Fed/Nadal rd 3

[airforce1]

What a super article. It is an amazing level of tennis that these 2 are giving us. The article seemed to support my idea that Federer is the superior attack player, while Nadal's shot tolerance seems to be a little stronger. Along these lines, Federer should do better on grass and Nadal on clay, but with the differences on the more nuetral surfaces playing out to exciting finishes.

Do you have any data on each of their shot tolerances?

[lukman41985]

Highlights: http://www.youtube.com/watch?v=KFL0gzhlfGg

[johnyandell]

The way to do it would be to just count the number of balls per point by each player over the 3 matches... something I'd like to do in my spare time--oh wait I don't have any spare time.


But it's a worthy question! Anyone out there have the 3 matches on tape and want to take it on???

[jhm36]

Ha, comparing shot tolerances of Federer-Nadal matches will give you pretty much the number of strokes per rally that Nadal can send his forehand to the Federer backhand before the point ends. You wouldn't get decent 'shot tolerence' figures from such number crunching of their head-to-head matches, but 'Nadal forehand to the Federer backhand tolerances.'

I tell you what would be interesting, comparing the shot tolerances from the Blake-Nadal match at the Masters Cup with the Federer-Nadal one. I wonder if we'd see shorter rallies from Blake-Nadal and longer ones with Federer.

That Federer topspin backhand held up probably as well as I've ever seen it against Nadal's forehand. I thought it was interesting the way Federer wasn't trying for that many winners with his backhand and was happy just driving it back to Nadal and keeping the rally going. Only one backhand winner for the entire match for Federer, and that coming late in the last game of the match. Very interesting match for a number of reasons.

Federer, the dude who hates the drop shot, must have found it particularly pleasing to win the match hitting a crazy winner off a drop shot.

And speaking of tolerances, man, Federer's ego must have gotten a boost out of breaking Nadal when he was serving to stay in the first set and then when serving to stay in the match. Talk about feeling like you've got a higher tolerance for match pressure than the guy across the net from you. Positives all over the place for Federer in this match. Straight sets. Playing ahead all the way. Playing ahead all the way in his last two matches against Nadal, losing only one set in six. And in those last two matches Fed's only faced, get this, five break points, while he's held twenty-one against Nadal. One guy is putting a whole lot more pressure on the other.

[johnyandell]

So I gather you aren't going to count them.

[jhm36]

Well, I guess I could count a set from each match. The thing is, I know I won't just do a simple shot count, but I'll start dividing the rallies up into serves, aces, service winners, forced errors, unforced errors, forehand winners, backhand winners, overhead, vollies, slices versus topspins, down the line versus cross court, forehand to backhand, backhand to forehand, etc, etc.

No promises, but yeah, I'll try to do a set from Blake-Nadal and one from Federer-Nadal.

[shootermcmarc0]

But it's a worthy question! Anyone out there have the 3 matches on tape and want to take it on???

hmm...i still have the 06 french and wimby and masters cup final on my tivo....

[johnyandell]

So just count it up! We'll post the results. You just have to watch the match once more is all...

[jhm36]

Alright, I'll count, I'll count. First set of Blake-Nadal and first set of Federer-Nadal. I'll even watch them in double-time to keep myself on track and not trying to write down the specifics of the storkes. Just serve, tick, tick, tick, tick, end of point.

I might have to break the figures down so that we can see the difference between real rallies and points that are only aces or service winners.

Not quite the discussion place for it, but what a draw Federer had in the Master's Cup. Nalbandian, Roddick, Ljubicic, Nadal, Blake. Doesn't get much harder than that. Also, Federer seems to play more top-ten players than anyone else over the last few years. Could that be helping his game improve? Or is it wearing him out? Did Nadal maybe lose a bit of his edge this year by not playing as many top level players as Federer, a partial reason for his fade?

I guess to be really scientific about this shot tolerance thing, I'd need to count a set from the Federer-Blake match as well!

[airforce1]

shot tolerance

Both you guys make a bunch of great points about these matches. They give us plenty to think and talk about.

If I understand shot tolerance though, you won't get the correct number from just any match, but must count during a match where the player is forced to hit a lot of balls to win a point. And shot tolerance is not an average is it, because that would make Andy Roddick's numbers useless? The short points would bring down the avg. so far that the number would be useless to get an idea of how many balls you must hit to break him down.

It seems you would have to watch only the points ending in an unforced error, where the error was not in an attacking position, because what we are looking for is a player who is deep into the point and starting to get desparate.

In summary, I don't think Blake found Nadal's shot tolerance very often! (but I didn't see the match)

[johnyandell]

Shot Tolerance is more impressionistic than that. More matches is better, but no just count. It's a range from zero to whatever. According to Elliot over time there are clear patterns but they also vary with the court and the opponent so those 3 matches would still be a place to start.

Jim, you'd be better off picking the Blake match and counting the whole thing.

[jhm36]

Hmmm, you know, the more we talk about it, the less and less I understand what shot tolerance is! I've been working a long post that documents my own confussion, but I'll spare you the pain of wading through that. But, if anybody has a clear idea of what we're talking about please write in.

Heck, I no longer know what I'm suppossed to be counting in the matches! Is it average number of shots per winning point per player? Do we include aces and unreturned serves? What about double-faults? Should we take into account first serve percentage in some way since it's easier to win a point on a first serve than on a second? (Heck, is there a difference in lenght of points between first and second serves? I'd imagine second serve points last longer than first service ones.)

At one point in the essay Teltscher says "How many balls can you hit deep and crosscourt? That's your shot tolerance." Well, if that's the case, then there's no way to determine that sort of thing by watching a match. It's the sort of thing you'd determine in practice, because people aren't going to stand in the corners and hit back crosscourt to each other in matches. Why bother to try to approximate this figure by watching pros play matches? And the way it's formulated here doesn't take into account the speed and spin of the ball coming at you. It's not really how many balls can I hit deep a crosscourt that matters, is it? That's too vague, not specific enough. What we want to know, isn't it, is at what speed and what rate of spin do balls sent my way start breaking down my own ability to hit deep and crosscourt? Once we determine that we can start buidling up my tolerance by jacking up the speed and spin to an appropriate amount that allows me to get better without blowing me off the court.


By the way, the numbers for the Agassi-Johansson match mentioned are off by a lot, I think. More on that later.

[johnyandell]

You are over thinking this tremendously. What I am suggesting is something far more simple as a starting point. Every point how many balls does each player hit? That's it.

Two columns, two players, each line a point with how many balls.

[jhm36]

HA, no doubt I'm an 'over-thinker' to put it nicely.(Here I should confess took a number of courses in philosophy in college and I'm married to an ex college philosopher teacher.)

Thanks for the simplification, but at heart I don't see what those numbers are really going to significantly tell us, and how they are going to fit with the basic defintion of shot tolerance being the number of consequative balls that can be hit deep and crosscourt. Rally length and shot tolerance seem like pretty different things to me.

How about a specific question: how do you score a double-fault? Technically no ball was ever put into play. Did the server hit one ball, or two, or none? How does double-faulting figure into shot tolerance at all? Maybe we should just throw out double faults from consideration?

See what I mean. Philosophical questioning. It never ends.

[johnyandell]

So my advice is don't do it. Anybody else up for it?