Category: Tennis Abstract

Another week, another enormous tranche of new women’s tennis data on Tennis Abstract. Today I present an extensive view of the 1963 season, including about 250 events and almost 3,000 matches. The season page is here, so jump in whenever you’d like.

This is the fifth amateur-era season I’ve added. I hesitate to use the word “complete,” because there is no clear line separating “tour level” from the rest, and for many of the tournaments I have only partial results. Even for the top players, some early-round matches may be lost to history. But as an in-depth view of the era, we continue to break new ground. For comparison, there were about 3,100 WTA tour-level matches in 2019, and we now have almost the same number of results from 1963.

I’ve made a few more improvements to the season pages, which are now available from 1963 to 1986:

• The Elo rankings table now includes columns for “iElo” — ratings specific to carpet (and wood and tiles and whatever artificial surfaces that organizers put on the floor of their indoor facilities). The “i” stands for “indoor,” although iElo does not include indoor hard or clay results. Those were rare at the time, and are included with the hard- and clay-specific ratings.
• The list of number-one ranked players now shows how long each woman held the top spot–including in other seasons. For 1963, the “list” is rather boring, as it consists solely of Margaret Court, but it does show that Court owned the number one position from the end of 1961 through to her first layoff in 1967. The exact numbers and start/end dates are very much subject to change as I add more data, correct errors, and improve the Elo algorithm, but all told, I have Court at #1 for a total of 536 weeks.

Coincidentally, I recently charted the 1963 Wimbledon final between Court and Billie Jean King. While it was their only meeting this season, it was one of more than 30 in their careers between 1962 and 1973.

As usual, the raw data is now available in my GitHub repo, and I gratefully acknowledge the work done by the Blast From the Past contributors at tennisforum.com.

The quest continues, and there are now another 3,200 matches in the women’s tennis database at Tennis Abstract. If you’d rather dive in to the data than read my ramblings about it, click here for the 1964 season page.

(If you’d like to read more of my ramblings, here are my intros to 1965, 1966, and 1967 data.)

The further back we go, we more we confirm the dominance of Margaret Court in the decade before the Open Era. In 1964, she won two majors, reached the final of a third, posted a year-end Elo just shy of 2500, and went undefeated over 44 matches on clay courts. Just about the only stats she didn’t dominate were three-set numbers, because she almost always won in straights.

Of course, there’s a lot more to 1964 than one Australian star. Importing these thousands of match results meant adding 360 new players to the database, including some important contributors whose career ended this season. Here are a few:

• Eva (Mannschatz) Johannes, the “Grande Dame” of East German tennis
• Valeria Kuzmenko Titova, an important early figure in the Soviet game
• Nina Vosters Moyer, two-time U.S. squash champion
• Joy Valderrama, once known as the “Shirley Temple of Hawaii“

The women’s season pages are now available for every year from 1964 up to 1978. You can navigate between seasons using the links in the upper-left corner of every page. I’ll further integrate the season pages into the rest of the site soon.

As usual, the raw data is available in my women’s tennis GitHub repo.

Finally: Another round of thanks are due to the contributors at tennisforum.com, who searched out newspapers and annuals, then typed up all these results. The same group is responsible for the Blast Encyclopedia of Female Tennis Players, an essential source for biographical data, especially married names.

Enjoy!

Welcome to the latest update on a project that has well and truly spiraled out of control. I’m pleased to announced that the Tennis Abstract site now features a huge amount of women’s tennis data from 1965. I hesitate to call it “complete,” because it is not, and it probably never will be. But the word “substantial” will do just fine:

• 3,200 matches
• 248 events (plus Federation Cup)
• 400 players that weren’t previously in my database

The 1965 dataset is even more sizable than the 1967 and 1966 results that I’ve recently discussed in other blog posts. To put those 3,200 results in perspective, there were “only” about 3,100 tour-level WTA matches in 2019.

For an bird’s-eye view of the 1965 women’s season, check out my season page. I introduced the season pages with my post on 1966 last week, and I’ve since made several improvements:

• The full event calendar has some new information to indicate the strength of the tournament: the number of top 10 players in the draw (as per that week’s Elo ratings), and the “geographic concentration” of the field–that is, the percentage of women in the draw who hail from the most common country. The second number isn’t perfect, especially when I only have a few results from the event, but as a general rule, the lower the geographic concentration, the stronger the field.
• The year-end Elo rankings table includes some helpful additional information: each player’s age, her number of titles, and her won-loss record on the season.

The season page tends to highlight the best players, and I’d imagine that’s what most of you will find the most interesting. Margaret Court dominated the 1965 campaign, winning over 100 matches, losing only 8, and posting the best year-end Elo on all surfaces. The page will also tell that you she drew Lesley Bowrey ten times–nine of them in finals!–and Bowrey accounted for 4 of her 8 losses.

(Court and Bowrey were already familiar foes: They met in the 1960 Australian Championships girl’s final. Court lost, but bounced back quickly, winning the women’s final–her first major title–the next day.)

Equally fascinating for me are the names you almost never hear in their tennis context. Since I’m working backward, the players I added to the database for 1965 were those who finished their careers that year. (Or played predominantly at lesser regional events, and only briefly popped up on my radar.) Here are a few of the ladies whose tennis careers I stumbled upon:

• Sandy Warshaw, who later become mayor of Tampa
• Lillian O’Donnell, a noted detective novelist who set one of her stories at Forest Hills
• Maria Chiara Ramorino, who became better known in the orienteering world, and has an Antarctic glacier named after her
• Rita Horky, a Hall of Famer in women’s basketball

I could list many more.

Data and acknowledgements

Once again, I note the huge debt I owe to the contributors at tennisforum.com’s Blast From the Past section. They’ve converted newspaper and annual results into online content that I could then further organize into a proper dataset.

All of the raw data is available in my women’s tennis GitHub repo.

I’ve been working hard to organize 1960s and 1970s women’s tennis results so that you can view and search it as easily as if they took place last month. It’s an enormous task, and probably never to be completed, but I do have some progress to share.

A couple of weeks ago, I announced the inclusion of the 1967 women’s tennis season on Tennis Abstract and discussed why it’s so important. Today, I give you 1966, along with a much easier way to dive in.

The season view

Here’s a one-page overview of the 1966 season. On that one page, you’ll find:

• The results of the four majors, at a glance
• Some key statistical leaders
• A full calendar of all the tournaments in the database, along with finalists and semi-finalists (in 1966, that’s 159 events!)
• Year-end Elo rankings, including surface-specific ratings (yes, Elo for the 1960s!)
• Elo number ones for the season (Margaret Court made that rather uninteresting for much of the decade, monopolizing the top spot this year and several others)
• Sortable stats for the 30 most active players, including won-loss records in finals, in three-setters, and on all surfaces
• The most common head-to-heads
• Country-versus-country won-loss records, which offers a glimpse of which nations predominated at the time

Of course, the page contains links galore. One more click gets you detailed player pages just like the ones available for current players, or event-specific pages with full tournament draws. The database contains over 2,600 matches from the 1966 season.

(Once I work out all the kinks, I’ll generate similar pages for later seasons as well.)

What’s here and what’s not

To repeat myself from the 1967 post: This project owes a tremendous debt to the contributors at tennisforum.com’s Blast From the Past section. They’ve typed in tens of thousands of results compiled from newspapers and annuals. Without their efforts, I would barely be getting started. I highly recommend browsing that forum. In addition to the singles results, it contains doubles and mixed doubles scores, as well as descriptions of some of the top events. It’s one of the truly invaluable corners of the internet.

Newspapers and annuals didn’t report everything, and even the tireless Blast compilers haven’t scanned every possible source. Thus, some tournaments are missing rounds or specific matches. For some events, I have only the final. There are still other events that I would love to include, but am unable to for lack of data, such as the annual ATA championships and many of the tournaments that took place in the USSR.

I also haven’t imported every single possible result. There was no clear demarcation between “tour-level” and the rest back then, but some events were much stronger than others. Just because the results of the Wyoming state championships have survived doesn’t mean you can find them on Tennis Abstract.

That said, I’ve erred on the side of over-inclusion. There is at least one result from over 150 different 1966 events, and that number will be over 200 from 1962 to 1965! If a tournament has even one great player, I’ve imported the entire draw. (Ann Jones, who seems to have played just about every tournament in Great Britain for 15 years, has repeatedly made me question that commitment.) I’ve included virtually everything from the USSR and the former Eastern Bloc nations, along with nearly every tournament that included players from Eastern Europe. There was much less East-West mixing than there is now, so these results are particularly important for establishing the level of play behind the Iron Curtain.

About these Elos

It’s particularly exciting to be able to rate these players, both to find unheralded women from this era, and to see how the stars of the 1960s stack up against those of later eras. Of course, a certain Elo rating doesn’t mean the same thing in 1966 as it did in 2016, because the level of play has risen, and the game has changed in innumerable ways. That said, my Elo algorithm doesn’t suffer from any kind of inflation, so a certain rating–say, Billie Jean King‘s 1966 year-end 2274–means roughly the same thing relative to her peers as it does now.

These Elo ratings are provisional, however. For one thing, there’s a lot more historical data to be added. As the algorithm can look at more matches from the early 1960s, it can better calculate proper ratings for each player in 1966.

Also, the less-structured nature of the tennis tour in the 1960s may necessitate some tweaks to the algorithm. As I’ve said, there’s no clear top level, and there’s certainly no helpful classifications like Satellites or Challengers or ITF W15s. While the best players did a lot of traveling, they represented a much smaller core than the hundreds of full-time nomads who populate today’s tour. Thus, 1960s stars played more early-round matches against locals who–at least in tennis terms–would never be heard from again.

So far, my Elo algorithm is spitting out plausible results for the 1960s without any era-specific alterations. Adding thousands more matches and hundreds of new players is not causing any noticeable inflation in the ratings of later players. But any of those things might change.

The data

I’m making all of this data available in my GitHub repo for women’s tennis results.

In addition to “new” seasons like 1966, I’m also working on filling in lower-level events and qualifying rounds for the 1970s. I have about 50 tournaments per year from 1968 through the mid-70s, but I’m finding that there are 100 or more per year that could be added, plus qualifying for the big events. I recently added 1,500 such “additional” matches from 1974 alone.

These are all on Tennis Abstract as well, so to take just one example, you can see Virginia Ruzici fighting her way through qualifying rounds at the big tournaments to start 1974. Once I finish with 1973, you’ll be able to see evidence of something almost unthinkable: Martina Navratilova playing qualies. It didn’t last long, but it did happen.

Enjoy!

Last week, I finished* adding complete** 1967 women’s results to the Tennis Abstract site. I’ll talk about those asterisks in a bit, but for the moment I’d prefer to revel in how cool this is.

The “Open Era” starts in 1968, and in the near-decade since I launched TA, I took that year as my starting point. Along the way I added men’s slams and Davis Cup back to the beginning, but it’s buried on the site as an afterthought. I can’t imagine that anyone uses the site for amateur-era results.

Even late 60s and 70s results were spotty for women. I initially built my database from the results published on the WTA and ITF websites, neither of which is (how to put this mildly?) primarily focused on the thoroughness and accessibility of its historical data. Add in the mistakes and omissions that come from building my own database from scratch, and you end up with a lot of gaps.

A more complete Tennis Abstract

A few weeks ago, I started filling in those gaps by adding about 20 missing tournaments with a Chris Evert–Martina Navratilova match. That head-to-head is now complete. Soon it will be “more than complete,” as I add various exhibitions that don’t count in the official tally. From there, I used various sources (more on that below) to fill in the remaining gaps of top-level Open Era women’s tennis back to 1968. The result is about 50 full tournaments per year, sometimes more, with various bonuses like Federation Cup and a lot of grand slam qualifying.

The further back I went and the more I stumbled on stories about the women’s game at the beginning of the Open Era, the more I wanted to know. 1968 is an important year, but a lot of tennis was unchanged from 1967 to 1968–almost all of the same players excelled, on the same surfaces and mostly at the same events. It seems a little silly to have a statistical record that starts smack in the middle of all-time-great careers like those of Billie Jean King and Margaret Court.

Into the unknown

One of the most incredible online tennis resources is one you’ve probably never heard of. On the “Blast From the Past” section of tennisforum.com, a group of contributors have assembled a unparalleled collection of women’s match results going back to the 1800s. They’ve dredged up results and tournament information from old annuals, newspapers, and just about any other source you can imagine.

The disadvantage of their forum-based, text-based format is that it is only awkwardly searchable. (Just to be clear, I am not taking anything away from their outstanding efforts.) The forum approach does allow for a certain kind of serendipity, and I’m sure I’m not the only one who has lost hours scrolling, reviewing results, reading the tournament recaps and anecdotes collected there. But it precludes the kind of serendipity made possible by sites like Baseball Reference and Tennis Abstract, where you see one result, get curious about a player, click the player’s name, and find yourself looking at a whole new list of unfamiliar scores and stats.

The further back in history we go, the more I want that kind of serendipity. Now, Tennis Abstract has that for 1967, and soon it will go back further still.

Okay then: 1967

The site now includes results from over 100 events in 1967, from familiar names like Rome and Queen’s Club to lesser tournaments such as the Pan-American Games (held that year in Winnipeg) and the Soviet Championships in Tblisi. I don’t have complete data for every draw–some are missing a handful of first-rounders, and others have only the final round or two. All told, the database now includes almost 2,300 matches from that single year. By comparison, there were about 3,000 tour-level WTA matches in 2019.

Since there was no formal “tour” in 1967, there’s no official definition of what’s “in” or “out.” A match is a match. I didn’t include every single event with some kind of data available, but I did import the entire main draw of any tournament with even a single “big-name” player, using a fairly broad definition of that term. (1969 Wimbledon champ Ann Jones may make me regret that decision. She played a lot of tennis.) Because the various circuits were more fractured, that means more events: There were many weeks with three or four tournaments each, and a couple with five.

Creating records for those 2,300 matches meant adding almost 300 players who weren’t in my database. The majority of those are early-round losers in small events, women who didn’t seriously pursue tennis. But where I had a full name, I did at least a cursory search for each one, turning up a noted Spanglish poet, the “first grunter,” a squash Hall of Famer, and Marat Safin’s mom.

100 events sounded like a lot until I started working on 1966. I have a provisional list of 160 tournaments to include from that year. Even with all those caveats on the meanings of “finished” and “complete,” this is going to take a while.

Diving in

Here are direct links to 1967 results for a few players:

• Billie Jean King
• Virginia Wade
• Nancy Richey
• Francoise Durr
• Rosie Casals
• Galina Baksheeva

If you go to the main page for one of those players (for example, here’s Peaches Bartkowicz), you’ll find a cool addition that all the new 60s and 70s data has made possible: women’s Elo ratings back to the end of 1967. Player pages for women who played at least 20 matches in a season include their year-end ratings and rankings, including surface-specific figures.

Here is a very provisional overall top 10 for year-end 1967:

Rank  Player               Elo  
1     Billie Jean King  2221.3  
2     Virginia Wade     2114.9  
3     Nancy Richey      2113.2  
4     Judy Dalton       2083.3  
5     Ann Jones         2042.7  
6     Lesley Bowrey     2018.8  
7     Kerry Reid        2006.0  
8     Francoise Durr    2005.4  
9     Rosie Casals      1940.4  
10    Annette Du Plooy  1926.8

I say provisional because there’s so much left to add. (You know, the entire history of tennis prior to 1967.) At the moment, the algorithm doesn’t know anything about any of the players prior to January 1st, 1967. As it learns more, each player’s rating will be different at that point, and the year-end results will be tweaked as well. That goes for all Elo ratings and rankings throughout the 60s and 70s. The broad strokes will remain constant, but the exact numbers will change, and sometimes players will swap positions. As I add more data, King, Court, and Richey (among others) keep creeping up the all-time list.

As for the project as a whole, I have no idea how far I’ll get. While fascinating, it’s a time-consuming project, and the further into history we go, the less information is available on players beyond the all-time greats. Still, every small step back in time improves the accessibility of this period of women’s tennis data, which includes some of the most important players in the history of the sport.

About those sources

I’ve mentioned tennisforum’s Blast From the Past, which is truly essential. Another exhaustive source for match results starting 1968 is John Dolan’s book, Women’s Tennis 1968-84. Wikipedia has oddly spotty coverage: the Italian Wikipedia is good for tournament data, while the French Wikipedia seems to cover more players. (For Swedish players, Swedish Wikipedia is awesome. All that time spent learning Norwegian is finally paying off.) English Wikipedia is disappointingly lacking in comparison.