Joe Trinsey, who attended a GMS volleyball clinic in New Jersey this past December, has really taken some of the concepts from the clinic and put a lot of additional thought into them.  At one point during the statistics session, we talk about a concept, developed by Tristan Burton, called “Hitting Effectiveness” as a measurement of an attacker’s value.  It goes beyond just hitting efficiency.  Joe shares some of his thoughts in the article below.

Does Conventional Volleyball Hitting Efficiency Matter?
In any discussion of volleyball statistics, hitting efficiency always appears as one of the most important stats. Most of us should know hitting efficiency as (kills-errors)/attempts or “kill percentage minus error percentage.” From a macro-view, hitting efficiency appears to be a pretty important statistic. In most matches, the team who hits for a higher efficiency will outscore their opponent. However, at the individual level, the way hitting efficiency is traditionally calculated breaks down a bit and may not always give a coach the information he or she is looking for.

One of the biggest deficiencies of conventional hitting efficiency is that it only accounts for kills and errors. Depending on the level, 30-60% of attacked balls will be dug or be played up on the hitter’s side after a block touch. For example, in the MPSF this year, about 37% of balls were not terminated (killed or errored). On the women’s side, about 46% of balls in the Pac-10 were not terminated. So obviously there is a lot here that is not being accounted for by conventional hitting efficiency. The question is: is there any way we can improve this statistic?

The most obvious way is to simply look at what happens when the result of the attack is not a kill or an error. A hitter who bails out on a difficult set and just puts it easily in play will allow the opposition to score easily in transition. On the other hand, a hitter who is consistently aggressive might make a few more errors, but not give up as many easy transition plays to his/her opponent. Instead of tracking “kills”, “errors”, and “attempts,” a coach might try tracking the ball from the time the hitter contacts it to the time it gets back over the net on his/her side again, and look at one of three outcomes:

  1. Our team scored the point
  2. The opposing team scored the point
  3. Neither team scored the point, and now we are starting another “attempt” with another (or possibly the same) attacker.

Sometimes this can produce some interesting departures from conventional hitting percentage. For example, in the 2007 women’s NCAA final between Penn State and Stanford, there was a point about midway through the match when Foluke Akinradewo and Nicole Fawcett had the following statlines:

Akinradewo: 10 kills, 1 error on 20 attempts
Fawcett: 9 kills, 4 errors on 20 attempts

By conventional hitting measures, this was no contest. Akinradewo was blowing Fawcett out of the water- .450 to .250.

However, a more in-depth analysis tells a different story. On the 9 balls where Akinradewo was dug, Penn State converted 4 of them. On the 7 balls where Fawcett was dug, Stanford only converted 1 and made 2 errors in transition. So another way to look at it would be:

Akinradewo: 10 Stanford points, 6 Penn State points, 4 new attempts for Stanford
Fawcett: 11 Penn State points, 5 Stanford points, 4 new attempts for Penn State

So despite the fact that one player’s conventional hitting percentage was 80% higher, the other player’s team scored 10% more points on rallies where she was set!

Now this is simply one (partial) example, and it doesn’t provide a rule of any sorts. Certainly this is a small sample size and is not meant to be a comment on either player’s effectiveness outside of a couple games in one particular match. One thing that needs investigating is whether this statistic is a robust one; is making it difficult for the other team to score in transition an ability that certain hitters consistently display? Or put another way:

  1. Do certain hitters consistently “underperform” their conventional hitting percentage by making things too easy on the other team when the sets are not perfect? Are certain hitters easier to get quality digs off of? Or…
  2. Do certain hitters consistently “overperform” their conventional hitting percentage by hitting shots that put the other team in a difficult position to score in transition? Do some hitters have a better sense for when it is acceptable to make an error (because just “keeping it in” would mean an easy point for the opposition) and when to be more conservative? And most importantly…
  3. Are these qualities coachable and can a more informed coach collect and use data to help guide his players to be more effective than their conventional hitting percentage would indicate?

Instinct would suggest that the answer to all three questions is yes. However, a more formalized study is probably needed. Unfortunately, this information cannot be obtained in a standard collegiate box score, so somebody would need to implement their own way of tracking their hitters over an extended period of time.

Does this mean that conventional hitting percentage is useless? No. Does it mean that a coach looking to gain that extra 2% (that we talk about so much at GMS If anybody has any comments, questions, feedback, or (especially) information they have collected that can further this dialogue