Statistical Analysis of Top Liberos' Serve-Receipt

[meanmug]

I may be misunderstanding this formula, but... I would say that this formula probably does a better job of pointing out POOR serve receivers than GOOD serve receivers.

For example, a player could make errors on 20% of receptions (which is terrible at the high collegiate level) and pass 40% of receptions and be scored at a 0.26. On the other hand, a passer could never be aced and pass 25% of her team's receptions and be scored at a 0.25, which would rank worse.

In general, there is an inverse correlation between % of service receptions and quality of receiver. Meaning: teams generally serve the worse passers more.

You're better off just looking at straight-up reception error%, and all of them are between 2% and 5.6%, which, considering the sample size here isn't really enough to draw much of a conclusion from, other than that Moster's reception error % (5.6) is a bit high for an elite libero, although as you noted that was almost all just a consequence of one match.

Although, it should be troubling for USC that Hagglund has only been able to receive about 16% of team receptions in the matches listed. That's going beyond good target serving.

[meanmug]

Ye Old Dawg,

You don't even need to dive very deep into specifics matches. Just look at the receivers list for Grand Prix Finals:

www.fivb.org/EN/volleyball/competitions/WorldGrandPrix/2013/StatisticsF.asp

By this rating system, Kim Hill would score a 0.556 S2R while passing at a 33% efficiency and the Brazilian bro would score a 0.208 S2R while passing at a 58% efficiency. There's a reason that the only libero to receive over 80 serves in the tournament was also by far the worst.

[Ye Olde Dawg]

Well, I was suggesting that alantech run the numbers, but thanks. Skepticism validated.

[goodtobeagator]

Until statisticians learn to grade pass quality with low variance, judging libero play will be subjective. Passing skills aren't "yes" & "no" results. It's 0,1,2, or 3 options for the setter to use.

A for effort. But until I know how the pass influences ease of running offense, I'll pass on this analysis (pun intended).

[jasonr]

Nov 1, 2013 15:53:46 GMT -5 goodtobeagator said:

Until statisticians learn to grade pass quality with low variance, judging libero play will be subjective. Passing skills aren't "yes" & "no" results. It's 0,1,2, or 3 options for the setter to use.

A for effort. But until I know how the pass influences ease of running offense, I'll pass on this analysis (pun intended).

That could be done over a small sample, e.g. 5 games. The statistician would have to grade out each pass (it'd be subjective based on location and the ease with which the setter handles it), then find the median score over sample number, something like that.