Should Stuffed Attacks Be Considered the Same as Attack Errors?

tl:dr

• A stuffed attack is when an attack attempt is rebounded off an opposition block attempt and the attacking team cannot keep the ball in play as a result, i.e. the blocking team got a stuffed block, otherwise known as a block kill.  They are classed as an attack error.  

• Stuffed attacks should be not considered attack errors or points lost (+0) because they are forced errors (the blocker earned the point rather than the attacker losing the point) and you don’t want to double up on points in a single rally (since it’s -1 to the attacker and +1 to the blocker)

• Stuffed attacks should still be points lost (-1) because players and coaches understand stuffed attacks as attack errors and it will need a lot of explaining and convincing that they are not.  If you don’t achieve either then players and coaches may distrust the stats and may not utilise the stats.  It will also change the reference points of what a good hitting % is.  Again, if players and coaches don’t agree with the stats, this puts distrust in the stats.

• Both arguments are convincing and I am currently on the path of least resistance of continuing to class stuffed attacks as attack errors.

Stuffed Attacks

A stuffed attack is when an attack attempt is rebounded off an opposition block attempt and the attacking team cannot keep the ball in play as a result, i.e. the blocking team got a stuffed block, otherwise known as a block kill.  A stuffed attack is generally classed as an attack error and a point lost (-1).  Attack errors would mainly include attack attempts that go out of bounds, into the net, net contact, centre-line violation, etc.  Some scouts and software programs may not distinguish between the two so they would record all stuffed attacks as attack errors.  

Since stuffed attacks are attack errors and points lost (-1), they form a part of a performance indicator that measures attacking performance – the Hitting %.  It is calculated as Hitting % = (Attack Kills – Attack Errors – Stuffed Attacks)/ Total Attack Attempts.  When a stuffed attack (-1) is charged to the attacker, a block kill (+1) is also charged to the blocker. 

However, should stuffed attacks be considered the same as attack errors?  I make arguments for both sides.

Stuffed Attacks Should NOT Be Considered as Points Lost

Points Lost Should Only Include Unforced Errors and Stuffed Attacks Are Forced ErrorsMark Lebedew makes an argument that either a player earned a point (+1) through their good actions that led to a forced error (+0) or a player made an unforced error (-1) that led to an unearned point (+0).  You can’t have a situation where a player earned a point (+1) AND the opposition player made an unforced error (-1).  However, the block kill (+1) and stuffed attack (-1) treat the situation this way.  

This means we need to logically decide if the blocker earned the point (+1) through their good actions which led to an unforced error by the attacker (+0), or if the attacker made an unforced error (-1) which led to an unearned point by the blocker (+0).  I think we can agree it is the former situation as it sounds silly to say that the blocker did not earn the point.  

Each Rally Should be Worth 1 Point But Stuffed Attacks are Worth 2 PointsYou want each rally to be worth 1 point only.  Ideally, it should either be +1 to one team (for earned points) or -1 to the other team (for unforced errors).  However, the stuffed attack-block kill situation is worth two points because it is simultaneously -1 to the attacking team and +1 to the blocking team.  This doubling of points will really skew the point-scoring stats, especially if there is a high occurrence of block kills.  

Stuff Attacks SHOULD Be Considered as Points Lost

Players and Coaches May Not AgreeIt is so ingrained that stuffed attacks are attack errors that it will need a lot of explaining and convincing why stuffed attacks are not attack errors.  If you put in the time to explain it, the players and coaches may not be convinced.  If the players are not convinced, it may put distrust in the stats that they might not want to utilise it.  This is something you want to avoid.  The whole reason raison d’être of volleyball stats is so those responsible can use it to make informed decisions to enhance performance.  If they are not used due to perceived faulty data, then you have lost your raison d’être.

It Changes The Reference Points for the Hitting %The reference points of what a good hitting % [= (Attack Kills – Attack Errors – Stuffed Attacks)/ Total Attack Attempts] is well established.  Generally, we think of hitting 30% as the benchmark of great hitting.  This will change depending on the level (it could be 20% for lower level) and position (you add 10% for middle players).  

So if we decide that stuffed attacks are not points lost (+0), that means we should logically remove stuffed attacks from the Hitting % equation.  If we did that, this would change the players’ and coaches’ reference points of what good hitting is.  How good is 20% hitting now?  I remember I had a situation when the second opposite came off the bench.  He had a few of his attacks stuffed and he was eventually subbed off.  His attack stats were something like 2 kills, 3 stuffed attacks, 0 attack errors, and 10 total attack attempts.  Using the original calculation, he has a hitting % of -10% (0.100) and we understand this to be a poor attacking performance.  Using the new calculation, he has a hitting % of 20% (0.200), which isn’t that bad.  So was the attacking performance poor or not that bad?  Most players and coaches expect this to be a poor attacking performance but you are proposing that it’s not that bad.  Again, this may put distrust in the stats that the players and coaches may not want to utilise.  

Discussion

The argument that stuffed attacks are forced errors and that you don’t want to double up on points was convincing enough for similar situations in reception errors and tooled blocks that I don’t consider those two actions as points lost (+0).  It seems logical that I should do the same for stuffed attacks.  I should simply remove stuffed attacks from the Hitting % and point-scoring calculations.  

However, the argument that players and coaches not may agree that stuffed attacks are attack errors and that it changes the reference points for the Hitting % could be stronger.  You don’t want a situation where it can lead to players and coaches distrusting the stats and not utilising them.  If they do then you have lost your raison d’être.  

I did think about including stuffed attacks in the Hitting % calculation but excluding them in the point-scoring calculations.  Players and coaches don’t have as good reference points for point-scoring stats as they do for attacking stats.  It’s the same thing when I removed reception errors in the point-scoring stats.  However, the tricky thing to consider is how to show that in my match report (see below).  Should I include stuffed attacks as part of the attack error count in my match report?  If I do then the total error count does not total up correctly.  If I exclude them in the attack error count then the Hitting % doesn’t calculate properly.  I could add another column but this will require a tremendous amount of work because it affects the rows below and into the other pages.  

Another thing I have thought about is to make the change for the teams that I coach with players that are not familiar with the stats, e.g. the American School of London.  There’s no need for convincing as they don’t have that reference of what a good hitting % is.  The issue I face is my own reference points of what good hitting percentages look like will change.  This might be overcome if I research some conversion numbers (e.g. converting last year’s players’ hitting percentage).  The change would mean I have to create a second different match report that excludes stuffed attacks in the hitting %.  I want to avoid this because if I want to make some changes to the match report, I would have to duplicate those changes in the second report.  This was annoying because I used to have several team-specific match reports with some specific stats, e.g. my City University London had aggression stats and Team SideOut Polonia London had expected sideout and breakpoint stats.  I would modify one match report and forget to change the others, e.g. when the NVL allowed 14 players on a team list, I had to adjust the height of my rows so I was able to fit a maximum of 28 players on one page.  This is why I condensed them all into one single match report.  

Final Thoughts

This is quite a tough decision.  It’s logical to make the change and it’s going to take a lot to convince players and coaches of the change.  I do like the idea of including stuffed attacks in the hitting % but leaving it out in the point-scoring calculations.  However, that’s going to take a lot of work to fix the match reports.  It may be something I do in the future.  So for now, I am on the path of least resistance so I will continue to class stuffed attacks as attack errors.