Homogenized surfaces myth or fact?

http://heavytopspin.com/2013/04/08/the-mirage-of-surface-speed-convergence/

Here again, the most diverse results occurred during the 5-year span from 2003 to 2007, when hard-court aces were 51.3% higher than clay-court aces.  Since then, the difference has fallen to 46%, still a relatively large gap, one that only occurred in two single years before 2003.

If surfaces are converging, why is there a bigger difference in aces now than there was 10, 15, or 20 years ago? Why don’t we see hard-court break rates getting any closer to clay-court break rates?

However fast or high balls are bouncing off of today’s tennis surfaces, courts just aren’t playing any less diversely than they used to.  In the last 20 years, the game has changed in any number of ways, some of which can make hard-court matches look like clay-court contests and vice versa.  But with the profiles of clay and hard courts relatively unchanged over the last 20 years, it’s time for pundits to find something else to complain about.


I encourage you all to look at the graphs and the whole article. Basically the writer states that ace rate differences between clay and hardcourt has not decreased and the difference in breaking on hard courts and clay courts also has not decreased. Meaning that in recent years going back to the 90s the numbers have been pretty stable. If the surfaces have been homogenized you would expect a smaller difference in break rates and ace rates between hardcourts and clay courts. However this is not happening. I didn't find this but saw it on another site so credit to the writer. We have heard a lot on this site and others that homogenization is killing the game but this writer provides strong evidence that debunks the idea of homogenized surfaces.

So what has created the dominant base line game and the end of S and V. Well I have always maintained that it is the better racquets and the better strings and that seems to be the case. Coming in is risky, you can miss the volley or get passed. When you can win the point by blasting winners more easily from either wing then why take the risk of coming into the net. A short forehand will suffice, in the past to finish points you just had to get to the net. It is very difficult for example to hit winners from the baseline with a wooden racquet.

Who is the writer? The comments underneath are also very much worth a read.

To be honest Johny I don't know who this guy is and saw this on another site but he does provide great evidence that seems to debunk the idea that clay and hardcourts are playing more similarly now then ever. It is well researched piece and goes against the running storyline that the surfaces have been homogenized and that is why we are seeing certain players win across those surfaces.

We know that Wimbledon slowed down the grass, that the new AO surface is slower than the old Rebound Ace and that carpet (a fast surface) has virtually disappeared from the tour (replaced by slower surfaces). Those factors alone have homogenized surface speed overall.

The response from "TennisFan1" is excellent. Number of Aces/Breaks do not determine the speed of a surface. If you have a receiver, whose movement is like he is playing on wet cement (thanks SoCal), aceing such a player is much easier. S&V style is also a factor.

Carpet (and Wood) are the fastest surfaces, and are now defunct.

@JJ... the blog writer is Jeff Sackmann. (can be found using his Twitter link).

In what follows, we’ll use two stats: ace rate and break rate.  When courts play faster, there are more aces and fewer breaks of serve.  The slower the court, the more the advantage swings to the returner, limiting free points on serve and increasing the frequency of service breaks.

SoCal... what is the return percentage for Djokovic on Clay, Hard (current) and Grass (current)?

The above premise is a bit too simple a model. The admission of the limits of the model are also clearly stated...

While things like ball bounce and shot speed can be quantified, they haven’t been tracked for long enough to help us here.  We’re stuck with the same old stats — aces, serve percentages, break points, and so on.

Thus, when we talk about “surface speed” or “court speed,” we’re not just talking about the immediate physical characteristics of the concrete, lawn, or dirt.  Instead, we’re referring to how the surface–together with the weather, the altitude, the balls, and a handful of other minor factors–affects play.  I can’t tell you whether balls bounced faster on hard courts in 2012 than in 1992.  But I can tell you that players hit about 25% more aces.

You take three parameters and this is the conclusion, what happens if you add temperature, humidity, height from sea level and coefficient of friction which can be measured by standard scientific methods to this data and repeat the same analysis.

This is subjective and biased data selection to prove a hypothesis.

PS: Every televised match in the last 5 years, has shown temperature, wind speed and humidity, IIRC. By this method, AO in January at 120 degree F (40+ celsius) is the same as USO in August-September.

JuliusHMarx wrote:We know that Wimbledon slowed down the grass, that the new AO surface is slower than the old Rebound Ace and that carpet (a fast surface) has virtually disappeared from the tour (replaced by slower surfaces). Those factors alone have homogenized surface speed overall.

We know wimbeldon has but that is one tournament. Both Plexicushion and Rebound Ace are slow hardcourts both of them. In fact the differences between the number of aces hit on hard court and clay are bigger today than they were in 90s, to me that seems to go a long way in debunking the idea that hardcourts are slower than in the past. The guys study by the way doesn't take Grass into question because not enough numbers to provide an accurate sample. I agree wimbeldon is slower and thank god for it. But AO has always been a slow hardcourt since the 80s.

If you replace carpet with a hard court, that will increase homogenization simply because hard courts are slower.
If you slow down the existing hard courts as at the AO (which is what they did) and slow down Wimby that will homogenize the 4 GS tournaments.

JuliusHMarx wrote:If you replace carpet with a hard court, that will increase homogenization simply because hard courts are slower.
If you slow down the existing hard courts as at the AO (which is what they did) and slow down Wimby that will homogenize the 4 GS tournaments.

I agree there when did they start replacing carpet with hard, has it been recently I don't know the answer to that. If my memory serves me right carpet started dying down a long time ago like in 90s, but I could be wrong.

laverfan wrote:The response from "TennisFan1" is excellent. Number of Aces/Breaks do not determine the speed of a surface. If you have a receiver, whose movement is like he is playing on wet cement (thanks SoCal), aceing such a player is much easier. S&V style is also a factor.

Carpet (and Wood) are the fastest surfaces, and are now defunct.

@JJ... the blog writer is Jeff Sackmann. (can be found using his Twitter link).

In what follows, we’ll use two stats: ace rate and break rate.  When courts play faster, there are more aces and fewer breaks of serve.  The slower the court, the more the advantage swings to the returner, limiting free points on serve and increasing the frequency of service breaks.

SoCal... what is the return percentage for Djokovic on Clay, Hard (current) and Grass (current)?

The above premise is a bit too simple a model. The admission of the limits of the model are also clearly stated...

While things like ball bounce and shot speed can be quantified, they haven’t been tracked for long enough to help us here.  We’re stuck with the same old stats — aces, serve percentages, break points, and so on.

Thus, when we talk about “surface speed” or “court speed,” we’re not just talking about the immediate physical characteristics of the concrete, lawn, or dirt.  Instead, we’re referring to how the surface–together with the weather, the altitude, the balls, and a handful of other minor factors–affects play.  I can’t tell you whether balls bounced faster on hard courts in 2012 than in 1992.  But I can tell you that players hit about 25% more aces.

You take three parameters and this is the conclusion, what happens if you add temperature, humidity, height from sea level and coefficient of friction which can be measured by standard scientific methods to this data and repeat the same analysis.

This is subjective and biased data selection to prove a hypothesis.

PS: Every televised match in the last 5 years, has shown temperature, wind speed and humidity, IIRC. By this method, AO in January at 120 degree F (40+ celsius) is the same as USO in August-September.

I don't see the bias and it is pretty strong evidence. Julius raises a good point as to the disappearance of carpet but the authors study isn't referencing that. What it does I think prove is that this idea that clay courts have become faster and the hardcourts have become slower is not very well supported by the facts. I think this information goes a long way to debunking this idea. We know that the USO and wimby were slowed down in the early 2000s but this idea that the surfaces are getting slower and slower and that these hardcourts are now more like clay courts to me is not supported.

Probably around then. Pretty sure if you compare 1995 and 2010 there is a lot of evidence for homogenization overall.
So many other changes as well - rackets, strings, balls, fitness - that it's almost impossible to isolate one factor and say 'x has happened because of y'.

Yes but we are talking about homogenization of surfaces. The game has become more baseline oriented really since the advent of the graphite or composite racquet. The strings and racquets play a bigger role in my mind for the baseline orientation of the game. But on the specific issue that we hear online about how these hardcourts are so slow and are slower and slower each year this study I think casts serious doubt on that proposition.

I don't know about year-on-year - there's probably not much changed for the last 3 or 4 years.
But the problem with using one set of stats e.g. ace count and making conclusions about surface speed is that it doesn't take into account any of the other changes that also affect ace count and which might counteract the homogenization of surfaces.
Although you could then argue that homogenization of surfaces was necessary to counteract those other factors.
For me homogenization of surfaces by itself isn't as important as homogenization of style of play. I prefer to see more variety of play on offer, however it is achieved. At the moment I'm a somewhat bored of watching tennis because the combination of factors means I watch more-or-less the same style of play, same tactics regardless of surface.
Other people prefer it that way. If 10 people will pay to watch a 6 hour slug-fest and only 1 will pay to watch S&V, no-ones going to set up an S&V tournament. I'll just stop watching, it's no big deal. There are plenty of books I've been meaning to read.

Honestly homogenized surface or not I dont care, it brings its own challenge , yes its very difficult for Tennis pundits to accept it but in my view it has made much impact on a player winning everything.

Lemme put this had the surface been homogenized in 90's, clay and hard courts behaving very slow along with slow grass clay court players like Brugera would have won lot more slams taking something out of Pete's tally, on the otherside Pete would have won FO has he would have got used to playing slow conditions, so in the end the tally of slams won by Pete would have been still the same however Pete might have had a complete CV in comparison.

Thats why total slam count is more important than completion of all slams.

The same would have been the case for Rafa in 90's, he would have trained hard for fast conditions and hence would have still won 13 slams however he might lost a lot more FO but would have ended up winning lot more of Wimbledon and USO.

Casual fans anyways give a Poopie to it.   

He doesn't use one stat though Julius he also uses difference in break percentages vis a vis hardcourt and clay. Also he bases his numbers on matchups. He looks at players who in the same year played someone on clay and on hardcourt.

I think there is a similar style of play in tennis and we could use more variety. But as you point out Julius there is an incentive on part of the broadcasters and tournament owners to get more bang for their buck with longer matches and rallies.

"Homogenised" is often used to describe tennis surfaces but I don't think it's the right word.

There are still differences in tennis surfaces. Why is Rafa maybe the best ever on clay but has only a single indoor title? Why is Andy so successful on grass but has never appeared in a clay final? Surfaces still have significant differences.

The variety in the slams has been reduced though. The new grass at Wimbledon slowed the speed and increased the bounce. USO slowed their surface in 2001 and again in 2002 I believe. As mentioned above, AO slowed their surface too.

Along with racquet developments, this means someone like Novak can play pretty much the same way at all 4 slams and be in with a shout in all of them. The 'old skills' are still relevant - a good slice and a good volley are still an asset at Wimbledon, sliding confidently helps a player at RG - but they are not vital anymore. A top level baseliner will be capable of winning any event in the calendar.

HM Murdoch wrote:"Homogenised" is often used to describe tennis surfaces but I don't think it's the right word.

You're right. "Homogenised" implies they are almost identical, which clearly isn't the case.

Both graphs actually support the idea that in the last 5-10 years surfaces are more homogenised. his data supports the idea that isn't a change vs 20 years ago but vs 5-10 years ago, the way his graphs fall at the end, acctually supports homognisation over that period!

How does it do that HB it actually show that there is a larger gap now between hardcourt and clay than the 90s it shows a high in 2003- 2007 between break percentages on clay vs on hard. And we are slightly off the 2000s high but higher than what existed in the 90s.

There was a time long ago, when Clay "specialists" refused to play at W, but Grass players played on Clay and did not have any issues. (See Laver and Rosewall in 1962-69).

Basel was carpet in 2006, there is a CZE v SUI DC on Carpet in 2007.

Take a look at this -

http://tt.tennis-warehouse.com/showthread.php?t=229026

http://www.marca.com/edicion/marca/tenis/es/desarrollo/1179715.html

Here is a Google Translate version...

The ATP searches with this decision that the game becomes more attractive and has less points earned by aces. "Physical force kills the talent," said the Romanian Ion Tiriac, Madrid Masters promoter. But the Croatian Mario Ancic not match that. "I'm against this change. Unable to unify all surfaces because of the quality of the game. If we want players to develop all his shots, you need to play on fast surfaces, otherwise the game will be very boring," said .

No matter what parameters are used to analyze this, the decision to homogenize is very clear and indisputable.

The players adapted to the surfaces as they changed, the game is now baseline oriented, by decree and design. When Stakhovsky, Stepanek, Llodra and Dodig play S&V, it surprises people nowadays. They came from a Doubles background.

I agree with iC and his evaluation.

Here is another set for reading pleasure...

Wilander, who currently operates the traveling tennis clinic Wilander on Wheels, added that slower courts at the various ATP stops and the Grand Slams, as well as heavier balls in use, have led to more parity in the sport.

Players "have a chance to win on all surfaces because the surfaces are so similar," he says. "Wimbledon is more similar to what the French is like; the hardcourts are slower, indoors is slower. They're able to play the game on all surfaces."

http://usatoday30.usatoday.com/sports/tennis/story/2012-04-16/weekly-net-post-demise-of-the-clay-court-specialist/54319198/1

The blog in the OP is a ramble/rant and has no bearing on reality.

Hardly a ramble or a rant Laverfan there is more statistical evidence that he provides than the people who claim that the hardcourts have been slowed down without really any evidence. The USO was slowed down in 03 and we know wimbeldon was slowed down, but the idea that the courts across the boards keep getting slower and slower I haven't seen any evidence for.

socal1976 wrote:Hardly a ramble or a rant Laverfan there is more statistical evidence that he provides than the people who claim that the hardcourts have been slowed down without really any evidence.

Based on an average graph of Aces and Break Points on different surfaces while acknowledging the limitations of such analysis? I provided a link to an ATP decision and you are saying we need more evidence?

Does ANY tourney on any surface publish the exact composition of the surface they use, including scientific KPIs for such? ITF just publishes a classification. For example, does US Open publish daily measurements of coefficients of friction and restitution on a per match basis?

socal1976 wrote:The USO was slowed down in 03 and we know wimbeldon was slowed down, but the idea that the courts across the boards keep getting slower and slower I haven't seen any evidence for.

Which is based on an interview referenced here...

http://bleacherreport.com/articles/447146-are-us-open-tennis-courts-slowing-down (This article is dated 31 Aug 2010).

Jim Curley, tournament director at the U.S. Open, said organizers began slowing their own courts down in 2002, adding extra sand to the paint, and did so again before the 2003 tournament to make the court fair for baseliners and serve-and-volleyers alike.

This is not a claim, but an admission from a person responsible and accountable for such changes. There is no need for statistical gymnastics to prove or disprove it.

Here is a set of ATP match facts over a career (the surfaces are shown in the images)...

1.
2.
3.
4.
5.
6.

Based on the article referenced in the OP, can you discern if,

a. What years are these statistics from?
b. Do these have anything in common, if at all?
c. Do these fit the OP's conjecture?

Good find Socal. An interesting read.

@ LF

1. The link you provided purely refers to removal of carpet. It provides no evidence that hard courts have been slowed generally.

2. Interesting to note which player won the US in the year it was slowed down.

3. Tennis was clearly moving to a baseline orientated game before 2003. Players like Roddick and Hewitt didn't adapt to any slowdown. They already played that way.

4. The article is clearly not even close to a rant. Assuming that the statistics are accurate, do you accept that it suggests there remains a significant difference between surfaces? What explanation would you give to the figures?

5. Asking Socal to try and work out what years your stats are from is ridiculous as that is clearly not possible. It would be far more useful to explain why you consider them relevant to the article?

Worth noting of course that in the early 90s Courier won slams on both clay and hard and reached a Wimbledon final. Agassi won both at Wimbledon and RG. How did Coria get on at Wimbledon compared to the French?

The guy who wrote the article admits that he can't determine, using stats, if the speed that the ball bounces off the court has change or not :-
"Thus, when we talk about “surface speed” or “court speed,” we’re not just talking about the immediate physical characteristics of the concrete, lawn, or dirt. Instead, we’re referring to how the surface–together with the weather, the altitude, the balls, and a handful of other minor factors–affects play. I can’t tell you whether balls bounced faster on hard courts in 2012 than in 1992. But I can tell you that players hit about 25% more aces."

So if the argument is that the surfaces (and only the surfaces) have converged, he is not actually presenting (or even trying to present) an argument for or against that.

Born Slippy wrote:1. The link you provided purely refers to removal of carpet. It provides no evidence that hard courts have been slowed generally

I suggest reading Ion Tiriac (the same TD who also "inflicted" blue clay on the likes of Nadal and Djokovic) and Ancic's statements.

Born Slippy wrote:2. Interesting to note which player won the US in the year it was slowed down.

Do you know for a fact that ALL US courts were slowed or not? Any links that point to Ashe being faster or Armstrong? There are many factors affecting court speed. There is an ESPN interview with Brad Gilbert (during the US courts being slippery and players threatening walk out incident) where he went on court and showed how over a period of time including ToD, the court speeds vary, that too on different courts, with the same sand content in the paint. Temperature at the time of match has a significant impact. The surface is just one factor.

Born Slippy wrote:3. Tennis was clearly moving to a baseline orientated game before 2003. Players like Roddick and Hewitt didn't adapt to any slowdown. They already played that way.

Do you recall Henman's rant at W court speeds? or Wilander's statement linked in my post?

Born Slippy wrote:4. The article is clearly not even close to a rant. Assuming that the statistics are accurate, do you accept that it suggests there remains a significant difference between surfaces? What explanation would you give to the figures?

The goal of the article is to show using "paired" matches using aces and bps that surfaces are "divergent". Let me show you why this makes no sense.

Two matches, roughly thirty days apart in 2014 between the same two players.

A. Total number of Aces in Match 1 = 12, Match 2 = 12
B. Total number of Bps converted in Match 1 = 5/9, Match 2 = 9/12.

Can you determine the speed of the surface given A and B?

Born Slippy wrote:5. Asking Socal to try and work out what years your stats are from is ridiculous as that is clearly not possible. It would be far more useful to explain why you consider them relevant to the article?

The images I linked to are "career" stats for two players, with speciality on two different surfaces. Is that a good starting point?

This is an interesting debate.

Personally I don't think that you can attribute any significance to any one or two factors. It is probably a combination of the lot that has led to the game becoming more baseline orientated. A perfect storm if you will, where by one factor alone would have little impact but when they are added up the effect becomes what we see today.

While tennis produces a lot of statistics they are rarely, if ever, used in any 'scientific' way. So until any statistical significance testing is done in a controlled environment it will always be conjecture. The research quoted may well be thorough but it is not scientific with no values with which to reject (or accept) your null hypothesis.

Using human beings as test subjects already introduces an element of bias and variability that would be impossible, if not very hard, to eliminate because quite simply people have good days and others have bad days in sport (from a serving and returning perspective which both impact on the statistics produced).

The U.S. Open courts are classified as "medium-fast" and consist of 10 different layers. The courts are resurfaced beginning in May and the process is completed a few weeks before the tournament begins.

Since the center court at Arthur Ashe Stadium handles five matches a day -- and the courts are washed when action ends -- the sand tends to wear down as the two weeks progress.

"At the business end of the tournament," Gilbert said, "when they've been played on for so many matches, they quicken up even more."

These subtleties would be lost on club players, but elite players are extremely sensitive to small variations in pace. Changes in temperature, humidity and wind can dramatically affect how the ball moves through the court. Tennis courts are like snowflakes; no two are alike.

But, conspiracy theorists, apparently these U.S. Open courts have been the same speed for seven years.

This is from 2009...

http://sports.espn.go.com/sports/tennis/notebook?page=notebook/tennis09032009

the only ral fact is that nadal is a 13 time slam winner and not everybody can assume that

naxroy wrote:the only ral fact is that nadal is a 13 time slam winner and not everybody can assume that

I would love to see him win more and give Federer a good run for the GOAT (or GOTE) moniker.

The discussion is not meant to take away anything from Nadal's achievements. He has adapted well to different surfaces.

@ Laver:

1. I have read the comments quoted. I haven't read the original Spanish article but understand that the context is removal of carpet. From the quoted comments I can't see that there is a suggestion of a wider ATP policy.

2. Agreed but those external factors (weather etc) should be absorbed in a large enough statistical sample. I don't know the precise nature of every US hardcourt. For all I know they may all be half the speed they were 5 years ago. My only point was that there is limited evidence in support and the article seems to provide an interesting analysis that whatever slowing down has taken place seems to have had a very limited impact on 2 points where I would expect it to impact.

3. Yes I think Tim's rant was more aimed at slower balls from memory. However, I don't understand the relevance to my point. What type of game did Roddick/Hewitt play before 2003? Even Federer for all his skills was nominally a baseliner.

4. One match clearly wouldn't work. I am unsure of the relevance of BPs. Are you suggesting that could be evidence of courts being slower? The article suggests rather more than one match per year. If breakpoints is random then it will be absorbed in a larger sample. You appear to have ignored my questions. If there are 100 matches on one court and 1000 aces and 100 matches between the same players on another court and 200 aces would that be good evidence that the first court was quicker than the second?

5. I know what the stats show. The first player looks like someone like Roddick. The second could be Bruguera? Fun guessing game but what's the point?

laverfan wrote:

socal1976 wrote:Hardly a ramble or a rant Laverfan there is more statistical evidence that he provides than the people who claim that the hardcourts have been slowed down without really any evidence.

Based on an average graph of Aces and Break Points on different surfaces while acknowledging the limitations of such analysis? I provided a link to an ATP decision and you are saying we need more evidence?

Does ANY tourney on any surface publish the exact composition of the surface they use, including scientific KPIs for such? ITF just publishes a classification. For example, does US Open publish daily measurements of coefficients of friction and restitution on a per match basis?

socal1976 wrote:The USO was slowed down in 03 and we know wimbeldon was slowed down, but the idea that the courts across the boards keep getting slower and slower I haven't seen any evidence for.

Which is based on an interview referenced here...

http://bleacherreport.com/articles/447146-are-us-open-tennis-courts-slowing-down (This article is dated 31 Aug 2010).

Jim Curley, tournament director at the U.S. Open, said organizers began slowing their own courts down in 2002, adding extra sand to the paint, and did so again before the 2003 tournament to make the court fair for baseliners and serve-and-volleyers alike.

This is not a claim, but an admission from a person responsible and accountable for such changes. There is no need for statistical gymnastics to prove or disprove it.

Yes I know about that admission and I don't deny the USO was slowed down after 02 and wimbeldon at the start of the 2000s that is pretty much common knowledge. But what we hear is that since then there has been an across the board slow down of all the hardcourts. That simply isn't accurate and this guys stats, charts, and methodology are pretty sound.

JuliusHMarx wrote:The guy who wrote the article admits that he can't determine, using stats, if the speed that the ball bounces off the court has change or not :-
"Thus, when we talk about “surface speed” or “court speed,” we’re not just talking about the immediate physical characteristics of the concrete, lawn, or dirt.  Instead, we’re referring to how the surface–together with the weather, the altitude, the balls, and a handful of other minor factors–affects play.  I can’t tell you whether balls bounced faster on hard courts in 2012 than in 1992.  But I can tell you that players hit about 25% more aces."

So if the argument is that the surfaces (and only the surfaces) have converged, he is not actually presenting (or even trying to present) an argument for or against that.

He is looking on what would be a logical outcome statistically if the surfaces were homogenized. IE that we would see more the gap between breaking and aces between hardcourt and clay reduced, we haven't seen that and in fact have seen the opposite trend. No one can tell you what every factor in the universe on court speed for every single court on every single match played over the last ten years. But if the courts had been slowed (hardcourts) you would expect different trends on these two statistics and in fact we don't see that.

Laverfan, why don't you just tell us the relevance of your stats instead of having us guess. Raw stats without any information giving it context like you provided is not something I can respond to.

Johnyjeep wrote:This is an interesting debate.

Personally I don't think that you can attribute any significance to any one or two factors. It is probably a combination of the lot that has led to the game becoming more baseline orientated. A perfect storm if you will, where by one factor alone would have little impact but when they are added up the effect becomes what we see today.

While tennis produces a lot of statistics they are rarely, if ever, used in any 'scientific' way. So until any statistical significance testing is done in a controlled environment it will always be conjecture. The research quoted may well be thorough but it is not scientific with no values with which to reject (or accept) your null hypothesis.

Using human beings as test subjects already introduces an element of bias and variability that would be impossible, if not very hard, to eliminate because quite simply people have good days and others have bad days in sport (from a serving and returning perspective which both impact on the statistics produced).

JJ, no one denies that the game has gone more baseline oriented. The issue here is that many claim it is a result of continually slowing down the hardcourts post the early 2000s. We have not seen that based on the stats and analysis provided. The fact that this guy compares players who in a year played each other both surfaces actually makes it more accurate not less so. Why because you get the same serving motion vs. the same returner on a different surface and elimates the outlier affects of speciaists who pop up on one or the other surface. I personally think it has a lot more to do with the strings and racquets and that this whole courts slowing down and becoming homogenized is not true. Except in a couple of cases over a decade ago, and frankly they should have slowed wimby down it was freaking unwatchable in the late 90s.

HM Murdoch wrote:"Homogenised" is often used to describe tennis surfaces but I don't think it's the right word.

There are still differences in tennis surfaces. Why is Rafa maybe the best ever on clay but has only a single indoor title? Why is Andy so successful on grass but has never appeared in a clay final? Surfaces still have significant differences.

The variety in the slams has been reduced though. The new grass at Wimbledon slowed the speed and increased the bounce. USO slowed their surface in 2001 and again in 2002 I believe. As mentioned above, AO slowed their surface too.

Along with racquet developments, this means someone like Novak can play pretty much the same way at all 4 slams and be in with a shout in all of them. The 'old skills' are still relevant - a good slice and a good volley are still an asset at Wimbledon, sliding confidently helps a player at RG - but they are not vital anymore. A top level baseliner will be capable of winning any event in the calendar.

Excellent big picture view of things. The issue is that there is to some extent less variety between the slams then there once was and carpet being gone as Julius has pointed out has changed that as well. This study looks at the idea that hardcourts are continually slowed down to play more like clay. And I think it goes a long way to debunking that one critique. I like how you put it that the old skills are still relevant like slice and volleys, still a great number of points finish at the net. The issue I think is that and S an V has disappeared as a principal style and this is long term trend that even outdates luxilon strings and the big slow down of early 2000s. Should we try to bring back S and V? Is it worth radical changes to the game that would be required?

Born Slippy wrote:@ Laver:

1. I have read the comments quoted. I haven't read the original Spanish article but understand that the context is removal of carpet. From the quoted comments I can't see that there is a suggestion of a wider ATP policy.

Ancic and Tsonga did not want to "slow" down courts. Wilander's comments are also relevant for "sameness" of courts. Both are being ignored and the focus seems to be on carpet removal. Tiriac also says that "physical" strength will kill skills.

Born Slippy wrote: 2. Agreed but those external factors (weather etc) should be absorbed in a large enough statistical sample. I don't know the precise nature of every US hardcourt. For all I know they may all be half the speed they were 5 years ago. My only point was that there is limited evidence in support and the article seems to provide an interesting analysis that whatever slowing down has taken place seems to have had a very limited impact on 2 points where I would expect it to impact.

It is an "interesting" analysis with a clear admission of it's limitations. The limitations of such an "interesting" analysis are being ignored.

Born Slippy wrote:3. Yes I think Tim's rant was more aimed at slower balls from memory. However, I don't understand the  relevance to my point. What type of game did Roddick/Hewitt play before 2003? Even Federer for all his skills was nominally a baselined.

Everyone is a nominal base-liner, but can adapt.

https://www.youtube.com/watch?v=m0UZ-3W-bgs

https://www.youtube.com/watch?v=CExF6ERmJQA

https://www.youtube.com/watch?v=UgEx5B1lpqQ

Born Slippy wrote:4. One match clearly wouldn't work. I am unsure of the relevance of BPs. Are you suggesting that could be evidence of courts being slower? The article suggests rather more than one match per year. If breakpoints is random then it will be absorbed in a larger sample. You appear to have ignored my questions. If there are 100 matches on one court and 1000 aces and 100 matches between the same players on another court and 200 aces would that be good evidence that the first court was quicker than the second?

Such a sample is unlikely in real world. Laver-Rosewall played ~140 matches vs each other. The current Top 4 have played about 30+ matches against each other. If you had Dr Ivo or Isner (recall 113 aces from Isner in MahIsner) and another player, the ace/bp balance is completely different, no matter what the court speed is.

Born Slippy wrote:5. I know what the stats show. The first player looks like someone like Roddick. The second could be Bruguera? Fun guessing game but what's the point?

The point was that number of aces or bps are not a very good measure of surface speeds. I will post the remainder on SoCal's post.

socal1976 wrote:Laverfan, why don't you just tell us the relevance of your stats instead of having us guess. Raw stats without any information giving it context like you provided is not something I can respond to.

This is from three different time "windows".

The usage of Aces and Bps and their corresponding convergence or divergence is not a measure of surface speeds.

The highlighted columns is what I want to point out.

From 1990-2014, Bps/match is steady at 11.7-11.8 (on Hard).

I would discard Kuerten-Grass because of only 15 matches. Federer-Grass at 12.3 is higher than Rafter-Grass, does that mean Federer played on Grass which was faster than Rafter's time window? If that is not the case, then Bps are not relevant to a surface at all.

Aces/match for Federer-Grass at 10.3  is higher than Rafter-Grass at 7.3. Was Rafter playing on "slower" grass?  If that is not the case, then aces are not relevant to a surface at all.

Aces/match for Federer-Hard at 7.4 is higher than Rafter-Hard at 6.6. Was Rafter playing on "slower" hard?  If that is not the case, then aces are not relevant to a surface at all.

The answer may lie in the the following partially...

socal1976 wrote:Well I have always maintained that it is the better racquets and the better strings and that seems to be the case.
Coming in is risky, you can miss the volley or get passed. When you can win the point by blasting winners more easily from either wing then why take the risk of coming into the net. A short forehand will suffice, in the past to finish points you just had to get to the net. It is very difficult for example to hit winners from the baseline with a wooden racquet.

Better medical technology and better recovery techniques should not be ignored either. Coming in is risky, only if the opponent has the ability to chase the ball indefinitely. The Nalbandian-Hewitt W 2002 final is a very interesting battle between baselines, even when the "grass" was relatively fast. If you see the match again, you can look at the wear patterns on the grass and see the area close to net in almost the same state as the baseline, which means S&V was still alive (albeit perhaps in it's dying throes).

I worry for Murray (the Tank's) longevity or DelPo's.

Pancho played well into his 40s, as did Rosewall. Now we have Haas at 35 (with a big hiatus in his career) as the oldest player in the ATP elite (same boat as Agassi). Stepanek is the other player (but he came from a doubles background) and still plays S&V.

Laverfan of course surface speed impacts aces, there are more aces hit at wimbeldon per set or the USO than are hit at RG, and conversely there are more breaks on clay than on grass or on an indoor or faster hardcourt. The idea that surface speed does not impact break percentage and or aces is completely not true. The author by the way compares players who played each other on different surfaces in the same year. So your critique that an Isner type of player would distort the numbers one way or the other is addressed. For example, if Djoko and Nadal play each other on clay and hardcourt. He compares the two players stats against each other and takes 100 or so matchups like that over time in order to get the same server compared against the same returner but on a different surface. In fact the stats you provide that shows Federer hitting twice as many aces on hardcourt than on clay actually supports the writers thesis.

The only person who doesn't show a huge swing between aces on clay, grass, and hard court is Rafter which if you know the player and his serving strategy with how he usually kicked even his first serve in to have time to charge the net you can see why his numbers would be different. But Kuerten and Federer show wild fluctuations with their ace numbers based on the surface.

I'm not sure this debate is really going anywhere, as we seem to be having two separate discussions! Anyway:

1. I'm sure Ancic and Tsonga would love faster courts. However, the original point was that there was a decision of the ATP to slow down hard courts. The focus is on the removal of carpet because that was what the article you linked appears to relate to. Ironically, I read Tiriac as saying courts had to be slowed because physicality would ruin tennis.

2. The fact there are limitations is apparent. However, refusing to engage with the data and instead merely pointing out possible flaws seems a slightly odd approach to a debate.

3. I'm sorry but that isn't an answer. Was Henman a baseliner when he came onto the tour? Edberg? Tennis had changed by the mid 90s. Baseline tennis was the way to go. Players like Roddick or now Raonic had/have sufficient groundstrokes that they don't need to seek to serve volley. The 1-2 punch is as effective (especially on fast courts) and less risky.

4. Well I don't know how many matches he has analysed but the OP certainly appears to suggest he looked at a large number of matches per year.

5. I'm genuinely at a loss to respond to the analysis in your response to Socal. It has no relevance to the original article so far as I can tell. The raw data would seem to be relatively consistent with the OP albeit it is obviously a very small sample set.

Kuerten-Hard at 9.6 aces/match is higher than Federer-Hard at 7.4. Was Kuerten playing on "faster" HC than Federer? (I am less concerned across surfaces, but within the same surfaces).

There is no credit (or debit) given to the receiver for the server's ace count? Why? Same thing for Bps?

I understand the OP used "paired" matches, but even paired matches are matching humans vs mechanical devices (as JJ alludes to human bias).

"Coming in is risky, only if the opponent has the ability to chase the ball indefinitely. "

It's precisely the opposite. Coming into the net shortens the points and makes stamina less relevant. It also makes it far harder for a retriever because they have less time to react. Coming in is risky if you are playing exceptional baseliners - which most of the tour are now.

laverfan wrote:Kuerten-Hard at 9.6 aces/match is higher than Federer-Hard at 7.4. Was Kuerten playing on "faster" HC than Federer? (I am less concerned across surfaces, but within the same surfaces).

There is no credit (or debit) given to the receiver for the server's ace count? Why? Same thing for Bps?

I understand the OP used "paired" matches, but even paired matches are matching humans vs mechanical devices (as JJ alludes to human bias).

But you are comparing two different players! The comparison to the OP is the difference between one player's stats on one surface compared to the other. Socal's post deals with that point.

I'm lost by the mechanical reference completely.

Born Slippy wrote:"Coming in is risky, only if the opponent has the ability to chase the ball indefinitely. "

It's precisely the opposite. Coming into the net shortens the points and makes stamina less relevant. It also makes it far harder for a retriever because they have less time to react. Coming in is risky if you are playing exceptional baseliners - which most of the tour are now.

It is a risk vs. reward issue BS as we are both in agreement here. In the past if you wanted to finish off a point you had to take the risk and move in. Now you don't, due to technology mainly the strings you can just hit the short forehand for a winner and not take the risk. Plus like you pointed out moving in ends the point one way or another being able to retrieve all day long isn't a factor against a players charging in all the time. Instead it becomes a function of being able to hit your spots on the run. Running down everything only is effective against a player who stays back and tries to out grind you or hit winners from the back. S and V players cause themselves and their opponents to play much shorter points.

@BS...

2. The debate may make more sense, if the correlation between surface speeds and aces/bps was clear. My contention is that aces/bps are not the ONLY factors which determine surface speed.

3. Roddick did play S&V. For example Roddick v Ivanisevic, but they preferred to win purely by aces. Hewitt was always a base-liner. Henman v Federer (both S&V) - https://www.youtube.com/watch?v=RPA_0GR9G7Y

5. My data set is not "paired", but over entire careers. It is much more relevant, IMHO, than paired data. What happens if two players did not play each other at all for quite some time?

laverfan wrote:Kuerten-Hard at 9.6 aces/match is higher than Federer-Hard at 7.4. Was Kuerten playing on "faster" HC than Federer? (I am less concerned across surfaces, but within the same surfaces).

There is no credit (or debit) given to the receiver for the server's ace count? Why? Same thing for Bps?

I understand the OP used "paired" matches, but even paired matches are matching humans vs mechanical devices (as JJ alludes to human bias).

Wrong Laverfan, I am sure Kuerten hit more aces on hardcourt than Ferrer or Federer more than chang. That doesn't shed light on anything. There is no matching of a human vs. mechanical device. He isn't measuring differences in radar gun speed. And he is making a credit and a debit for the receiver's quality and impact on the ace count. If Djokovic hits 5 aces against Nadal on clay, then he plays Nadal indoors and hits 10 aces against the same receiver and this pattern repeats itself with 100 different pairs of players over time then one can assume that it is easier to hit aces on an indoor than a clay precisely because the same server is being measured against the same returner with a significant sample size to make up for vagaries of chance.

laverfan wrote:@BS...

2. The debate may make more sense, if the correlation between surface speeds and aces/bps was clear. My contention is that aces/bps are not the ONLY factors which determine surface speed.

3. Roddick did play S&V. For example Roddick v Ivanisevic, but they preferred to win purely by aces. Hewitt was always a base-liner. Henman v Federer (both S&V) - https://www.youtube.com/watch?v=RPA_0GR9G7Y

5. My data set is not "paired", but over entire careers. It is much more relevant, IMHO, than paired data. What happens if two players did not play each other at all for quite some time?

As to your point 2, there is a correlation between aces and service speeds. The writer doesn't deny that surface speed can be analyzed in different ways. But are you really claiming that there is no correlation between the speed of the court and the number of aces someone hits because that isn't the tennis I have watched and played. The fact that other statistics can show service speeds is irrelevant. For example, scientists have never seen a black hole but they know that it is there because of its impact. The same analogy is drawn here, he uses two valid and relevant indicators that are directly related to the speed of the court.

Born Slippy wrote:"Coming in is risky, only if the opponent has the ability to chase the ball indefinitely. "

It's precisely the opposite. Coming into the net shortens the points and makes stamina less relevant. It also makes it far harder for a retriever because they have less time to react. Coming in is risky if you are playing exceptional baseliners - which most of the tour are now.

Running towards the net also requires stamina, that too right after the serve. The current base-liners are aided by technologies.

socal1976 wrote:I am sure Kuerten hit more aces on hardcourt than Ferrer

Now, should we compare Isner vs Kuerten? "Pairing" assumes that the each "human" plays at the same level every day like an automaton over the years. Hence my reference to mechanical vs human.

socal1976 wrote: there is a correlation between aces and service speeds.

I assume you meant surface speeds. Serena can hit an ace in Miami 2014 at 119 miles an hour, but Raonic requires 131 mph on the same surface, so surface speed is not the only factor to serve an ace. "Pairing" is trying to eliminate the returner, but does not account for the "human" factor.

socal1976 wrote:For example, scientists have never seen a black hole but they know that it is there because of its impact. The same analogy is drawn here, he uses two valid and relevant indicators that are directly related to the speed of the court.

The impact of a black hole is due to many factors, like gravity, density, mass. You just keep saying the impact is due to mass alone.

Global warming?

As BS says, the data is fine, but it's correlation to surface speeds is not the only parameter in the equation. For example temperature makes a difference, as does wind, as does humidity.

If Aces = (K) Surface Speed, it assumes a linear (inverse or direct) relationship. It is an approximation, which is why I disagree with the OP.

I had much rather rely on a TD telling me the coefficient of friction and restitution of a surface.