top of page

Redefining Volleyball Blocking Grading Levels for Improved Analysis

Jun 10, 2024
7 min read

tl:dr

  • How I currently grade the block has numerous issues because 1) it doesn’t have a grade for negative or positive block touch, 2) it has no grade for block attempt with no touch on attacked ball, 3) it has three blocking grades that relate to a 0% Rally Win % which isn’t great if you want to calculate an Expected Value

  • DataVolley’s blocking grades don’t help much

  • I will likely adopt Former USA Women’s Scout Joe Trinsey’s blocking grades of Block Kill (B#), Positive Block (B+), Rebounded Block (B!), Negative Block (B-), No Block Touch (B/) and Block Error plus Tooled Blocks (B=) because it has all the blocking grades I want

Measuring Blocking Performance

We usually measure blocking performance whether the player gets a stuff block for a point won or commits an error for a point lost.  This is known as a block kill and block error, respectively.  In DataVolley, we would code block kills as B# and block errors as B=.  There are also four other codes (+!-/) we can use to grade the block in DataVolley.  I have my own definitions of what they mean but I already reflected that it is not the best way to measure blocking performance.  So I wanted to reevaluate what is the best way to redefine the blocking grading levels in volleyball so we can get the best analysis out of it.  

Grading Other Technical Skills

This is one in a series on how I grade the various technical skills. Click on the links below to read more.

2023-24 Blocking Grades

My current 2023/24 grades can be found below.  Block kills (B#), rebounded blocks (B!) and block errors (B=) are pretty much set in stone.  There are issues with the other grades of B+, B- and B/.  

No Positive or Negative BlockOne issue is I don’t have a grade for whether the block touch was positive or negative.  They both come under the Block Touch (B+) code.  This means a block touch that slows down the attack to make it easier for the defence to play in-system and a block touch that makes it hard for the defence to play in-system are given the same grade when they are clearly not the same.

No Grade For Missed Block AssignmentA common performance indicator to measure blocking performance is Block Kill % (= Block Kill/ Total Blocking Attempts).  A major problem with performance indicators like these is Total Blocking Attempts, as it is currently statted, only include block attempts that touch the attacked ball.  However, most blocking attempts don’t touch the attacked ball at all.  This means the majority of blocking attempts are not recorded at all and this can skew the blocking stats.  A player with 2 block kills in 4 total block attempts, i.e. 50% block kill rate, might sound good but if you also knew the player had 46 block attempts that did not touch the attacked ball, i.e. 4% block kill rate, then it doesn’t sound so good all of a sudden.  So I want to eventually have a blocking grade where the player goes for the block but does not touch the spiked attack, i.e. missed block assignment, so we can get a true account of total blocking attempts.

Three Grades Relate to 0% Rally Win RateThe issue with the current blocking grades is three grades (B=, B/, B-) relate to a 0% Rally Win %.  If I want to eventually calculate a similar Expected Value to measure blocking performance (e.g. see Expected SideOut % to measure passing), I want to have clear differences in Rally Win % (or something similar like Hitting % for setting) between each grade.  I don’t want half of my grades having a 0% Rally Win % and one grade having a 100% Rally Win % (Block Kill), leaving two blocking grades somewhere in between.  It makes sense to clump those three actions together under one grade so I can make better use of the other blocking grades.

DataVolley Blocking Grades

It is worth referring to the DataVolley Handbook to see how they do it.  The table below shows how DataVolley grades the block.  It’s standard to have block kills as B# and block errors as B=.  I’m surprised that rebounded balls are coded as B- by DataVolley and not B!.  When the attack is compound coded with B!, the attack direction is shown as a rebounded attack, so it makes sense B! should mean rebounded ball.  I’m also surprised that when the ball is touched and can be played again by the home team isn’t coded as either positive or negative.  I’m also not sure what Invasion means – perhaps blocking the setter?  It was a good exercise to see how DataVolley grades the block but it hasn’t inspired any good ideas in a better way to grade the block.  

Joe Trinsey Grades

Former USA Women’s Scout Joe Trinsey has his own system to grade the block (see below).  It has the three regular blocking grades of block kill (B#), rebounded block (B!) and block error (B=).  It also has all the blocking grades I am looking for, chiefly the positive block, negative block, and missed block assignments.  

Tooled Blocks Having the Same Grade as Block Errors?One drawback is having tooled blocks under the same code as block errors (B= code) (click here to read more).  I wasn’t a fan of this because I believe block errors (e.g. net contact, blocking the setter) are unforced errors and should be -1 in the point-scoring stats.  Tooled blocks are not the same as block errors because although both actions come at the end of the rally, I consider tooled blocks as forced errors (i.e. the spiker won the point) and therefore should not count as -1 in the point-scoring stats.  This is why I coded tooled blocks under the B/ code to distinguish them and make it easier when calculating the point-scoring stats.  I wasn’t a fan of this because I believe block errors (e.g. net contact, blocking the setter) are unforced errors and should be -1 in the point-scoring stats.  Tooled blocks are not the same as block errors because although both actions come at the end of the rally, I consider tooled blocks as forced errors (i.e. the spiker won the point) and therefore should not count as -1 in the point-scoring stats.  This is why I coded tooled blocks under the B/ code to distinguish them and make it easier when calculating the point-scoring stats.  

However, I realised I could simply use the extended codes to distinguish block errors and tooled blocks.  I can code Block Errors under the “Net Contact” extended code and Tooled Blocks under the “Ball Out Side”, “Ball Out Long”, and “Ball on Floor” extended codes.  I can code Block Errors under the “Net Contact” extended code and Tooled Blocks under the “Ball Out Side”, “Ball Out Long”, and “Ball on Floor” extended codes.  It would mean I need to modify formulas related to tooled blocks, block errors, and point-scoring stats It would mean I need to modify formulas related to tooled blocks, block errors, and point-scoring stats to include those extended codesto include those extended codes but at least it would allow me to have a grade for Missed Blocking Assignments. but at least it would allow me to have a grade for Missed Blocking Assignments.

How to Code Block Touches That Lead to a Dig After Block Error?Since these block touches lead to the end of the rally by definition, it makes sense to code it under the B= since the B= code has a 0% Rally Win rate.  It makes any Expected Value calculation logical if done this way.  However, it is a bit harsh on the blocker because maybe it was only a slight block touch that didn’t deflect the spiked ball’s direction at all or they really slowed down the attack and the defender messed up in both cases.  We want to judge the block touch on its own terms, not what another player has done.  So I think it is ok to grade blocking as either a positive or negative regardless of the outcome of the defence.  I should also remember many Missed Block Assignments lead to a dig error so I think it’s ok to have 0% Rally Win % for some blocking grades.  

What is the Difference Between a Positive and Negative Block?The big question I haven’t thought deeply about is how to define what a positive block is and what a negative block is.  I thought about whether the defence is in-system or out-of-system.  However, as I just discussed above we want to judge the block touch on its own terms and not what the defence does.  My best idea is a positive block is when you slow down the spiked attack to make it easier for the defence to defend and a negative block is when you deflect the spiked attack to make it harder for the defence to defend.  What about a block touch that doesn’t make it harder or easier to defend the spiked ball, e.g. there’s a slight block touch that doesn’t deflect the ball that much?  Maybe I would code it under the Missed Block Assignment since the attack direction is still the same.  However, grading the block is one of the last things I want to code so I still have time to settle upon a definition.

Final Thoughts

I will probably adopt Joe Trinsey’s grades when I eventually learn to grade the block.  It solved all of the issues I had with my current grades.  I going to make a start next season by coding I going to make a start next season by coding tooled blocks under the B= code rather than the B/ code.  tooled blocks under the B= code rather than the B/ code.  

 
 
 

Comments


bottom of page