If someone told you about an event where Rafael Nadal crashed out to a non-contender, Milos Raonic made a statement, and the final pitted Tomas Berdych against Roger Federer, you’d be forgiven for assuming the event was played on a very fast court. All of those things happened last week in Madrid on a surface that has at least some things in common with clay.
Given the tournament results, it’s no surprise to discover that statistically, the Madrid courts didn’t play like the old-fashioned red stuff. The stats from this year’s event at Caja Majica are a significant departure from those in past years, and suggest that the blue clay resembles a hard court more than it does European dirt.
Let’s start with aces. Aces are the stat most affected by surface, given the small difference in serve speed and bounce trajectory that can turn a returnable offering into an unreachable one. Of the 29 ATP tournaments played so far this year, Madrid ranks 10th in ace percentage after making adjustments for the players in the field and how many matches each one played. In fact, taking these adjustments into account, the ace rate in Madrid was almost indistinguishable from that of the indoor San Jose tourney!
(For a bit more background on methodology and more tourney-by-tourney comparison, see this article from last September.)
This is a huge departure for Madrid. The tournament has always had a reputation for playing a bit fast, given the altitude compared to Monte Carlo, Barcelona, Rome, and Paris, but that has long been a minor difference, at least when it comes to ace counts. In 2011, Madrid’s ace rate ranked 22nd of the season’s first 29 events, just ahead of Acupulco and behind Munich, Casablanca, and Santiago. 2010 was almost exactly the same, with Madrid coming in 23rd of these 29 events.
Another way of estimating court speed is by looking at the percentage of points won by the server. Even on points where the returner gets the ball back in play, a fast court should generate weaker returns and more third-shot winners. In this department, Madrid once again ranks among this year’s faster events. As in ace rate, it is #10 of 29 on the list, just behind San Jose and ahead of the hard court events in Chennai, Auckland, and Brisbane.
I can’t say whether it’s right or wrong to have a Masters-level event on an unusual surface, but I can say, based on these numbers, that the blue clay hardly plays like clay at all.
Excellent use of stats, however, was the blueness of the clay what caused this? If it were red clay and prepared the same way, would these same numbers have been produced?
I had the time this last week to watch quite a few of the matches, especially on the men’s side. What struck me was not so much the fastness of the surface – though obviously it was quite fast – but its peculiar slipperiness underfoot. Press reports in the last few days seem to suggest that both the fastness & the slipperiness come from the court undersurface having been overly compressed in an attempt to reduce bad bounces. Tirac has apologized and says they’ll try and do better next year – which means we’ll likely see courts that no matter what their color, play more like traditional clay.
I don’t trust Tiriac at all! He wants the Clown Show and not a traditional Tennis Tournament… I would not mind the blue color so much as I do the painful sliding of the players…. heard that Tiriac is not going to to it next year though. Hope Manolo Santana and Charly Moya have more sense!!!
Great analysis. In answer to Mike B. (above), it’s my understanding that the color had nothing to do with the speed – it’s the same crushed stone and the same base, constructed identically to red clay, but simply dyed blue instead of red (the red clay is not brick dust, and the crushed stone it’s made from is white). The courts were simply too new to have allowed the top dressing to bond with the base, and so they were fast and slippery, just as any new clay court would be. The red clay court constructed for the Davis Cup tie in Switzerland in Feb. was a similar disaster.
Yeah, this explanation makes sense to me, about the courts being too new for the base and the dressing to bond. I guess one big difference between the Madrid courts and the notorious Davis Cup court in Switzerland was that they did in fact compress the heck out of the surface at Madrid, to try and avoid the bad bounces – and this probably led to it playing fast. Whereas for the Davis Cup court, it was not so much that the court was fast, but that it had bad bounces galore.
“This is a huge departure for Madrid. The tournament has always had a reputation for playing a bit fast, given the altitude compared to Monte Carlo, Barcelona, Rome, and Paris, but that has long been a minor difference, at least when it comes to ace counts.”
Your words,”long been” have me wondering if your The Speed Of Every Surface* analysis took into account Madrid becoming a clay court tournament in 2009.
Yesterday I was intrigued by your twitter mention of Sao Paolo being the clay court tournament with the 2nd highest ace count in part because last year you wrote, “Surface speed is tricky to measure–as I’ve already mentioned, “surface speed” is really a jumble of many factors, including the court surface, but also heavily influenced by the atmosphere and altitude.”
So I’m wondering how significant a factor altitude is when it comes to these tournaments… a “minor difference” or one by which tennis is “heavily influenced”?
* http://tennisabstract.com/blog/2011/09/13/the-speed-of-every-surface/
With the hard courts and grass courts becoming slower and bouncier, what’s wrong with a clay court becoming faster. I love it. Please don’t change a thing. The sport needs variety. I am so sick and tired of the baseline grinders and their endless rallies on slow courts.
How many matches were too few when looking at a player’s head to head ace count? Obviously you threw out those that only played one or two matches, but did you throw out those with three?
By the way, using head to head ace percentages with 4 matches played and above, I got 1.48 as the average number of aces hit on Queens for 2015, which seems wrong? Have you checked the numbers for 2014?