Peter Bodo's TennisWorld - Talking Head-to-Heads
Home       About Peter Bodo       Contact        RSS       Follow on Twitter Categories       Archive
Talking Head-to-Heads 01/13/2008 - 11:00 AM

Wimbledon01

By Rosangel Valenti, TW Contributing Editor

Some months back, I wrote a piece about the head-to-head records between the (then) top 20 ATP players and the rest of the top 20, as well as their head-to-heads against the Top 50 players. "Oh no," I hear some of you groan. "Not another piece about Federer, Nadal and Djokovic, and the unfulfilled potential of David Nalbandian." Well, no. The WTA has, statistically speaking, received zero love around here recently, so it seems like a good time to take a look at the comparable set of statistics for the women. Thus, I've run exactly the same head-to-head exercise for the WTA as was done earlier for the ATP, on data that was current up until the end of last week.

The purpose of looking at head-to-head records is to see whether a given player has an ability to beat the top players - as this is mandatory for achieving success in Slams and other big events. In some ways, the exercise is designed to illuminate the giant-killing potential of some up-and-coming players. Though of course, tennis is all about matchups, and, as the statistics unarguably show, whatever record any player may possess against the rest of the field, Justine Henin is one giant who currently has a stunningly good record against all challengers. In fact, there is only one player in the current top Top 50 with a significant winning record against her - step forward, Venus Williams, with a 7-2 head-to-head in her favour. The only other Top 50 player with a winning record against Henin is Lucie Safarova, at 1-0. What's the betting that after the next time they play, it will be 1-1?

Simply put, I have looked at each player's head-to-head record, in aggregate, against the current Top 5, Top 10, and Top 20 (to be clear: that means that the Top 5 are being assessed against only four other players, as opposed to five for the rest of the Top 20, whie the Top 10 are being assessed against nine other players, rather than ten). For comparison I've looked at the records of comeback queen Lindsay Davenport,  as well as the recently retired Martina Hingis and Kim Clijsters.

Top20e

From this table, I made the following key observations:

- Among the Top 20 players, there are five with a 50% or better winning record against the current Top 5 - Henin, Ivanovic, Venus and Serena Williams, and Amelie Mauresmo, with Henin's record at a soaring 89%. When I calculated the same numbers for the men last April, Roger Federer's winning percentage against the Top 5 was "only" 83%, which I thought was amazing in its own right. Kim Clijsters also possesses a winning record against the current Top 5.

- The percentage of wins against the Top 5, for those exceeeding 50%, is generally higher than for the men. Thus, while Rafael Nadal had a 56% winning record against the Top 5, second to Federer, in the women's case, Ana Ivanovic, Amelie Mauresmo, Serena Williams and Venus Williams all topped Nadal's showing (as, historically, did Kim Clijsters). Three other Top 20 men other than Nadal (Hewitt, Murray and Nalbandian) had a greater than 50% winning record against the Top 5 - by just a few percent. Gaudio, Henman and Canas could also claim to be part of that elite. In general, we appear to have a situation where, in spite of having a dominant player in Justine Henin, the elite players below her - especially Venus Williams, on 64%, also have very strong records against the Top 5 (well, in Venus' case, there's that head-to-head against Justine).

- Among the Top 20, there are six players with a winning record against the Top 10 - the same five mentioned above, plus Sharapova. Clijsters and Davenport are also part of this elite club.

- The only change when we look at the records of the Top 20 against the Top 20 is that Nicole Vaidisova has a 50.9% winning percentage, while current number two Svetlana Kuznetsova is exactly 50/50, and Agnes Szavay, on the basis of only six matches, also stands at 50%.

- For me, perhaps the most interesting statistics are those applying to Kuznetsova and Jankovic. Apart from the 14% recorded by Sybille Bammer, world No. 3, Jelena Jankovic, at 18% wins, has the worst record against the Top 5 of any Top 20 player. Kuznetsova, world No. 2, is the next worst, on 24%. Jankovic in particular has a disappointing record against the Top 10 as well. Is she turning into the female equivalent of Nokolay Davydenko, who also has a notably less-than-stellar record against the elite players? Of course, part of the story is her 0-9 record against Henin, but the highest-ranked player over whom she has a winning record, 3-2, is Elena Dementieva at No. 11. Probably her best head-to-head records are 2-2 against Serena Williams, and 3-3 against Venus Williams. But she's 3-3 with Marion Bartoli, and has a losing record against the likes of Nicole Vaidisova and Anna Chakvetadze.

- Kuznetsova, like Jankovic, has played Henin a lot - her losing record is 2-15, which accounts for part of her losing record against the top 5, but not all. She and Jankovic are 2-2 against each other, but otherwise have losing records against the rest of the Top 5. Ivanovic, meanwhile, has a 0-3 losing record against Henin (0-4 after the current week), but more than compensates with a 4-1 record over both Jankovic and Kuznetsova. She has yet to record a win over either Williams. Perhaps, then, part of her apparent better performance, versus that of her fellow Serb, is due to her fairly impressive winning record against most of the rest of the Top 20.

In the second table, I've also looked at the records of the current top eight players against the Top 50. And, as for the men, I calculated how many players they have played, but never lost to. Roger Federer was the only ATP player who had established "complete dominance" over more than 50% of the field (58%), with Nadal at 48% and (even back in April last year, before some of his later heroics) Djokovic at 45%.

Top8fg

Interesting, then, that while Justine Henin has a similar "complete dominance" percentage (i.e. has never been beaten by that player) to Federer's, at 56%, Lindsay Davenport is well ahead of her, at 66%. She hasn't played against as many of the current Top 50 as has Henin, but has a strong record against all that she has played, including a 14-13 head-to-head against Venus Williams. With the exceptions of Serena Williams (who's at 10-4), Maria Sharapova (4-1) and Justine Henin (7-5), Davenport has a winning record against all other current top 50 players that she's played.

Regarding "complete dominance" over other top 50 players - only Maria Sharapova and the Williams sisters can rival Henin and Davenport. Looking more closely at these dominant women in turn:

- Maria Shrarapova has a losing head-to-head against Henin (2-6), Serena Williams (2-4), and Amelie Mauresmo (1-3) while being 3-2 against Venus Williams. In fact, only one other Top 50 player, Victoria Azarenka, has a winning record (1-0) against her. As mentioned above, she's 4-1 against Lindsay Davenport.

- Only two top 50 players have a winning record (1-0 apiece) against Serena Williams - Anna Chakvetadze and Sybille Bammer. She's tied with Henin (6-6), Venus Williams (7-7) and Jankovic (2-2).

- With the exception of Agnieszka Radwanska, who's 1-0 against her, Tatiana Golovin, who's at 2-0, and Lindsay Davenport, against whom she's 13-14, no top 50 players have a winning record against Venus Williams. Incidentally, I watched the match against Radwanska last year, and Venus completed it despite a very obvious wrist injury.

For what it's worth, Amelie Mauresmo, while not as dominant as the women mentioned above, has losing records against only seven Top 50 players - Henin (6-8), Serena Williams (2-9), Venus Williams (3-5), Vaidisova (2-3), Alona Bondarenko (0-1), Lucie Safarova (0-2) and Peng Shuai (0-1). I'm seeing the current cutoff between "Tier 1" and "Tier 2" for the women as coming around here, with Mauresmo in Tier 2. This is debatable - but then, there are those head-to-heads against the more dominant women.

Among others, Anna Chakvetadze has yet to make much impact regarding those significant head-to-heads - she's 1-0 against Serena Williams, and 1-2 against Venus Williams - but otherwise has been dominated by the dominant women. Ana Ivanovic has a 2-2 head-to-head with Maria Sharapova, has never played Davenport, and has no wins against the other three dominants. Jelena Jankovic, as it turns out, can at least say that she has even head-to-heads with both Venus Williams (3-3) and Serena Williams (2-2). But she's 0-9 against Henin, 1-3 against Sharapova, and 1-4 against Davenport. Current world number 2, Svetlana Kuznetsova, is 3-3 with Venus Williams, but has a losing head-to-head against the four other dominant players. None of these players belongs in Tier 1 - the question is really, which ones, if any belong in Tier 2? What does the Tribe think? A working definition of Tier 1 for me would be something like "current Grand Slam contenders, with Slam titles already in the bag". Tier 2 would be something like "possible future Slam contenders, or players with Slam titles who don't look like repeating it."

Finally, I also felt that it would be useful to review career match-winning percentages. Henin has dominated more recently, agaist the current field, but this doesn't necessarily reflect the situation in earlier years, against a somewhat different field. In fact, Henin, Sharapova and the Williams sisters all have 81-82% career-winning percentages, as of now.

By way of a benchmark, I thought that it would be useful to throw in the career-winning percentages of some great female champions of earlier years. Steffi Graf won 89% of her 1,017 career matches; Monica Seles 83% of 717 matches; Chris Evert 90% of 1455 matches; Martina Navratilova 87% of 1653 matches; Billie Jean King 82% of 850 matches; and Evonne Goolagong Cawley 81% of 869 matches. Martina Hingis, incidentally, won 80% of her 681 matches.

Looking at the current percentages versus the higher ones of the past, I was tempted to think that the fact that there is more than one sizeable rivalry in the women's game today may have something to do with the difference. However, not all the top female players play each other particularly often. For example, the Williams sisters have a 7-7 head-to-head, but the last time they met in a tournament was at the US Open in 2005. And, until the 2007 US Open, it was more than four years since Justine Henin had played Venus Williams - so, as Justine won, how much does that 7-2 head-to-head in Venus' favour really mean today? Until 2007 (Key Biscayne) it was also nearly four years since Henin had played Serena Williams, and with three wins to one loss in her favour since then, she has now evened out their head-to-head at 6-6. Meanwhile, Lindsay Davenport's 14-13 head-to-head with Venus Williams is the most prolific one I came across during this exercise, but they haven't met since the 2005 Wimbledon final. Clearly, motherhood, injuries and other matters have kept some women off the tour at different times, but it would be good to see some of these rivalries being more active. Other prolific rivalries of the recent past, by the way, included Henin's with Kim Clijsters, at 12-10, and the 11-10 rivalry between Martina Hingis and Venus Williams. Most others currently contain far less matches.

Finally, I've checked out the current career-winning percentages for the top men - Federer stands at 81% of 685 matches, Nadal at 79% of 324 matches, Djokovic at 70% of 172 matches, Roddick at 76% of 540 matches, Davydenko at 58% of 465 matches, and Ferrer at 59% of 346 matches. All that this tells us is that the dominant women generally seem to have higher career match-winning percentages than the men, and find it easier to dominate lesser rivals.

In fact, comparing the men with the women, the statistics pretty much bear out what is also intuitive - that there is a group of elite women, who are very difficult for the rest of the field to beat, who do, however, have the ability to beat each other, even though one is currently dominant. Whilst, among the men, complete domination of one or more other players is harder to do, with relatively little parity existing at the top of the game.

[Note - it's tough to display tables this complex in the format we use here - so I have included two popup windows. This link  should give you a larger version of the first table, and this is a larger version of the second.]


46
Comments
Feed You can follow this conversation by subscribing to the comment feed for this post.

Posted by Anna 01/13/2008 at 10:14 AM

Am I wrong or Amelie actually has winning record against Maria? 3-1

Posted by Schwab 01/13/2008 at 10:15 AM

Rosangel,
You have done it again. Great!!! Not surprised on Jelena record against the Top 5 which says to me that the top players stay away from her go to shot(BH down the line) on crucial points.
Also, your second table have validated a point in my opinion which there is the top level(consist of Justine, Maria, Venus, and Serena)and the best of the rest. If Davenport does well at AO and beat Maria along the way to a QF (or better finish), she will be at the top of the 2nd level. Reason LD stays at the 2nd level is she has not won a Slam since 2000 AO even though she made 2 finals in 2005 only to lose in some good matches to Serena and Venus.

Posted by Sam 01/13/2008 at 10:27 AM

Ros: Great work, and a meaty post for stats geeks like myself. ;-)

To answer your question, I think that Jankovic and Kuznetsova belong in Tier 2.

Did you look at the H2Hs at Slams? For example, Henin-Clijsters was close overall at 12-10, but Henin was 5-2 when they met in Slams (lost the first 2), and lost just 1 set in those 5 wins.


Posted by abbey 01/13/2008 at 10:31 AM

the first thing that came to mind when i first saw your table was, "oh no, is jankovic the female davydenko?" :)

thanks for doing this for the women. they often get pushed aside in the disucssions.

Posted by Rosangel 01/13/2008 at 10:42 AM

Corrected, Anna.

Posted by Sam 01/13/2008 at 10:43 AM

Schwab: I'm not surprised at Jankovic's record against the top players either.

Posted by Scott 01/13/2008 at 10:58 AM

Test

Posted by Rosangel 01/13/2008 at 11:02 AM

Sam: I did not look statistically at the H2H in Slams - just stuck to the basics here. It's my first real attempt at compiling WTA stats on any scale, so I tried to keep it simple. But of course, the Slam H2Hs are well worth discussing.

Posted by Papo 01/13/2008 at 11:16 AM

Core Tennis should try to hire you, Rosangel.

Posted by GSte 01/13/2008 at 11:31 AM

Ros, great piece! This statistical analysis approach is a really good way to determine who the very best of the best players are.

What's so interesting about the women's game is that so many of the top players are young. Sharapova (like Nadal) has obviously already "made it", but we still need to wait on the likes of Ivanovic, Chakvetadze, Vaidisova, Golovin and Szavay (who are similar to Djokovic, Gasquet, Murray etc). I'd be very interested to see what your ATP and WTA charts look like in 1-2 years.

To answer your question, I agree completely with your definitions of "tier 1" and "tier 2".

Thus far, I would say 1-time slam winner Kuznetsova, Jankovic and Ivanovic are tier 2 players. Besides the tier 1 group, they have had the best slam results. Chakvetadze and Vaidisova have started going deep in slams, but they need to win some bigger titles first and score more wins over the top players before they could be called tier 2. To me, the biggest "question mark" top 20 players are Golovin and Szavay. They are both relatively new in the top 20, so we'll see if they go up, down or stay the same.

Posted by magritte 01/13/2008 at 11:34 AM

Some very interesting stats there. It certainly illustrates why we tend to ignore the seedings and look down the list for where the Williams sisters are rather than worry about Jankovic and the Kooz. The top winning percentages do illustrate that the women's field is quite a bit deeper than it was in the eighties, though.

As far as comparing the women to the men: I really think these days the reason the top players (excl Federer) still are less threatened by weaker players on the women's side is mostly because of the dominance of the serve in the men's game. Even a much weaker player will win his serve most of the time and can sometimes get into a tiebreak and--with a little luck--can take a tiebreak against greatly superior opposition. Isner vs Federer in the US Open last year is a case in point.

Posted by creig bryan 01/13/2008 at 11:37 AM

Rosangel:

Thank you for taking the time to sift through these numbers.
Just yesterday, having printed out the OZ draw sheets, I was discussing the various matchups with my SO. The subject of Davenport vs. Sharapova drew us into a heated discussion of the half-life of of a win (or loss). The question I posed was: Should a rolling time window be imposed on any given H2H, similar to that of the current points system? And if so, what length of window would be appropriate? 2 years? 3?

In this context, reading your comments about Davenport and Henin this morning only served to re-ignite the concept (and the conversation). Yes, it opens an entirely new set of stipulation possibilities, but how else could you shave away the anomalies, while maintaining the integrity of the numbers?

I seem to remember Martina N. having a losing record against either King or Evert (I'm not sure) early in her career, and trivializing it later, when she became dominant.

To wit, I can envision a future entry from Mrs. Stevens:

"Yeah, but Venus beat her before she became the queen, so that doesn't count."

Keep Smiling

Posted by Yummy Prince Fed/Karen (completely upset by Dish Network) 01/13/2008 at 11:55 AM

Ros, excellent post. Very fair and well balanced. It is instructive that whilst there is always talk that there is no depth in the women's tour, as can clearly be seen from the stats, not only is there depth, but parity with the men's tour as well. Excellent

Posted by embug 01/13/2008 at 12:07 PM

Thanks for this Roseangel. Glad to see reporting for the women. Really great work on your part!

What would the stats have said had Serena played consistently??? Had Seles not been stabbed? Had I been someone else with someone else's place in history... WHAT!!! :)

Posted by Sam 01/13/2008 at 12:09 PM

GSte: Good point about some of the top players being quite young. I think Ivanovic will be the first of the young group to win a Slam, though I don't see that happening until 2009. I'm curious to see how she fares at the Slams this year.

Posted by Mr. Sherlock (the best detective in a weak era) 01/13/2008 at 12:11 PM

Ros, fantastic post. Wow. I could read over this a few dozen times and get something new each time.

Creig, great point. Kind of like Nalby and Fed, no? Didn't Nalby do well against him early, but since Fed has owned him?

Posted by Sam 01/13/2008 at 12:13 PM

Sherlock: Correct about Nalby and Fed. After Fed morphed into TMF, he has dominated the H2H.

Posted by Ruth 01/13/2008 at 12:14 PM

Thanks, Ros, for these figures and the analyses. It was good to see the women's game receive the same detailed analysis that we've often seen for the men. You covered all the bases, including anticipating the thoughts of a Mrs. Stevens re: the timimg or historical relevance of the Venus-Justine H2H.:) As I was reading your report, I wondered how soon someone would point that out.

As long as players keep playing past their prime or their glory days, we can expect to see (in most cases) a lowering of their winning averages against the top players. To keep their winning %'s very high, for example, Venus would have had to retire a couple years ago and Justine (who started playing 5 years!! after Venus did) would have to retire now!

Posted by Mr. Sherlock (the best detective in a weak era) 01/13/2008 at 12:21 PM

So embug, who would you like to be in this alternate universe? :)

Posted by jbradhunter 01/13/2008 at 12:26 PM

Rosangel- great piece... and any mention of Steffi and her exemplary winning % is fine by me :)

Justine is head and shoulders above the rest- the Williams' comparatively low %'s just reiterate that these girls show up for GS tourneys foremost, then have a non GS good tourney win every now and then-- I remember Richard wouldn't let them play the competitve junior events either- brilliance they name is individuality

Posted by Sam 01/13/2008 at 12:30 PM

"the Williams' comparatively low %'s just reiterate that these girls show up for GS tourneys foremost"

jbrad: That was part of why I brought up the Slam H2H - it would take into account which players were better on the big stage. But of course that would make Ros' work harder! ;-)

Posted by jbradhunter 01/13/2008 at 12:34 PM

Sam- that would be nice to see as well- GS H2H... speaking of stats and piggybacking Ruth a bit (after inexplicably piggybackig Sam :)) the historical WTA stats are harder to track down than the ATP stats it seems... maybe the ladies deserve their due some more! Go Ladies!

Posted by Sam 01/13/2008 at 12:36 PM

jbrad: You're right - for example, the ATP site is better for finding information than the WTA site.

Ruth: Good point about the lowering of winning % as players play past their peak years.

Posted by IF 01/13/2008 at 12:58 PM

Isn't this more than a little misleading? To the extent that certain players are only beginning to peak, their averages are diluted by the year/years took to work their way into the top. For example, Ivanovic has only one year of elite (top-5) play to balance against 1 year of top-100 and two years of top-20, while Sharapova has only one-year of top-50 to balance against 4 years of top-5. On might then think that the stats for Sharapova are a truer measure of her ability (as the ramping up phase is not diluting her championship phase), than the stats for Ivanovic (too much dilution) or (heaven help her) Jankovic who has spent 80% of her career out of the top-10.

The fact the Sharapova has spent, around, 75% of her career in the top ten, while Jankovic has spent a similar amount of time out of it, certainly means something; but, what it can not tell us (and in fact may mislead us) is how good each player is/will be starting at 7PM EST.

In conclusion, I think your methodology is an excellent measure for Henin, Sharapova, Kuznetsova, and the Williams', but no so much for Jankovic and Ivanovic. But then, I'm a shameless serb-lover.

Great work.

Posted by creig bryan 01/13/2008 at 01:01 PM

IF:

So that would make you subSerbient?

Keep Smiling

Posted by IF 01/13/2008 at 01:06 PM

Among other things, yes.

Posted by jb 01/13/2008 at 01:07 PM

Ros - this is great to be able to look at all this - just what we need to chew on a Sunday afternoon!

Interesting Craig Bryan - it does seem to be there should be some rolling cutoff on h2h's, as for players the same age, their matches from say - 6 years ago somewhat lose their relevence. Except of course it can be interesting to see which players turn them around.

Mhm - maybe they should be sliced even further. so you can see h2hs changing. Like 1st half is this, 2nd have is that. OK - this may be too geeky even for me... (But it would be interesting... !)

Posted by pogiako 01/13/2008 at 01:07 PM

Based on that stat, I realized that Roddick (76%) is still up there w/ the big guns. Well, he could maintain that if he forgets about Roger and concentrate more on the other guys. Before he gets to Roger he has to beat Rafa, Novak and the rest of the contenders like Richard. For now, he has to look from afar in Roger's world......he..he..he..he..he..he..he..he..he..he

In Women's side, you really don't have to be a giant like Venus to be a great champion. Look at the picture, a dimunetive Justine strolling next to a troll like player Venus. More power to you Justine!!!!!! May you have more Grand Slams to come but plz stop at 14 coz Steffi(22 gs) is still my fave among the women!!!!!!

To Rosangel: TY 4 d info. It helps me in gaging the abilities of other players...

Posted by laffytaffy 01/13/2008 at 01:23 PM

Great post Rosangel!
I like to point out though that in the table of the top 8, the losing H2H of L-Dav is 2, but you mention 3 in your post.

Posted by Ruth 01/13/2008 at 01:31 PM

IF: I don't think that the stats are "misleading" at all, not if any reader takes the time to figure in the "Yr turned pro" dates that Ros was careful to include in the first table. Of course, if someone wanted to make all kinds of sweeping conclusions from a table without considering the number and timing of years "on the job," one would have to include only players who all turned pro in the same year or, at most, one year apart. and that would be rather limiting -- and not as much fun.:)

Posted by Rosangel 01/13/2008 at 01:50 PM

Re Davenport - the losing H2H is 3. I will correct the table for this a bit later (will see if there are any other inconsistencies that come up).

BTW, the Mauresmo/Sharapova numbers in the tables (as per anna's query above) are correct - it was the wording that required a tweak.

IF: at least the numbers already show Ivanovic doing pretty well. And I still think that beating the top players is mandatory for winning a Slam.

BTW, one thing that was interesting, when we compare the men to the women, is that the winning percentages of the players around 11-20 against the top 5 and top 10 were almost identical, at around an average 30% vs. the top 5, and 35% versus the top 10. Does that say anything about the respective depths of field, I wonder?

Posted by Ruth 01/13/2008 at 01:51 PM

Incidentally, this conversation reminds me about when, as a devout Seles fan, I was trying to find a silver lining in what was appearing to be her early and permanent departure from the game after the stabbing: I figured that, at least, she'd retire with the most amazing winning percentage when compared with players who peaked and then played several more years when they lost more often.

But I was as happy that she was able to come back and play (even though her winning % fell) as I am that Venus and Serena and Lindsay are continuing to play-- and giving us occasional flashes of brilliance -- even though they are no longer winning "everything in sight," as some fans seem to demand and expect that they would.

Posted by Anna 01/13/2008 at 02:02 PM

Rosangel >> oh so my memory served me well ))
It was really interesting to read...even though I don't always believe in statistics

Posted by Rosangel 01/13/2008 at 02:07 PM

Just to follow up on that last post, the average "complete dominance"of the top 8 men (last April) versus the top 8 women here was 34% versus 48% for the women. The figures sort of suggest decent depth up to the low 20s for the women, and less ability to beat the top players if you're ranked from the mid-20s downwards, compared to the men. But that isn't all that surprising.

Posted by IF 01/13/2008 at 02:18 PM

Ruth: Oh sure, I didn't mean to imply that Ros was sloppy or trying to deceive or anything untoward. I merely meant to highlight that the manner in which the data is sliced (that is, that it is unsliced) creates a negative impression of one of my favorites (Jankovic) and that a more nuanced slice is not quite so negative. In other words that, just as results decrease after a certain point, they also decrease prior to a certain point as well. In any event, if the first sentence was too provocative, then I am happy to withdraw it. Time slices (as Creig suggests) or weighting results seems the way to go; but, as I am too lazy to do that work, I consider the "raw" data offered by Ros as being exceptionally useful and a great asset, and so on and so forth with the positive appellations.

On the other hand, this sentence, "Is she turning into the female equivalent of Nokolay Davydenko, who also has a notably less-than-stellar record against the elite players?" struck me as a bit provocative as well. JJ just isn't the forgone conclusion that Davydenko tends to be.

Finally, and not for nothing, the year turned pro figure doesn't quite do the work that I am looking for. For example, Sharapova, Kuz, and JJ all turned pro w/in one year of each other; the stats that Ros gives, I would argue, give a good picture of Maria and Kuz, but not for JJ.

I suspect I am just arguing to my preferred conclusion, though.

Posted by Rosangel 01/13/2008 at 02:26 PM

IF: The time slices might be a lifetime's work:). Raw data is just - raw data.

BTW, I really like Jelena Jankovic as a personalty (and her game), so wish her well this year - hope the injury wasn't serious.

Posted by slowpoke 01/13/2008 at 02:33 PM

Great post Rosangel!
I like to point out though that in the table of the top 8, the losing H2H of Vee is 3, but you mention 2 in your post. Which is accurate?

Posted by Bob 01/13/2008 at 03:56 PM

Thanks for this interesting post, Roseangel. I was comparing H2H on Steve's blog for Henin, since she won her first slam in 2003, and she is very dominant since then, being 6-0 against Lindsay, 4-2 against Serena, 1-0 against Venus, 9-0 against JJ, 4-0 against Ana, 3-0 against Anna, 6-2 against Maria, 4-0 against Vaidisova, and 16-2 against Kuz. While doing that, I noticed that Lindsay was 14-13 with Venus, and also noticed that Hingis was 6-7 against Serena, and 11-10 against Venus, and 11-14 against Lindsay.

Henin lost to Davenport, Venus and Serena many times when she was a teenager, but since she won her first slam (reached her peak), she's been very dominant. It shows that Hingis was very competitive with Lindsay, Venus, and Serena, despite her small size and lack of power, when she was healthy. Hingis wsa 2-0 against Henin when Justine was 17, but 0-2 in their next two matches, in 2006. Hingis was 4-1 against Kim when Kim was 16-18, but 0-4 since then (though that was in 2006). Kim was 1-7 against Serena, but never played her after Kim turned 20, and was 4-6 against Venus winning the last two matches. Kim was 10-12 against Henin, but after Henin won her first slam, she was 7-3 against Kim.

H2H is fascinating, but you need to examine the time periods. Henin and Federer both have complex games which didn't peak until they were at least 21-22, and both of them had so much talent that they reached matches against the best players in the world while they were still very young, and lost most of those matches. Once they did reach their peak, they have both been very dominant. Henin also had health problems which started in early 2004 and didn't entirely resolve until almost two years later, though she won the 2005 FO and 2004 Olympic gold medal in that period, and she was ill for the 2006 AO final. Kim, Serena, and Venus also had injuries which hampered them for periods of time.

Posted by Rosangel 01/13/2008 at 06:19 PM

slowpoke: the answer is 3. I will clarify above.

Posted by Ruth 01/13/2008 at 06:22 PM

Please...Bob! You say that "you need to examine time periods," something with which, it seems, all of us, including Ros agree. Yet, you seem to be implying that Henin's 2003-2007 time period is somehow more amazing than the two- or three- or four-year periods of dominance of other players. It probably isn't!

IF: You're right. "Yr turned pro" doesn't eliminate all problems, not when some players are slower than others in reaching the peaks of their performance, but it (ytp factor) does help to make things clearer, I'm sure, when you're looking at a group of top players who turned pro as early as 1993 and as late as 2004!

Posted by Rosangel 01/13/2008 at 06:28 PM

And Bob - I agree that closer examination of the progress of each H2H is desirable. The only problem is, I was trying to write a single post, and not an entire book!

Remarkable, though, that Henin's numbers are so strong even though they include the losing part of some of those H2Hs.

Posted by Ruth 01/13/2008 at 06:30 PM

And, Bob, if I ask nicely, would you please consider ceasing your references to the "small size and lack of power" of a player (Justine Henin) who, although she is is 3 inches shorter than Serena (for example), is one of the 3 or 4 strongest and hardest-hitting women on the Tour and whose head, shoulders, waist, thighs, and right arm are as big as, if not bigger than, those of most of the top 10 women???? Thanks!

Posted by Rosangel 01/13/2008 at 06:33 PM

I should add to the above, of course, that if I had similar numbers for Venus or Serena in their periods of dominance, I believe those would also be very strong, albeit against different fields.

And, also, the year turned pro number is one signal to help interpret the data - the number of total matches played is probably also useful. There's a big difference between being Agnes Szavay and being Lindsay Davenport, in that regard, and naturally Davenport's overall percentages will be more reliable as a result.

Posted by Bob 01/13/2008 at 10:24 PM

I was referring to the small size and power of Hingis. Henin has small size, but is incredibly powerful. Even a few inches means a gigantic difference in tennis. You don't see 5'6" players anywhere near the top of tennis, even though they far outnumber the 5'9" and above players, by several times. 5'9" is very tall for a woman, about 95% of women are beneath that height, yet from that 5% come nearly all of the top women. On the men's side, it's the same thing. We have Ferrer at the only player under 5'10", and the top players are 6'1" or more.

This is all interesting. These types of analysis articles should be archived separately, so they don't get lost. They take lots of work, and have interesting data.

This Paszek-JJ match is really full of drama and tension. They are so nervous. JJ's clutch play on that first match point was as good as I've ever seen. It was a long rally with several rockets by Paszek, and JJ got them all back and survived. Paszek has blown a couple of match points, too. Now JJ is up a break, but we've had 14 breaks of serve in this set. They are just getting them in. Now JJ has double match point. Experience is going to win this match. She did, and that was a miracle.

ESPN managed to show about three points of Henin's match. That's about par for them. She came in about 30 times, from what they said, practicing her net play. I would't be practicing anything in the first round. These are tense matches.

Thanks again, Roseangel.

Posted by Rosangel 01/15/2008 at 12:17 AM

Bob: I am going to look into setting up a separate category for all the stats analysis that has been produced so far.

Posted by creig bryan 01/15/2008 at 08:20 AM

Last!

ks


We are no longer accepting comments for this entry.

<<  Pre-Australian Open Crisis Center TW Sunday Brunch  >>




Wild Women of the U.S. Open
Wild Men of the U.S. Open
Roddick's Imperfect World
"It's Kind of a Dance"
Nadal's Kneeds
The Racquet Scientist: Canadian Tennis
The Long and Short of It
This blog has 3693 entries and 1646147 comments.
More
More Video
Daily Spin