Serving Profiles - Part 1

Introducing the data

Related:
Digging Profiles
Blocking Profiles

Articles in this series:
Part 1 - Introducing The Data
Part 2 - Serving For Aces
Part 3 - Serving In
Part 4 - Knockout Queens

The Data

I sample here is NCAA Women’s top-100 RPI teams from the 2023 season. There’s some pros and cons of using this as the data set. In previous years, I used just Power-5 conference play when running similar analyses. The upside with that is a bit more of a constrained data set. I had teams from the top and bottom of conferences and they were (mostly) all playing each other. The downsides are (a) fewer total teams and (b) you miss out on some relevant teams from outside the P5 conferences. Also, the term P5 doesn’t really exist anymore, so why not just go blanket top-100 RPI, amirite?

One thing to note is that these teams as a general sample are more successful than the median NCAA team, because these are the top 100 out of 300+. The median Win % in this sample is 74%, the equivalent of going 22-8 in a 30-match season. But… that might not be a bad thing. We’re trying to build profiles of winning teams and the full 300-team sample gets a little unwieldy.

The Stats

The stats I’m mining for the serving portion of the game are:

Ace %
Error %
Opponent Good Pass % (OGP)

Which then gives you two additional derivations:

Ace:Error Ratio
Knockout %

Ace:Error ratio is, as you might expect it to be… the number of aces a team serves divided by the number of errors, expressed as a ratio. 10 aces / 9 errors = 1.1 A:E, 7 aces / 9 errors = 0.8 A:E

Knockout % is the percent of times you knock the other team out-of-system. In this case, it’s basically (1-OGP) from above. However, you also factor in missed serves. When you miss a serve, you have failed to knock the other team out-of-system. So it becomes (1-OGP) * (1-Errors). So if you serve 10 balls, miss 1 serve, and knock the other team out-of-system 4 times, the other team has a 44% OGP (4 divided by 9) and you have a 40% KO% (4 divided by 10).

Top - Bottom - Middle

Let’s get a picture of what these numbers look like at the NCAA level.

Ace %
9.4 % - #1 (Dayton)
6.6 % - #50 (Coastal Carolina, Washington State, Liberty, Minnesota)
4.0% - #100 (DePaul)

Error %
5.1% - #1 (DePaul)
9.5% - #50 (Georgia Tech, Clemson, Miami, Liberty)
13.7% - #100 (Stanford)

OGP
39% - #1 (South Florida)
46% - #51 (A lot of teams)¹
54% - #100 (New Mexico State)

A:E Ratio
1.1 - #1 (Southeastern Louisiana)
0.7 - #50 (A lot of teams)²
0.4 - #101 (Texas A&M)

Knockout %
54% - #1 (Yale, UMBC)
49% - #51 (A lot of teams)³
41% - #100 (New Mexico State)

What’s Next

The goal of these profiles is less about “who is good” and more about the different shapes of strengths and weaknesses. No coach ever has enough time and no roster is a collection of perfect players. We make trade-offs when we plan a starting lineup: do you start the more powerful attacker with worse ball control or vice versa? We also make trade-offs with practice time: do you spend more time on serve/pass or on offense? And finally, we make trade-offs with strategy and tactics: do you serve for aces and risk missing some more serves or do you keep the ball in play and trust your block/defense?

The next 3 parts of this will divide serving into 3 different profiles:

1. Teams who serve a lot of aces, at the cost of missing serves

2. Teams who serve in a lot, at the cost of less service pressure

3. Teams with high service pressure but who don’t necessarily serve a ton of aces

I’ll look at the broad success rates of those profiles. ie, do teams from 1 of those 3 categories tend to be more or less successful? I’ll also look at, within the profile, what seems to make the difference between the less successful teams with that profile and the more successful teams?

Look for those parts in the upcoming weeks!

1. Georgia Southern, Texas, Loyola Chicago, Creighton, Cal Poly, Michigan State, South Carolina, Eastern Illinois, Coastal Carolina, Stanford, Nebraska, Miami, Pitt, Washington State, USC, LSU

2. Villanova, Drake, Utah State, Houston, Arkansas, Washington State, Pepperdine, FGCU, Indiana, Citadel, Louisville, Creighton, Xavier, BYU, Miami, Colorado State, Colgate, SMU, Liberty, Georgia, Kansas State, Pitt, Ohio State, Minnesota, Hawaii, UTEP, Wisconsin, NC State, Duke, Cal Poly, Ohio

3. Coastal Carolina, Florida, Texas, Eastern Illinois, Towson, Buffalo, Loyola Chicago, Northern Iowa, Washington State, Pitt, Cal Poly, Florida State, Miami