Peak level of play (Federer, Nadal, Djokovic & Co.)

[krosero]

Originally Posted by Bursztyn

We have been debating whether it is possible for the two players to raise their level of play without changing the AM's.

I would like to present a hypothetical scenario in which two players playing on two occasions on the same surface exhibit different level of play yet their calculated AM’s stay the same.

It is open to debate if such a scenario is realistic.

On the first day they played an average match; winners, forced and unforced errors were counted, and the AM's were calculated.

On the second day they played an excellent match, however the number of winners, forced and unforced errors was the same as on the first day (for the sake of simplicity I assume that the number of points in both matches was the same). How is it possible?

Winners;
The first player was hitting great shots, however the second player was also defending better than previously (had the second player been playing as well as on the first day, the first player would have had more winners than on the first day, had the first player been playing as well as in the first match he would have had less winners than in the first match). As a result the first player had the same number of winners on both occasions, and the same applies to the second player.

Forced errors;
Generally the same logic applies to forced errors. Both players committed the same number of forced errors as during the first match, however the shots that forced these errors were of the higher quality than the shots that forced errors in the first match.

Unforced errors;
Generally the same logic applies to unforced errors. Both players committed the same number of unforced errors as during the first match, however the points which ended in unforced errors were of the higher quality than points which ended in unforced errors in the first match. We have already mentioned that points that end in an unforced error may be of different quality (e.g. long neutral rallies versus double faults or return errors or short neutral rallies).

I think I have to go back to what I said in one of the earlier posts (I forget which one). If a player increases his level of play, it's hard to imagine him not cutting down on his UE's.

With the other kinds of points -- the winners and forced errors -- there's a kind of logic to arguing that the defender, by raising his level, will offset the more successfully aggressive play of his opponent. You give as good as you get.

Not saying I agree that that could happen, but I see where you're coming from.

With the UE's, though, it's different. Those are the kind of errors that have least to do with what your opponent is doing. Those are the kind that you make when your concentration breaks momentarily; or you get physically lazy or something; or you've got a relatively simple shot but because you have a slight flaw in your technique you miss it.

I think we all agree that there are different kinds of UE's, and that sometimes an error is marked down as unforced even when the player commits it after a long, exhausting rally. On those points you could say that your opponent has a lot to do with your error.

But let's leave those ambiguous points aside. I'm talking about the most basic errors in a match: the ones that are entirely due to yourself. Every match has those. No one can play a perfect match.

And if the two players have genuinely raised their level, I can't see how they would not cut down on those types of errors.

I would expect those types of errors to decrease first and foremost, when you're playing well.

Anyway, once the UE's decrease, the AM's increase.

Still can't see how increased level of play can fail to raise AM's -- though it's still an interesting question!

[Bursztyn]

AO 2013, mens singles, first round

Somdev Devvarman versus Bjorn Phau

http://www.australianopen.com/en_AU/...y6/1127ms.html

Devvarman - 54,07%

Phau - 41,86%

[krosero]

Originally Posted by pc1

Krosero,

On the matches that you've charted, did you find that the average AM varied according to the era? For example did you believe 1968 to 1985 had different AMs than 1985 to 2000 and that would differ from 2001 to the present?

Really difficult to say with my limited data.

If I do find a pattern, we can talk about it, but there's none that I've noticed.

The one thing I can say with confidence is that UE counts are all over the place, the farther back you go. (I would guess that the farther back you go, the less agreement there was on how to score UE's). That makes it very suspect to compare official UE counts between eras.

[krosero]

Originally Posted by Bursztyn

AO 2013, mens singles, first round

Somdev Devvarman versus Bjorn Phau

http://www.australianopen.com/en_AU/...y6/1127ms.html

Devvarman - 54,07%

Phau - 41,86%

Those stats are off. The Winners are equivalent to the Aces, and the Unforced Errors are equivalent to the DF's. That's why they've only got 6 winners between them, and 3 UE's.

[krosero]

Originally Posted by corners

Thank you. Thinking of Fed's comment on unforced errors. Is it possible that the same forcing shot hit by player A is an unforced error off the racquet of player B but an forced error off the racquet of player C? In other words, if player B is so skilled at returning forcing shots, is it possible that he is given an unforced error for failing to do so, while player B, not quite so skilled but failing to return the same shot, is given a forced error instead. If so, one would think that the great players, like Fed complained, are scored with more unforced errors than they deserve. But I guess for this discussion it doesn't matter, as we're mostly comparing great players.

Well I remember a statistician years ago saying something similar. It was a long time ago so my memory is vague, but this was the idea: when you see Lendl miss a shot you have to consider whether a player at his level should be making that shot more often than not.

He went on to say it was a highly subjective process.

I think the scoring of UE's could be more objective if it were based as much as possible on physical movements of the players and the ball. I really think that if you start judging an error according to your own expectation of whether this champion should make the shot, that is a sure way to introduce bias into the stats. One statistician who has a high opinion of Lendl will judge the error differently from the statistician who has a lower estimation of Lendl.

Or take it down to the level of strokes. Different people have differing opinions on Federer's BH, and differing opinions on how it compares to his FH. You might expect the Federer FH to stand up to a certain shot, but not expect the same of Federer's BH: because you know beforehand that Federer's FH is stronger.

That's backwards IMO: the errors should be judged objectively, and then the stats can illustrate which stroke (or which player) is better. If you judge each stroke, or player, by preset standards, the stats lose all capacity to say anything.

I think all such opinions about particular players and their particular strokes should be left out of the scoring process as much as possible. IMO it would be far more objective to judge an error based on what's happening with the ball and the player, forgetting what the player's name is (if you can do that). Just judge it on how hard the ball is hit, or how much spin or angle; how far the player had to run to get it; whether one player is at net; etc. Physical facts about the body and the ball.

To the extent that this can be done, the resulting stats would then be able to illustrate which stroke, or even which player, is stronger in this situation, in that situation, etc.

[abmk]

Originally Posted by krosero

1. Federer gained something by being aggressive with Murray's second serve, and the cost was a few more missed serves, because he was going for them. Agreed on that. I think you're saying that when we look at the fact that his Return In % was basically the same as what it was in the semis we should take the missed serves into account. He may have missed them, but he also gained something by attacking the second serve. I could see that. There's a good case his returning quality in the two matches was very similar.

yeah, exactly my point ...

Originally Posted by krosero

2. Not sure I understand you here. Our working assumption is that net play of any kind, not just SV, reduces the chances for UE's because only 1 player (the one rushing the net) can now make an UE; the defender will either make a winner or a forced error.

I don't think you're saying that SV net play results in more UE's than net play behind approaches (or are you?)

Keep in mind we were debating why there were more UE's scored in the final than in the semi. That's the issue I'm talking about.

Are you saying that the baseline rallies in these matches -- the points on which all 4 players stayed back throughout the point -- were different in the final, and more likely to end in UEs, because Federer and Murray were trying to get into net?

Net play does reduce the chance of UEs, but even in this there are variations, SnV reduces it drastically. In case of a point constructed from the baseline and then the net approach made, there is still quite a bit of chance for the players to commit the UE - while the point is being constructed . Therefore it isn't logical to completely exclude those net points either from the semi or the final while computing unforced errors in rallies ...

[krosero]

Originally Posted by abmk

Net play does reduce the chance of UEs, but even in this there are variations, SnV reduces it drastically.

A net rusher makes more UE's if he's following his serve to net rather than following an approach? I doubt it; I'd like to see some stats supporting it.

Originally Posted by abmk

In case of a point constructed from the baseline and then the net approach made, there is still quite a bit of chance for the players to commit the UE - while the point is being constructed . Therefore it isn't logical to completely exclude those net points either from the semi or the final while computing unforced errors in rallies ...

There's some confusion here. I'm not trying to exclude net points from any total count of the UE's. I'm excluding net points from baseline rallies.

You're argument here is completely unclear.

[abmk]

Originally Posted by krosero

A net rusher makes more UE's if he's following his serve to net rather than following an approach? I doubt it; I'd like to see some stats supporting it.

There's some confusion here. I'm not trying to exclude net points from any total count of the UE's. I'm excluding net points from baseline rallies.

You're argument here is completely unclear.

no ...... its the other way around ... I'm saying there is more chance of UE in the 'whole' point when net rusher is following an approach rather than his serve ...please read what I wrote again ...

yeah, I know you were trying to exclude the net points from baseline rallies .....

But like I said, I was talking about errors in rallies, not just @ the baseline ....

my argument was that there were more returns in play in the finals and most of the net points were NOT SnV in the finals, therefore that leaves ample scope for UEs in all those points ... ( including rallies at the baseline and those points that were constructed from the baseline and ended up in a net approach )

[corners]

For no particular purpose:
(The average AMs on the men's side for seven rounds of 2005 AO: Winner 22.5%; Loser 11.0%)




Not sure how much meaning is here, but it appears that Murray, who's performance in the 1st round was lauded by the press, faced weak opposition, while five other players posted 29+ AM's even though their opponents played at a level that was near the 2005 tournament average (10+ AMs). Tomic appears to have played his 1st rounder like an elite player and also to have played at a high level in the 2nd round, defeating Brands who posted a higher AM than Tomic in the match.

Monfils' continues to win matches in novel but enervating ways.

[krosero]

Originally Posted by abmk

no ...... its the other way around ... I'm saying there is more chance of UE in the 'whole' point when net rusher is following an approach rather than his serve ...please read what I wrote again ...

yeah, I know you were trying to exclude the net points from baseline rallies .....

But like I said, I was talking about errors in rallies, not just @ the baseline ....

my argument was that there were more returns in play in the finals and most of the net points were NOT SnV in the finals, therefore that leaves ample scope for UEs in all those points ... ( including rallies at the baseline and those points that were constructed from the baseline and ended up in a net approach )

I see now that we were referring to different things. When I say that net play reduces the chances of an unforced error, I mean that if we looked at the final strokes that ended a net point, we would be less likely to find UE’s there than if we look at the final strokes of baseline rallies. So I presume that if we look at all volleys made at the net, we’re not going to find more volley UE’s behind the serve than we will find volley UE’s behind approaches.

What you meant by chances of an UE is the potential for an UE to occur if you’re just thinking of the two players as they begin their point: if they get into a long baseline rally there’s a great chance that they’ll end up making an unforced error; and there’s also a potential if they’re trying to get to the net, but less so, because those points might be shorter; and there’s the least chance of all for an UE to occur if the server decides right away to follow his serve to net.

It’s an interesting argument, and I see the logic now: Murray, by returning better than Djokovic, put more balls into play -- and once the ball is in play the point might end in an UE (at net or the baseline). Therefore the Murray final had a higher number of UE’s than it would have had if Murray had returned as “badly” as Djokovic, who often could not get the serve back in play. The higher UE’s meant lower AM’s than those of the semifinal.

However I’ve worked out the numbers, and I think Murray’s superiority in the service returns -- if it was responsible for producing any UE’s -- cannot have produced more than 2 or 3 extra UE’s.

The Murray/Fed match needs to drop more than 2 or 3 of its UE’s. It would need to drop 22 of its unforced errors, in order to equal the AM’s of Fed/Djokovic.

Okay, let me say this about your argument. I do not think it can work if we’re talking about Murray returning better than Djokovic because he made fewer UNFORCED errors on the return. I said this above to Burstyn: if a player is cutting down his UE’s on the return and putting more balls into play, he is lowering the UE’s of the match as a whole, and the AM’s will rise. That’s not the problem with Murray/Federer: we’re looking for a way to explain how Murray’s returning might have ADDED to the UE count and lowered the AM’s.

But I can see how your argument might work if Murray’s better returning cut down on the number of FORCED errors. You mentioned that yourself in a post above: if Murray is able to return the kind of powerful serves that forced Djokovic into return errors, then he’s putting the ball back into play without decreasing the # of UE’s in the match: and those extra rallies might end in some UE’s. Thus adding to the UE’s of the match as a whole.

(I don’t think the argument really works for Djokovic/Fed/Murray, but I’d like to apply it to matches between HUGE servers who have a weak overall game and who can be counted on to make unforced errors if you can only get their huge serves back in play. But I’ll leave that for another post.)

I concede that Murray’s strength on the return does happen to be his ability to get serves back in play. Maybe Djokovic has the more aggressive return: but Murray got more of Federer’s serves back in play. We’re talking about big, forcing serves: not the kind of serves on which the receiver can make an UE trying to return it. Murray and Djokovic actually had nearly identical numbers, as far as UE’s on the return (those numbers available on the Wimbledon site). It’s on the big, forcing serves that Murray must have done better than Djokovic.

So in a way this is a perfect test case.

The main problem is that Murray cannot have put nearly enough serves back into play -- not nearly enough to account for the difference in AM’s between the final and the semifinal. He needed to put enough serves back in play for 22 EXTRA unforced errors to appear. And since only a fraction of rallies end in unforced errors, you can imagine that the number of EXTRA rallies that he needed to create would be far more than 22.

In the actual matches, the number of powerful serves that Murray put back in play -- serves of the kind that Djokovic could not return -- was probably somewhere around 10 (I’ll show that below). And only a fraction of those -- if any -- would be likely to end in UE’s.

You can easily find out the number of rallies (baseline or net) in these matches (Total Points minus Unreturned Serves minus DF’s).

And you can find out the number of UE’s that occurred in the rallies (Total UE minus DF’s minus UE’s on the return).

Semifinal -- 133 rallies, of which 6 ended in UE’s (5%)
Final -- 206 rallies, of which 35 ended in UE’s (17%)

So of the extra rallies that Murray created with his return, you’d have to say that only about 17% of them were likely to end in UE’s.

And how many extra rallies did Murray probably create? Well if you lower his rate of “Returns In” to make it exactly equal to Djokovic’s (56%), you’d give Murray 17 more return errors. So is 17 the number of extra rallies that Murray’s quality return created? Not quite, because we agreed that Murray had weaker serves to deal with than Djokovic did (Novak faced higher MPH from Federer). So that’s why I said 10 before. As a compromise. Ten extra rallies produced by the quality of Murray’s returning.

And 17% of 10 extra rallies comes to 2 extra UE’s.

That would increase the combined AM’s of the final by less than 1%.

So I think if the AM’s of the final are lower than those of the semifinal, we’re looking at two possible causes:

1) the level of play in the final was genuinely lower
2) there is a discrepancy in how the UE’s were scored in these two matches
3) the AM method does not describe these two matches very well

(#2 and #3 may be related)

[Moose Malloy]

here were the AM's on Laver-Ashe '69 W SF(using my own stats)

Laver 44%
Ashe 33%

AM's on Sukova-Navratilova '84 AO(again using my own stats, have any women's matches been mentioned yet?)

Navratilova 37%
Sukova 34%

'83 W QF Lendl-Tanner
Lendl 45%
Tanner 36%

[krosero]

No women's matches have been mentioned yet, I don't have any AM's. What do you think might be a candidate for highest, maybe Graf-Zvereva? I couldn't find the UE's for that one.

[Moose Malloy]

Probably a Navratilova match at Wimbledon, do you have any stats on the '86 or '87 finals? low ue's after the 1st set in both.

[krosero]

Federer and Nadal at the year-end championships:

2007 semi in Shanghai
Federer - 31.0%
Nadal - 9.2%

2010 final in London
Federer - 25.4%
Nadal - 11.5%

2011 round-robin in London
Federer - 48.2%
Nadal - 14.8%

Anyone know the unforced errors for their '06 semi?

[krosero]

Federer won the '07 semi 6-4, 6-1, making 81% of his first serves. That's his highest service % in any of their 28 matches to date.

Next highest for him were:
'07 Wimbledon final - 71.2%
'06 Wimbledon final - 69.7%
'08 RG final - 68.8%

In this tournament he also hit his career-high service percentage against Roddick. These are his best service percentages in that rivalry:

’07 Masters Cup round-robin - 82.9%
’09 AO - 65.6%
’09 Wimbledon - 64.4%
’05 Cincy - 64.0%

[corners]

Originally Posted by krosero

No women's matches have been mentioned yet, I don't have any AM's. What do you think might be a candidate for highest, maybe Graf-Zvereva? I couldn't find the UE's for that one.

Serena Williams and Sloane Stephens were both around 34% in their first rounders at AO, both very one-sided matches.

Krosero, I was thinking of posting the top 10 men's AM's for each round of the tournament in this thread, maybe adding them to the 1st round post I made above. But it's a bit off topic. Your thread, your call.

[krosero]

Originally Posted by corners

Serena Williams and Sloane Stephens were both around 34% in their first rounders at AO, both very one-sided matches.

Krosero, I was thinking of posting the top 10 men's AM's for each round of the tournament in this thread, maybe adding them to the 1st round post I made above. But it's a bit off topic. Your thread, your call.

Not off-topic, please post them. We'll see what we can make of them.

Yeah probably in that post above would be best, so they're easy to study all at once.

[corners]

Originally Posted by krosero

Not off-topic, please post them. We'll see what we can make of them.

Yeah probably in that post above would be best, so they're easy to study all at once.

OK. Will do.

[abmk]

krosero,

yes, that was my point .. If you cut down on the no of forced errors by getting those serves back into play, that increases chances for the UEs ..

Coming to the return UEs, I don't know again WTH the statistician was doing ...

I have the stats till the 3rd set

murray only committed 2 UEs on the return ... they have him @ 8 UEs on the return until the 3rd set ...

federer only had about 4-5 UEs on the return ( don't remember exactly and don't have the stats with me right now ) ... they have him at 11 UEs on the return until the 3rd set

Also have a look at the return winners column there : 19 return winners for federer in the whole match ... now, really ?? he had only about 12-15 returns winners vs SnVing sampras in their 5-setter in 2001 (IIRC ) and we know its easier to hit return winners vs SnVers than against those who stay back

they also have murray @ 9 return winners .... till 3rd set, I have them both making only a couple of return winners till the 3rd set ( I haven't noted that stat down, but I recall that very clearly )

also chew on this :

murray had ~77% of returns back into play ... federer made 131 serves out of which 3 were double faults ... so 128 serves in which murray could have a chance to get the ball back into play

128*77/10 ~ 99

so he missed returning 29 of the serves from federer
the ace count is 12

so that leaves room for 17 (return errors+service winners ) ... if he had 9 unforced errors on the return, that leaves room for only 8 (forced return errors+service winners ) ... now really ?

-----------

now coming back to the total unforced error count, the unforced error count till 3 sets in the official stats is fairly close to mine unlike in the semi where there is more discrepancy .. I will post the full stats once I'm done with the final set ...

[abmk]

Originally Posted by krosero

Federer and Nadal at the year-end championships:

2007 semi in Shanghai
Federer - 31.0%
Nadal - 9.2%

2010 final in London
Federer - 27.5%
Nadal - 4.2%

2011 round-robin in London
Federer - 48.2%
Nadal - 14.8%

Anyone know the unforced errors for their '06 semi?

AM of 4.2 for nadal in the 2010 YEC final ? eh ?

I have the screenshot for the stats here :

http://tt.tennis-warehouse.com/showp...postcount=1451

22 UEs for fed and 19 for nadal

total of 130 points (ATP )

that computes to :

federer : 25.4%
nadal : 11.5%