serves and quicker movement which is what makes it more entertaining for me especially when you watch in person where the difference in athleticism from previous decades is obvious. It is a sport now of world class athletes with the best technique rather than average athletes with world class technique which is what it used to be.

Any rule change that is done to prevent this natural trend of taller, stronger players is a move in the wrong direction in my opinion. It is becoming rare to see players below 6 feet in the NFL and NBA - I like that tennis is moving that way also.

Begone out of my sight and life, Nishioka, Diego and AlecG! Tall people have rights too and we should not be rigging sport rules against them - let’s play tennis like it has always been played and let the best athletes win.

 

[Chanwan]

So do I, but it all depends how the rule change would affect his mental game and the mental game of his opponents. He had a great run from September up until the loss to Korda in Adelaide, but he doesn't do that well at big tournaments and plays a lot of 250s, so the stats are distorted by that, and he's no longer at no. 1 on these stats regardless. He is often very negative on the court & I think that's gotten worse since the Korda match.

@BauerAlmeida - Nishioka would be nowhere close to no. 1. Simply because the stats are not just distorted, they're so far off it's not even funny.

Here he is against top-20: https://www.ultimatetennisstatistics.com/playerProfile?playerId=6401&tab=statistics
(50,4 % won behind his own 2nd serve, 49,5 % against opponents' 2nd serve). I.e. essentially a 50-50 match vs. top-20 players. That's actually good (and his results vs. top-10 are similar). But to say it's enough to make him close to no. 1 is outright insanity.

Djokovic is 54,1 & 53,6 % vs. top-20. Rafa is 54,9 & 53,2 %. Federer is 54,7 & 49,3 % (as mentioned above, his 2nd serve return was very much his weakness).
I.e. leagues above Nishioka. Medvedev is 52,1 & 53,9 %. Also miles better.

 

[Chanwan]

@AlecG - in short, who's in the the top is not just decided by the crude stats vs. the field, but the stats against the very top.

When up against top-20, Nishioka wins roughly 50 % of the points if you limit it to only looking at 2nd serve points won and 2nd return points won.
When up against top-5, his stats drop to 49,6 & 45,7 %. Still quite good, but needless to say he would lose the majority of matches with these stats.

Funny side note: Federer is actually vastly better at breaking serve on HC against top-10 on hard than Rafa (who among some die hard vamosbrigaders are considered the best returner of all time)
23,7 % for Fed, 19,8 % for Rafa (and 26,9 % for Novak). Novak's advantage over Fed is entirely due to 2nd serve return points won. Fed's actually ahead in first serve if ever so slightly (31,1 to 31,0, murygoat is also at 31,1, while Rafa is at 27,2 %).

Change that to clay and Rafa takes a MASSIVE leap to 37,2 %, Novak stays more or less the same (26,1 %), while Roger goes down to 22,3 %

 

[AlecG]

@AlecG - isn't it easier to just look at these 2? Or do you intend to judge it based on the last 12 months exclusively, rather than say 2022?

Second Serve Return Points Won | ATP Tour | Tennis

Second serve return points won leaders and stats by career, year, surface and country from men's professional tennis on the ATP Tour.

www.atptour.com

Second Serve Points Won | ATP Tour | Tennis

Second serve points won leaders and stats by career, year, surface and country from men's professional tennis on the ATP Tour.

www.atptour.com

PCB - very surprisingly to me - is leading the 2nd serves won in 2022. Novak is 2nd and 1st in 2nd returns won

It wouldn't be any easier or any harder, was just interested in who would be top 10 right now under that methodology. Maybe at some point I will do 2022 or even go back in time to work out all the year end winners during big 3 era.

 

[AlecG]

Come on now, you don't really believe that? Look at Roddick. He's roughly the same height as Rafa (officially an inch difference). Look at Kyrgios, Ivanisevic and Raonic - all ATG servers between 1,93-1,96 (6'4-6'5). Delpo, Berdych, Medvedev and Zverev - all 6'5-6'6 - do not even enter into the same conversation.

Federer is 6'1 and has a vastly better serve than Rafa and Murray (who's taller). Djokovic' added so much to his serve over his career and is now an elite server.
Pete was also 6'1... Wilander was 2 cm shorter than Sampras and his serve was sh****t compared to Pete's.

Connors was 2 cm shorter than McEnroe and his serve was also far, far worse.

Is the difference really almost entirely due to height, as you claim? Of course it isn't. Just like some have a better forehand, backhand, volley etc., some have a better serve based on better technique. But yes, there is a strong correlation with being tall and having a good serve. But there's plenty of difference between players the same height.

Part of the beauty of tennis to me is that you can succeed with very different bodies. Sure, it's tough when you're below 6'0 or 5'11, but it's still doable with the right amount of talent. And then you see Diego vs. Isner - where else do you get that contrast?

Wouldn't mind to see it once or twice a year, but I like that the serve means something.


"Isner & Opelka win more points on first serve than Kyrgios because they're taller."
Yes they do. But keep in mind the difference Kyrgios and Isner (15 cm) is similar to the difference between Kyrgios and Fognini/Nishikori.
Kyrgios' and Isner's serve are in the same league. Fognigi and Nishikori... not so much

The stats show that you're exaggerating the differences. Yes, of course Kyrgios is a better server than Zverev, but the stats are very clear that the difference between Zverev and Rafa is huge compared to the difference between Kyrgios and Zverev. So yes I do believe it's *almost* entirely (not entirely) determined by height at the pro level. The stats are very clear on this. Just have a look at them.

 

[AlecG]

Yes. Nishioka, @AlecG's ideal no. 1 (or 2), is very poor in first serve return. Also compared to other short players. Should we reward that? If so, why?

Not my ideal, just happened to be no 1 in Feb on this methodology as he had a very good year up until the loss to Korda in January.
It's not rewarding a poor first serve return to have only one attempt on serve. It would just become irrelevant. The question is should we give so many free points for being tall in every single tournament & thereby vastly increase the number of faults and boring short points & greatly disadvantage men of normal height or below. If so, why?

 

[AlecG]

Tall people have rights too and we should not be rigging sport rules against them - let’s play tennis like it has always been played and let the best athletes win.

You're literally supporting rules that rig tennis in favour of taller players & yet a guy barely above average height like Alcaraz still dominates most of them.

Would you also support similar rules to modify soccer so that a guy like Messi can't become a pro because it hurts tall people feelings when short people are good at sports?

In soccer, there are amazing players that are short and also amazing players who are tall in similar numbers. All else being equal, that makes it a much richer sport than if there were rules that disadvatanged short players without making the sport more entertaining overall.

 

[AlecG]

@BauerAlmeida - Nishioka would be nowhere close to no. 1. Simply because the stats are not just distorted, they're so far off it's not even funny.

Here he is against top-20: https://www.ultimatetennisstatistics.com/playerProfile?playerId=6401&tab=statistics
(50,4 % won behind his own 2nd serve, 49,5 % against opponents' 2nd serve). I.e. essentially a 50-50 match vs. top-20 players. That's actually good (and his results vs. top-10 are similar). But to say it's enough to make him close to no. 1 is outright insanity.

Djokovic is 54,1 & 53,6 % vs. top-20. Rafa is 54,9 & 53,2 %. Federer is 54,7 & 49,3 % (as mentioned above, his 2nd serve return was very much his weakness).
I.e. leagues above Nishioka. Medvedev is 52,1 & 53,9 %. Also miles better.

You make some good points but fail to recognise that I didn't say Nishioka would be number 1. I reported that he was number 1 in that month under that methodology and noted the methodology is flawed.

The confidence of shorter players against taller players would be greatly improved under such rules, so it's really hard to say where players like Nishioka would be ranked.

 

[mahesh69a]

What's next? Basketball for midgets? (or should I call them vertically challenged persons?)

 

[AlecG]

What's next? Basketball for midgets? (or should I call them vertically challenged persons?)

Arguably that's called soccer/football and it's much more popular, fair, talent-based and skill-based than basketball.

But actually, why not a 6 foot and under basketball league (in addition to the current league)? That would be amazing & you'd get better talent due to the much larger talent pool.

 

[mahesh69a]

But actually, why not a 6 foot and under basketball league (in addition to the current league)? That would be amazing & you'd get better talent due to the much larger talent pool.

What about Chess for people with an IQ less than 120 in addition to the current FIDE competitions? Would that be amazing as well? That is surely a much larger 'talent' pool than the current crop.

 

[AlecG]

What about Chess for people with an IQ less than 120 in addition to the current FIDE competitions? Would that be amazing as well? That is surely a much larger 'talent' pool than the current crop.

No, because that would be excluding talent and height leagues don't exclude talent, any more than weight classes exclude talent in UFC or weight lifting.

Also, every single weight class in UFC is far more popular than chess.

Meanwhile, two attempts on serve at *every* tournament, *does* exclude talent, in practice.

 

[AlecG]

I just had a look and the number ones from 1998-2022 are:

Rios 1998
Agassi 1999-2000
Ferrero 2001
Agassi 2002
Ferrero 2003
Agassi 2004
Nadal 2005
Federer 2006-2007
Nadal 2008-2014
Novak 2015-2016
Nadal 2017-2019
Ruud 2021
Novak 2022

I've added this to the OP.

 

[Chanwan]

The stats show that you're exaggerating the differences. Yes, of course Kyrgios is a better server than Zverev, but the stats are very clear that the difference between Zverev and Rafa is huge compared to the difference between Kyrgios and Zverev. So yes I do believe it's *almost* entirely (not entirely) determined by height at the pro level. The stats are very clear on this. Just have a look at them.

I have looked at the stats and gave you plenty of examples, where your statement is outright wrong.
There is a correlation, even a strong won, but still plenty of wiggle room. No one would pick 6'6 Zverev or Medvedev over 6'1 Pete in terms of serve

 

[Chanwan]

Not my ideal, just happened to be no 1 in Feb on this methodology as he had a very good year up until the loss to Korda in January.
It's not rewarding a poor first serve return to have only one attempt on serve. It would just become irrelevant. The question is should we give so many free points for being tall in every single tournament & thereby vastly increase the number of faults and boring short points & greatly disadvantage men of normal height or below. If so, why?

Because... Tennis? You are the one proposing a change that would alter a very, very great sport completely

And yes, it is rewarding Nishioka if we take away the two parts of the game, where he does poorly

 

[Chanwan]

You make some good points but fail to recognise that I didn't say Nishioka would be number 1. I reported that he was number 1 in that month under that methodology and noted the methodology is flawed.

The confidence of shorter players against taller players would be greatly improved under such rules, so it's really hard to say where players like Nishioka would be ranked.

You also asked us to comment on the biggest flaws in your methodology. And this is it, it's too crude.

I do think your overall idea is good - we can roughly use their points won on 2nd serve and 2nd serve return to get an idea of what might happen.

But - and this is important - we need to supplement those stats with how they do in those categories against quality opponents.

Nishioka is a quality rally player, perhaps even borderline top-10. But he's not top-3 or top-1, which becomes clear when factoring in opposition.

And yes, confidence can play a positive role, but it cannot make Nishioka better than Djokovic or Alcaraz under your proposed system

 

[Chanwan]

I just had a look and the number ones from 1998-2022 are:

Rios 1998
Agassi 1999-2000
Ferrero 2001
Agassi 2002
Ferrero 2003
Agassi 2004
Nadal 2005
Federer 2006-2007
Nadal 2008-2014
Novak 2015-2016
Nadal 2017-2019
Ruud 2021
Novak 2022

I've added this to the OP.

I am well aware that it's complicated but I think Rafa gets overly rewarded by his clay dominance here.
On clay, even under the current rules, Rafa routinely wins 65 % of the games, that's almost impossible elsewhere.

But does he win more or as much on HC and grass as Novak and Fed under your system? I doubt it with Novak (but haven't checked the numbers)

If I am right it would mean Novak would still win more points than Rafa in 70 % of the season while being the 2nd best on clay = Novak would probably be no 1 in quite a few of the Rafa years above.

Anyway that would require a deeper dive in the stats.

 

[AlecG]

I have looked at the stats and gave you plenty of examples, where your statement is outright wrong.
There is a correlation, even a strong won, but still plenty of wiggle room. No one would pick 6'6 Zverev or Medvedev over 6'1 Pete in terms of serve

You're not interpreting the stats very well. Cherry picking outliers doesn't disprove the claim I'm making because I said "almost" entirely. You have to look at average differences between tallest & shortest servers & compare that to the differences between the best & worst servers at the same height.
Zverev has perhaps the smallest number of free points at that height or above, but he still gets more than Mannarino who is the *best* server under 6 foot, yet comparing Zverev and Mannarino doesn't indicate the overall trend. Kyrgios only gets twice as many aces as Mannarino (who is the best server under 6 foot), while Zverev gets seven times as many aces as Diego & four times as many as Baez & Nishioka. This is a much larger difference than the difference between Kyrgios & Zverev, or between Nishioka and Diego. All of the top 10 servers right now are above 190 cm, and most don't have exceptional technique, just the standard technique that most top 100 players have plus height. The tiny number of players over 190 cm who aren't in the top 10 for aces are in the top 30 for aces. Every player below 5'9 is in the bottom 6 (of the top 79 players) for aces.

1st serve points won:

Over 190 cm:

Isner 80%
Berretini 80%
Kyrgios 80%
Cressy 80%
Hurkacz 79%
Shelton 79%
Bublik 78%
Felix 78%
Cilic 78%
Fritz 77%
Otte 77%
Dimitrov 77%
Tsitsipas 77%
Huesler 76%
Medvedev 76%
Sonego 74%
Khachanov 73%
Zverev 73%

Median 77.5%

The differences here a small apart from three outliers at the bottom despite the huge difference in ground stroke & rallying ability & both players above 2 metres tall are in the top 4 despite some of the weaker rallying skills on tour, so even the within-group variation is largely explained by height with the exception of the three outliers at the bottom.

Now let's look at shorter players:

175-179 cm

Evans 71%
McDonald 69%
Molcan 67%

Median 69%

<174 cm
Nishioka 65%
Baez 65%
Diego 62%

Median 65%

It's pretty clear that the differences relating height are *much* bigger than the differences unrelated to height, despite the fact that the shorter players are much better at moving and generally better at rallying, without which their percentages here would be even lower.

The difference between the tallest server and the shortest server (which includes outliers) is 80-62% = 18%

The difference between >190cm and <174cm median averages (which excludes outliers) is 77.5-65% = 12.5%

The difference between the best and worst servers in each similar height group (which includes outliers) is 3-7%

Excluding outliers, the difference between the best and below average servers in each group is 2-4%

CONCLUSION

If we include outliers that are poor servers for their height, the difference in first serve points won is 18% for height-related factors & 3-7% for factors not relating to height.

If we exclude outliers that are poor servers for their height, the difference in first serve points won is 12.5% for height-related factors and less than 2-4% for non-height related factors.

On top of this, the height-related differences would be even greater if it wren't for the fact that shorter players are generally better at moving and rallying to back up their serves.

So "differences in free points on serve at the top 100 pro level are almost entirely due to height" is not "outright wrong" as you claim. It's basically accurate, although I'm happy to rephrase is as "primarily due to height".
Of course, it doesn't really matter, since even if the difference were only 50% due to height, it'd still be the cause of average and below average height players being so rare on the tour, and the cause of the boredom of watching faults and preparation for second serves, etc. It'd still mean that half of those free points are just for height, on top of points for better serve technique & points for better physiology for preventing big-serve-related injury.

 

[AlecG]

Because... Tennis? You are the one proposing a change that would alter a very, very great sport completely

And yes, it is rewarding Nishioka if we take away the two parts of the game, where he does poorly

So, conservative bias. That's not a rational reason to oppose something.

& no, it's just no longer punishing him for not being tall or not rewarding his opponents for being tall. That's not what the word reward means. A reward is when you give someone something to incentivise a specific behaviour or trait, not when you just remove a punishment for lacking a trait or remove a reward for an opponent. But there's no point going back and forth on that if you can't see the difference.

 

[AlecG]

You also asked us to comment on the biggest flaws in your methodology. And this is it, it's too crude.

I do think your overall idea is good - we can roughly use their points won on 2nd serve and 2nd serve return to get an idea of what might happen.

But - and this is important - we need to supplement those stats with how they do in those categories against quality opponents.

Nishioka is a quality rally player, perhaps even borderline top-10. But he's not top-3 or top-1, which becomes clear when factoring in opposition.

And yes, confidence can play a positive role, but it cannot make Nishioka better than Djokovic or Alcaraz under your proposed system

I asked you to comment, yes. Not call a claim "insane" & suggest that I made that claim when I didn't, which is irrationally aggressive behaviour.

Otherwise, I agree with the constructive & accurate parts of what you said.

 

[AlecG]

I am well aware that it's complicated but I think Rafa gets overly rewarded by his clay dominance here.
On clay, even under the current rules, Rafa routinely wins 65 % of the games, that's almost impossible elsewhere.

But does he win more or as much on HC and grass as Novak and Fed under your system? I doubt it with Novak (but haven't checked the numbers)

If I am right it would mean Novak would still win more points than Rafa in 70 % of the season while being the 2nd best on clay = Novak would probably be no 1 in quite a few of the Rafa years above.

Anyway that would require a deeper dive in the stats.

OK, but it's an average across all surfaces, so the fact hard court & grass are >60% of the season is working against Rafa & he still comes out on top. If you're gonna arbitrarily weight hard court & grass more heavily, sure, Novak might have won an additional year or two. Some of the years that Rafa won were close between Rafa & Novak. However, the head to head between Rafa & Novak is almost equal & the number of slams is currently equal, and Novak is taller & has a better first serve, and his first serve return I guess is roughly as good or better, so you've got to expect that Nadal would have the edge without the much greater number of free points on serve for Novak.

 

[Chanwan]

So, conservative bias. That's not a rational reason to oppose something.

& no, it's just no longer punishing him for not being tall or not rewarding his opponents for being tall. That's not what the word reward means. A reward is when you give someone something to incentivise a specific behaviour or trait, not when you just remove a punishment for lacking a trait or remove a reward for an opponent. But there's no point going back and forth on that if you can't see the difference.

I know what I have and I like it. You can call that conservative bias if you please - reward/not punishing, you are probably right, I am not a native speaker. Let me refrase then: I think being able to return a massive serve is an integral part of tennis and Nishioka's inability on this metric is something I think he should get punished for

 

[Chanwan]

I asked you to comment, yes. Not call a claim "insane" & suggest that I made that claim when I didn't, which is irrationally aggressive behaviour.

Otherwise, I agree with the constructive & accurate parts of what you said.

Fair point, sorry

 

[Chanwan]

You're not interpreting the stats very well. Cherry picking outliers doesn't disprove the claim I'm making because I said "almost" entirely. You have to look at average differences between tallest & shortest servers & compare that to the differences between the best & worst servers at the same height.
Zverev has perhaps the smallest number of free points at that height or above, but he still gets more than Mannarino who is the *best* server under 6 foot, yet comparing Zverev and Mannarino doesn't indicate the overall trend. Kyrgios only gets twice as many aces as Mannarino (who is the best server under 6 foot), while Zverev gets seven times as many aces as Diego & four times as many as Baez & Nishioka. This is a much larger difference than the difference between Kyrgios & Zverev, or between Nishioka and Diego. All of the top 10 servers right now are above 190 cm, and most don't have exceptional technique, just the standard technique that most top 100 players have plus height. The tiny number of players over 190 cm who aren't in the top 10 for aces are in the top 30 for aces. Every player below 5'9 is in the bottom 6 (of the top 79 players) for aces.

1st serve points won:

Over 190 cm:

Isner 80%
Berretini 80%
Kyrgios 80%
Cressy 80%
Hurkacz 79%
Shelton 79%
Bublik 78%
Felix 78%
Cilic 78%
Fritz 77%
Otte 77%
Dimitrov 77%
Tsitsipas 77%
Huesler 76%
Medvedev 76%
Sonego 74%
Khachanov 73%
Zverev 73%

Median 77.5%

The differences here a small apart from three outliers at the bottom despite the huge difference in ground stroke & rallying ability & both players above 2 metres tall are in the top 4 despite some of the weaker rallying skills on tour, so even the within-group variation is largely explained by height with the exception of the three outliers at the bottom.

Now let's look at shorter players:

175-179 cm

Evans 71%
McDonald 69%
Molcan 67%

Median 69%

<174 cm
Nishioka 65%
Baez 65%
Diego 62%

Median 65%

It's pretty clear that the differences relating height are *much* bigger than the differences unrelated to height, despite the fact that the shorter players are much better at moving and generally better at rallying, without which their percentages here would be even lower.

The difference between the tallest server and the shortest server (which includes outliers) is 80-62% = 18%

The difference between >190cm and <174cm median averages (which excludes outliers) is 77.5-65% = 12.5%

The difference between the best and worst servers in each similar height group (which includes outliers) is 3-7%

Excluding outliers, the difference between the best and below average servers in each group is 2-4%

CONCLUSION

If we include outliers that are poor servers for their height, the difference in first serve points won is 18% for height-related factors & 3-7% for factors not relating to height.

If we exclude outliers that are poor servers for their height, the difference in first serve points won is 12.5% for height-related factors and less than 2-4% for non-height related factors.

On top of this, the height-related differences would be even greater if it wren't for the fact that shorter players are generally better at moving and rallying to back up their serves.

So "differences in free points on serve at the top 100 pro level are almost entirely due to height" is not "outright wrong" as you claim. It's basically accurate, although I'm happy to rephrase is as "primarily due to height".
Of course, it doesn't really matter, since even if the difference were only 50% due to height, it'd still be the cause of average and below average height players being so rare on the tour, and the cause of the boredom of watching faults and preparation for second serves, etc. It'd still mean that half of those free points are just for height, on top of points for better serve technique & points for better physiology for preventing big-serve-related injury.

Good post, if you had said primarily, I wouldn't have started looking for outliers. Nitpicking perhaps

 

[Chanwan]

OK, but it's an average across all surfaces, so the fact hard court & grass are >60% of the season is working against Rafa & he still comes out on top. If you're gonna arbitrarily weight hard court & grass more heavily, sure, Novak might have won an additional year or two. Some of the years that Rafa won were close between Rafa & Novak. However, the head to head between Rafa & Novak is almost equal & the number of slams is currently equal, and Novak is taller & has a better first serve, and his first serve return I guess is roughly as good or better, so you've got to expect that Nadal would have the edge without the much greater number of free points on serve for Novak.

Rafa had a pretty massive lead in slams and h2h before 2011 (9-1 in slams iirc).
Will get back in more detail later here

 

[socallefty]

What’s next? Change tennis rules so that fast players don’t have an advantage, physically stronger players don’t have an advantage, players who are not overweight don’t have an advantage, players with great hand-eye coordination don’t have an advantage, players who are young don’t have an advantage etc.? Or should we change it just for short players as the OP often makes arguments for?

Being short is a detriment or being a world leading athlete in many sports based on the age-old rules of the sport. Short people play sports like basketball, tennis, American football etc. at a worse level in general than taller people with comparable athletic and technique skills - so, they entertain the audience less by playing at a lower level. Why on earth would we change the rules to allow lesser talents to compete better? So that some short people in the audience feel better about themselves to see players who look like them competing? What about everyone else who is forced to endure a worser level of performance?

I want to see the best servers in the game and not have the rules be changed to make shorter players like Diego, Nishioka win more of their service games. Similarly many of the best returners are shorter players and the tallest guys like Isner, Opelka are amongst the worst returners - I don’t want the rules to be changed to allow the tallest players to compete better on return games either.

 

[canta_Brian]

The concept is flawed.

OP needs to stop thinking like a statistician and start thinking like a player.

No player will serve a single ball as if it is their current 2nd serve if they only win 50%ish. That stat is based on the idea that 50% wins is fine as you will make up the difference on the first serves in the current scenario. So 70% win on first serve + 50% on second gets the job done most of the time.

Nobody will just hit the 2nd serve that only wins 50% of the points for a single serve. What would be the point? So where would that leave us? Players would go risk reward on bigger first serves and who would that favour? Yep, taller servebot type players.

This would achieve the opposite of what you desire.

 

[AlecG]

What’s next? Change tennis rules so that fast players don’t have an advantage, physically stronger players don’t have an advantage, players who are not overweight don’t have an advantage, players with great hand-eye coordination don’t have an advantage, players who are young don’t have an advantage etc.?

No, because that would be stupid. Can you really not see the difference, or are you just pretending you can't see the difference?

I think you're mostly just being dishonest, but I'll explain the difference anyway.
Two attempts on serve in *every tournament* is a effectively charity for taller players who are generally worse at movement and aren't generally better when it comes to overall rallying & returning ability. It makes the sport less fun for most people to watch and play as there are far more faults, far more waiting for the next serve after a fault, far more unreturned serves & short boring rallies. It greatly reduces the number of players at average height or lower which greatly reduces the overall pool for talent in groundstrokes, movement, tactics, fitness, instinct, co-ordination, etc.

Soccer is more popular than most of the sports you mentioned largely because there is a larger talent pool. A sport that is based on talent, skill, fitness, co-ordination, etc and NOT height is generally going to be more entertaining for most people (both spectators and players) that one that is based largely on height. But this is particularly true if it's a single silly rule like two attempts on serve that biases the game towards height. Without that one silly rule of charity to taller players in *every* tournament, the talent pool would be better, stronger and more entertaining.

Do you prefer watching Cressy, Isner and Opelka to watching Dan Evans, Hewitt, Ferrero, Kwon & Nishioka or are you just pretending you do because you like getting free points on serve & your ego can't handle losing to shorter players with better overall skills?

 

[socallefty]

Do you prefer watching Cressy, Isner and Opelka to watching Dan Evans, Hewitt, Ferrero, Kwon & Nishioka or are you just pretending you do because you like getting free points on serve & your ego can't handle losing to shorter players with better overall skills?

No, because that would be stupid. Can you really not see the difference, or are you just pretending you can't see the difference?

I like watching Cressy and Opelka - would love for one of them to win Wimbledon someday. I enjoy Cressy’s S/V style and Opelka’s serve is a thing of beauty especially from courtside and he moves well compared to others who serve well like him. I don’t like Isner because he is so poor on return games mainly due to his poor movement. Ferrero was Ok - watched him sometimes. The others have never interested me after watching one or two matches for various reasons not correlated to their height. I don’t even know who Kwon is. When I watch tennis or am playing it, I don’t think a player‘s height is one of the top ten things I am focusing on.

Let‘s just say that the word you used above succinctly captures my opinion of you and your inane posts which are mostly fixated on changing the rules of tennis to favor short players. Basically your opinions come down to saying that you think people taller than you can play tennis better than you only because they are taller and this is unfair so that the rules of tennis should be changed. If you think you can spout these self-serving opinions without getting pushback from fans and players who like tennis as-is or who are tall themselves, you are out of your mind.

 

[AlecG]

The concept is flawed.

OP needs to stop thinking like a statistician and start thinking like a player.

No player will serve a single ball as if it is their current 2nd serve if they only win 50%ish. That stat is based on the idea that 50% wins is fine as you will make up the difference on the first serves in the current scenario. So 70% win on first serve + 50% on second gets the job done most of the time.

Nobody will just hit the 2nd serve that only wins 50% of the points for a single serve. What would be the point? So where would that leave us? Players would go risk reward on bigger first serves and who would that favour? Yep, taller servebot type players.

This would achieve the opposite of what you desire.

I am thinking like a player who knows the stats (like the pros and their teams do). The reason people try to hit aces on their first serve is because it doesn't really matter too much if it goes in or not. When it matters, they hit a reliable serve, because 51% (the median of points won on second serve including faults) is still better than 72% of 63% which is 45% (the median of points won on first serve when including faults).
There are a tiny number of players like Cressy who are statistically better off just hitting their second serve like their first serve than they would be hitting something very reliable on the second serve & the same would apply with only one attempt on serve. Isner might benefit (slightly) from going harder on his second serve even under the current rules. But the vast majority of players and even the majority of big servers do better on their second serve than on their first serve when you take faults into account. For example, Kyrgios wins 58% on his second serve, which I believe includes the tiny number of double faults. On his first serve he wins 80% of the 67% that go in which is 54%. So if he only had one serve, the second serve would be the better bet against most players. This is true of almost all servers including almost all tall servers, because it's the only thing that makes sense statistically.

 

[AlecG]

I like watching Cressy and Opelka - would love for one of them to win Wimbledon someday. I enjoy Cressy’s S/V style and Opelka’s serve is a thing of beauty especially from courtside and he moves well compared to others who serve well like him. I don’t like Isner because he is so poor on return games mainly due to his poor movement. Ferrero was Ok - watched him sometimes. The others have never interested me after watching one or two matches for various reasons not correlated to their height. I don’t even know who Kwon is. When I watch tennis or am playing it, I don’t think a player‘s height is one of the top ten things I am focusing on.

Let‘s just say that the word you used above succinctly captures my opinion of you and your inane posts which are mostly fixated on changing the rules of tennis to favor short players. Basically your opinions come down to saying that you think people taller than you can play tennis better than you only because they are taller and this is unfair so that the rules of tennis should be changed. If you think you can spout these self-serving opinions without getting pushback from fans and players who like tennis as-is or who are tall themselves, you are out of your mind.

You're not interacting with what I said. You're just making up false claims about me. Very dishonest.

Obviously I know I will get pushback, at least from those with irrational conservative bias and from a self-serving (pun intended) minority who like the charity points on serve. But I haven't encountered any pushback yet that is rational and considerate of what is best for the majority of spectators, amatuer players, potential players and potential pros.

 

[TheFifthSet]

But I haven't encountered any pushback yet that is rational and considerate of what is best for the majority of spectators, amatuer players, potential players and potential pros.

You were given more than enough substantive (and respectful) pushback, but had I known this undercurrent of dismissiveness permeated your thinking I probably would’ve disengaged much earlier lol.

 

[junior74]

Schwartzman is proof you can be a top 10 player with just a second serve

 

[Chanwan]

I am thinking like a player who knows the stats (like the pros and their teams do). The reason people try to hit aces on their first serve is because it doesn't really matter too much if it goes in or not. When it matters, they hit a reliable serve, because 51% (the median of points won on second serve including faults) is still better than 72% of 63% which is 45% (the median of points won on first serve when including faults).
There are a tiny number of players like Cressy who are statistically better off just hitting their second serve like their first serve than they would be hitting something very reliable on the second serve & the same would apply with only one attempt on serve. Isner might benefit (slightly) from going harder on his second serve even under the current rules. But the vast majority of players and even the majority of big servers do better on their second serve than on their first serve when you take faults into account. For example, Kyrgios wins 58% on his second serve, which I believe includes the tiny number of double faults. On his first serve he wins 80% of the 67% that go in which is 54%. So if he only had one serve, the second serve would be the better bet against most players. This is true of almost all servers including almost all tall servers, because it's the only thing that makes sense statistically.

I don't know where you find your Kyrgios' stats but they are off

Tennis Abstract: Nick Kyrgios Match Results, Splits, and Analysis

 

[Chanwan]

You're not interacting with what I said. You're just making up false claims about me. Very dishonest.

Obviously I know I will get pushback, at least from those with irrational conservative bias and from a self-serving (pun intended) minority who like the charity points on serve. But I haven't encountered any pushback yet that is rational and considerate of what is best for the majority of spectators, amatuer players, potential players and potential pros.

You got plenty of quality responses. Mind you, you're not exactly neutral in your phrasing as well. If we don't agree with you, we're conservative, irrational, supporting a "silly" rule that gives "charity points on serve".

Btw, football players are also taller than average. 6'0 in the Premier League. 5'11 in La Liga.

Average Height of Premier League Players [by Position]

Discover the height of the average Premier League player, plus the tallest and smallest goalkeepers, defenders, midfielders and forwards.

sqaf.club

Are EPL players bigger and stronger than La Liga players?

Answer (1 of 2): According to data from EA Sports’ PC game Fifa 17, yes. EPL (658 players analyzed): Average height: 182,52 cm Average weight: 76,02 kg Average strength: 67,92 La Liga (610 players analyzed): Average height: 180,57 cm Average weight: 74,20 kg Average strength: 65,93 It seems EP...

www.quora.com

 

[Chanwan]

You were given more than enough substantive (and respectful) pushback, but had I known this undercurrent of dismissiveness permeated your thinking I probably would’ve disengaged much earlier lol.

I am thinking like a player who knows the stats (like the pros and their teams do). The reason people try to hit aces on their first serve is because it doesn't really matter too much if it goes in or not. When it matters, they hit a reliable serve, because 51% (the median of points won on second serve including faults) is still better than 72% of 63% which is 45% (the median of points won on first serve when including faults).
There are a tiny number of players like Cressy who are statistically better off just hitting their second serve like their first serve than they would be hitting something very reliable on the second serve & the same would apply with only one attempt on serve. Isner might benefit (slightly) from going harder on his second serve even under the current rules. But the vast majority of players and even the majority of big servers do better on their second serve than on their first serve when you take faults into account. For example, Kyrgios wins 58% on his second serve, which I believe includes the tiny number of double faults. On his first serve he wins 80% of the 67% that go in which is 54%. So if he only had one serve, the second serve would be the better bet against most players. This is true of almost all servers including almost all tall servers, because it's the only thing that makes sense statistically.

Pardon my French, but this shows a basic lack of understanding of the sport.

 

[Chanwan]

Reasons why a lot of us (all the ones giving you pushback in this thread) aren't convinced of your rule change

• It would fundamentally alter the sport. It's not a small change, it's massive.
• Related: It would make all comparisons between players from before the rule change and after redundant (and that's big for some of us)
• It would be the final death blow to serve & volley
• It would make the very difficult skill of returning a tough first serve redundant
• It would more or less remove aces from the game (and some of us like them)
• It would remove the aesthethic of the first serve (I personally find it much more beautiful than the 2nd and I think I'm not alone)
• And as @TheFifthSet said it would most likely lead to less parity. Right now, taller players can make up for their worse movement with a great first serve while shorter players can make up for their not so great first serve with great movement.
• I.e. there's a risk we'll see more of the same style and differences are (to many of us) part of what makes the sport fun to watch. Isner vs. Diego is a contrast in styles.

 

[Chanwan]

OK, but it's an average across all surfaces, so the fact hard court & grass are >60% of the season is working against Rafa & he still comes out on top. If you're gonna arbitrarily weight hard court & grass more heavily, sure, Novak might have won an additional year or two. Some of the years that Rafa won were close between Rafa & Novak. However, the head to head between Rafa & Novak is almost equal & the number of slams is currently equal, and Novak is taller & has a better first serve, and his first serve return I guess is roughly as good or better, so you've got to expect that Nadal would have the edge without the much greater number of free points on serve for Novak.

Here's (part of) your list:
Nadal 2008-2014
Novak 2015-2016
Nadal 2017-2019

2009 and 2012 - Rafa sucked/was out with injury for half the year, so obviously he wouldn't be world no. 1 those years.
2011: Clay: Rafa is ahead 3,5 % in return points won, Novak is 2,5 % ahead in serve points won:
(2011 stats here, Djokovic outperforming Rafa on 2nd serve serve points: https://www.atptour.com/en/stats/2nd-serve-points-won/2011/clay/all/, while Rafa is ahead in the return department. Difference is small enough to make for close matches though https://www.atptour.com/en/stats/2nd-serve-return-points-won/2011/clay/all/)
HC: Novak 3,5 % ahead in RoS: https://www.atptour.com/en/stats/2nd-serve-return-points-won/2011/hard/all/, while being less than 2,5 % shy of Rafa on 2nd serve points won https://www.atptour.com/en/stats/2nd-serve-points-won/2011/hard/all/
Grass: Neck & neck

I.e. pretty darn close overall. Grass could go both ways, while Rafa has a very small edge on clay and Novak has a similar small edge on HC.
So let's look closer.

GS tournaments (i.e. where most points are earned) from QF onwards:
Rafa: 50,9 % & 57,4 %
Djokovic: 52,1 % 57,1 % (i.e. edge Djokovic - source Ultimate tennis stats)

Let's look even closer though. Novak against no. 1 in 2011 (i.e. their first 6 matches, Djokovic became no. 1 after Wimbledon).
Result?

Novak was outright killing him on this metric: 57,4 % behind his own 2nd serve and 55,2 % against Rafa's.
At the US, Rafa was 42,2 behind his own 2nd serve and 55,8 % against Novak's. I.e. Novak would most likely have won the vast majority of their 7 matches (maybe all). And he would have been no. 1.

Side note: He would have beaten Federer in the FO semi and (if we rely solely on the numbers from their 2 clay clashes in 2011) also have beaten Rafa in the final (Novak won 61 % vs. the Rafa 2nd serve in Madrid (Rafa won 48 %) and a whopping 75 % in Rome (Rafa won 53 %).

In short, giving Rafa 11 YE no. 1's under your system is, sorry to say, just wrong.

 

[canta_Brian]

I am thinking like a player who knows the stats (like the pros and their teams do). The reason people try to hit aces on their first serve is because it doesn't really matter too much if it goes in or not. When it matters, they hit a reliable serve, because 51% (the median of points won on second serve including faults) is still better than 72% of 63% which is 45% (the median of points won on first serve when including faults).
There are a tiny number of players like Cressy who are statistically better off just hitting their second serve like their first serve than they would be hitting something very reliable on the second serve & the same would apply with only one attempt on serve. Isner might benefit (slightly) from going harder on his second serve even under the current rules. But the vast majority of players and even the majority of big servers do better on their second serve than on their first serve when you take faults into account. For example, Kyrgios wins 58% on his second serve, which I believe includes the tiny number of double faults. On his first serve he wins 80% of the 67% that go in which is 54%. So if he only had one serve, the second serve would be the better bet against most players. This is true of almost all servers including almost all tall servers, because it's the only thing that makes sense statistically.

OK, it tried to explain some of the issues with your theory in a tennis way. It might be more valuable to explain them in a science way. You have an issue with construct validity in your assertion. You are trying to use the statistics gained under one set of conditions to attempt to explain what you believe would happen under another set of conditions. This is scientifically flawed.

 

[Ombelibable]

I vote for a third serve.

@Third Serve

 

[AlecG]

You were given more than enough substantive (and respectful) pushback, but had I known this undercurrent of dismissiveness permeated your thinking I probably would’ve disengaged much earlier lol.

OK. Can you remind me of an example of substantive reason for every single tournament having two attempts on serve? It was a long time ago but I'm pretty sure your criticism was that it would cause Nadal to dominate. Not a great reason to oppose it in my opinion, but I also never said this rule change should be made at every tournament. My suggestion is to try it at at least one 250 and see how it goes. Any substantive, rational objections to that specifically?

 

[AlecG]

I don't know where you find your Kyrgios' stats but they are off

Tennis Abstract: Nick Kyrgios Match Results, Splits, and Analysis

Saying you don't know where I found them and then also saying they're off is pretty funny. They're from the official ATP leaderboard, so last 12 months. If you want to provide evidence that using career stats would make current first serves sensible as one's only serve, go ahead.