VAR may be precise, but it’s not necessarily accurate. So why is the Premier League still relying on it to judge offside?

As I write this, Covid-19 is still killing thousands of people each day.  The vaccines that will protect us are being distributed, slowly, but it will take many months before we reach any semblance of herd immunity.  Meanwhile, the economy is a shambles, with unemployment among certain sectors still at a depressingly high rate, small business owners in crisis, and deficits at all levels of governments that will take a lifetime to be paid down.  There’s been an attempted coup in the United States, and a successful coup in Myanmar.  Climate change is continuing almost unabated. It’s all a bit much to deal with at one time.

So…..I want to talk about the use of VAR (Video assisted referee) in the adjudication of offside decisions in football (soccer).  Frankly, it’s a bit of a mess.

For the uninitiated, VAR is called in on close plays that have resulted in goals.  Everyone stands around while the VAR official (in some comfy studio somewhere) takes the video feed from the game, manually positions lines (cross-hairs) representing the position of the attacker and of the last defender at the time of the ball being played, and determines whether the attacker was indeed ahead of the defender (i.e. offside).

This image has an empty alt attribute; its file name is ings.jpg

Consider the above picture.  Southampton forward Danny Ings (the player in red and white in the upper right of the picture) was judged by VAR to be offside at the time the ball was played toward goal by team-mate Che Adams.  Why?  Because the end of his left sleeve was seen to be (at most) a couple of millimetres ahead of Aston Villa defender Matty Cash’s butt.  Southampton were denied a late equalising goal by an almost imperceptible gap between defender and attacker. 

Here’s another VAR wonder call, involving Sadio Mane of Liverpool. 

This image has an empty alt attribute; its file name is mane.jpg

Mane (at left in red) was deemed offside by VAR, thus denying Liverpool a goal that would have sealed a victory over rivals Everton.  The elements of offside are outlined below, but consider that Mane would be offside only if in the picture he was ahead (to the left) of the Everton player just above him in the picture.  Obvious, isn’t it?

The internet is populated with lots of similar pictures of marginal offside calls.  This retroactive denial of goals for the most miniscule of deemed offenses happens with such frequency, players are hesitant to celebrate goals after close plays, realising they’ll look like proper idiots if the goal is then disallowed by VAR.   In the first full year of VAR (2019/2020), the Premier League saw 34 goals overturned for offside after a VAR review.  The current season promises to see at least that many.  Many of these offside calls are comparable to those in the pictures above – “offside by a toenail,” or “armpit offsides.”  Meantime, linesmen (sorry, “Assistant Referees”) who used to flag offside plays as they happened are now told to keep their flags down until after the goal-scoring opportunity has passed.  Thus, their offside calls are delayed until such a time as they become irrelevant.  And if a goal is scored, the linesman becomes irrelevant in any case, because the offside is decided by the VAR official, not the official on the field.

Precise, but not Accurate

Every time the subject of VAR comes up for discussion, supporters of the technology dismiss the criticism, arguing that “it’s either offside or it’s not,” whether it’s by 1 millimetre or 2 metres.   They defer to the technology.  People like me who scoff at the notion of offside being determined by an official drawing seemingly random lines on a screen are dismissed as Luddites.  So be it.  The truth is, VAR use for offside is precision masquerading as accuracy, and should be shelved.  Or, at least, limited.

Here’s the problem.  For every offside decision, the Assistant Referee or VAR official is expected to consider three elements:

  1. the position of the attacker
  2. the position of the defender
  3. the timing of the pass to the attacker

For each of the three elements, there are challenges to accurate determination. 

For positioning of the attacker and/or defender, one cannot consider the hands or arms, since these parts of the body cannot legally be used to play the ball.  Shoulders are okay. But, the delineation between arm and shoulder for the purposes of determining handball (and by extension, offside) is as vague as this:    

The boundary between the shoulder and the arm is defined as the bottom of the armpit.  International Football Association Board (Ifab), 2020

Bottom of the armpit.  Well, that’s precise, isn’t it?  But, somehow, VAR officials are expected to apply this piece of guidance in positioning VAR cross-hairs to the precise millimetre required.

As for the playing of the ball to the attacker, there are a couple of problems with measurement.  First, it’s not really defined in the rule as to when exactly this happens for the purposes of determining offside.  As Glenn Hoddle (former England player and manager, current pundit) pointed out a few months ago, it’s not really clear whether the precise moment to be considered is when the ball first touches the passer’s foot, or when it leaves the passer’s foot on its way toward the receiving player, or any moment in between.  It’s just not clearly defined.  But, somehow, VAR officials are able to stop the video feed at the same stage of playing the ball to an attacker, with startling precision and consistency.

Second, even if the event of the pass was clearly defined, and consistently applied, we have a problem with the video.  The Premier League boasts that the broadcast cameras used for VAR operate at the equivalent of 50 frames per second.  Fine.  That means that there’s a difference of .02 second between one frame and the next.  Do you know what may happen in .02 of a second? Well, at full speed, a football player may travel up to 13 cm.  And the defender may be moving in the opposite direction during the same period, adding to the difference from frame-to-frame.  In choosing to judge an offside using one frame, as opposed to the frame before or the frame after, VAR officials could be misjudging the relative position of two players by more than 13 cm.  Then, somehow, they feel justified in making a call based on a gap of a couple of millimetres.  Huh?  

So, an offside call based on VAR takes into account a somewhat vague positioning of the required cross-hairs for both defender and attacker, an imprecise determination of “when the ball is played,” and a substantial margin of error depending on which video frame is picked for analysis.  Those of us with any scientific background remember the concept of significant digits, whereby a result can be measured only to the precision of its least precise input.  Here, you have three imprecise inputs used to determine an output that is presented to a ridiculous level of precision.  Precise, but not necessarily accurate. 

So what?

Why bother? Well, some of us subscribe to the view sometimes attributed to Bill Shankly (legendary Liverpool manager) that “football isn’t a matter of life and death; it’s much more important than that.”  Football fans tend to get a bit excited when officials call back goals that seem to be perfectly legitimate, especially when the goal in question could be consequential in gaining a win or draw.  

Yes, in the grand scheme of things, it doesn’t really matter.  It’s a game. Before VAR, referees and linesmen sometimes got things wrong.  After VAR, referees and linesmen and VAR officials sometimes get things wrong.   Life goes on.

But, if you’re going to rely on technology, at least make sure the technology is fit for purpose.  Otherwise, you’re just a technophile, using technology for the sake of technology.  In this case, applying VAR technology doesn’t necessarily produce better decisions – it just provides an electronic scapegoat.  (computer says no).  Why make a tough offside call if you can have a video process do it for you, even if the video process may lead to a wrong result?

And, if the VAR is not necessarily producing accurate decisions, but does interrupt the flow of a game while spurring confusion and anger, then why is it still used to adjudicate decisions well within its margin of error?  Sounds like a clear and obvious misapplication of technology to me.

Next up for commentary – use of VAR for decisions other than offside, where it is supposed to be used only to correct “clear and obvious errors” by the on-field officials.  This could be subtitled “How ‘clear and obvious’ only becomes apparent after 15 views of video replays using 5 different camera angles.”

Leave a comment