Every week, Tennis Analytics dives deep into the data, bringing you a stat of the week to help you understand tennis better and become a little smarter.
For the next few weeks, we will look at the WTA Top 4 – Barty, Halep, Osaka, and Kenin. What are the key match stats that separate them from the rest of the pack?
Findings are based on matches tagged by Tennis Analytics from 2018 to 2020.
Stat of the Week: 48%
The top 4 WTA players win 48% of points on their own 2nd serve. Even Serena Williams, considered one of the best servers in history, won just 48% during this same period.
This means that at every level of women’s tennis, as soon as the 1st serve misses, the advantage goes to the returner!

You can see that Barty is the only player who starts in a neutral 50/50 position on 2nd serves.
Point of the Week
Using the above knowledge, players can counteract this by closely monitoring two key areas.
- 1st serve percentage along with 1st serve win percentage. The WTA top 4 get an average of 62% of 1st serves in play, and win 68% of those points.
- On the flip side, consider the return of 2nd serve as one of the most important shots in your arsenal. The WTA top 4 win 55% of points when returning 2nd serves.
How does your own serve and return stack up to the world’s best?
Find out with a Tennis Analytics Player Package.
Use coupon code StatOfTheWeek for a special discount.
Photo of the Week
Sofia Kenin’s forehand.

James Mitchell says
Alternatively the women should consider only hitting “first” serves. Esp Osaka. What I would need to know is how many first serves go in for each player.. 62% is the average. Double faults go up but at 62% you would put the ball in play 85% of the time. Its just probability math!!
Warren Pretorius & Will Boucek says
Good points James! We’ve kind of seen that from Zverev this year. He lives with the double faults and his 2nd serve win percentage is off the charts.
We’ll have to do the math for the player to see if it’s worth it.
James C Mitchell says
Exactly. Each player needs to do the decision tree for their own statistics.