Raheem Sterling and the history of measurement
How certain can we be about close offside calls?
 |
We take rulers, clocks and thermometers for granted, but the history of these inventions is anything but straightforward.
In 1792, Pierre Méchain and Jean Baptiste Joseph Delambre set out to establish the metre by measuring the distance between Dunkirk and Barcelona. In 1772, Jean-André de Luc wanted to establish the temperature at which water boiled. To do so, he needed a sample of water with no air in it, and so he walked around for four weeks straight shaking a flask of water to expel all the air from it. John Harrison spent most of the 1740s grinding ever more precise metal wheels to make his watches work.
I often wonder what these men’s friends and families thought of their obsessions. But of course, from a distance, we can appreciate their visions because we depend on them to tell the time, to boil a kettle, to make journeys, and of course to work out whether Raheem Sterling is offside or not.
The first VAR offside controversy in the Premier League was in the very first match it was used in the Premier League: West Ham vs Manchester City in August 2019. I was at this match but, as is typically the case with VAR, it was only when I got home later and watched the highlights that I realised what had happened.
City had played a very slick and fast passing move, with David Silva flicking the ball on to Raheem Sterling, who then squared it for Gabriel Jesus to score. However, the VAR replay showed that when Silva played his pass, Sterling was fractionally offside, and the goal was disallowed.
To decide whether he was offside or not, the referee froze the video replay at the moment the ball was played to Sterling. This is the moment, on the replay, that David Silva's boot first makes contact with the ball. At this point, Sterling was deemed to be offside.
The relevant measurement issue is that we can't be totally sure at what point David Silva did first touch the ball. The footage used by the referees takes 50 frames per second: that is, one frame every 0.02 seconds. The referee has to pause the footage on the first frame that shows Silva's boot hitting the ball, but in reality, Silva's boot might have hit the ball somewhere in between two frames.
Raheem Sterling is very quick. His top speed is just over 20mph. He wasn't going at top speed for this goal - let's say he was going at just 10mph. That's about 4.5 metres per second, which is about 9 centimetres every 0.02 seconds. So in our 0.02 seconds of uncertainty, where we are not sure if Silva has touched the ball or not, Sterling has moved about 9 centimetres. And yet he was deemed to have been offside by 2.4 centimetres.
Hypothetically, this situation could be even worse. Imagine a situation where an extremely quick defender and attacker are both moving at top speed in opposite directions. In this case, in our 0.02 second window of uncertainty the defender is moving about 18 centimetres in one direction and the attacker 18 centimetres in the other.
The obvious riposte to all of these measurement issues is: what would you do instead? No measurement on earth is completely accurate. All measurements have some error. The reason VAR was introduced was because the error made by humans judging using the naked eye were very much greater than +/- 9 centimetres. Back in the 2013-14 season, when Raheem Sterling was playing for Liverpool against Manchester City, he had a goal ruled out for offside despite replays showing daylight in between him and the last defender.
Ultimately, the offside rule is fiendishly difficult to measure. It was first introduced in the 1860s when we had far fewer sophisticated measuring tools than today; in fact, the offside rule is about a decade older than the international agreement that defined the metre.
But what we can see with offside, as in many other fields of measurement, is that as your measurement tools get more accurate, they make the remaining inaccuracies even more noticeable. The quest for greater accuracy often ends not with perfect accuracy, but with a resigned acceptance of uncertainty. Was Sterling onside or offside when Silva played the ball? We will never know for sure.
Totally agree with your final paragraph - which neatly explains why I think critics who use VAR's imperfection as a prime reason to scrap it are wrong. It's not about perfection, it's about getting very close to it.
The argument around VAR use for offsides seems to basically revolve around the entirely subjective question of whether you believe the stoppage in time and decreased certainty of celebrations is worth it for a decision-making process that still ends with a decision that is not 100% sure to be the correct one. In my opinion it is, because as well as getting us a bit closer to the correct decision, that stoppage also provides a drama which - at least for TV viewers (much work is needed on the in-stadium experience) - is in and of itself quite entertaining. While you also get a decision which you know to be as close to correct as we are currently capable of getting (watching crap decision in non-VAR games now feels like robbery).
But I take it you disagree with this view?
I agree with you, Daisy, that the tradeoffs involved do not (yet) justify the use of VAR in football the way that they do justify the use of DRS in cricket. The big philosophical question for me is the degree to which technical precision is required for, or consistent with, a sense of fair play. The timeless philosophy of the Laws of the Game is that they are meant to promote fairness consistent with the spirit of the game. I am skeptical that we ever will have technology capable of providing unambiguously accurate decisions with 100% precision, or even that such theoretically perfect precision would guarantee fair play.
Offside is more conceivably flattened into objective fact than most other football laws. But would the spirt of the law and of the game be best served by a robotic system capable of measuring the relative position of ball and body parts down to the scale of the millimeter and the millisecond? We do not play the game at that speed and scale, nor do we observe the game at that speed and scale with our naked eyes. Led by the presence of VAR, the revised Laws now precisely define when the offside decision applies: at the first moment of contact. As you note, players cover a lot of ground very quickly, and during the full unpaused moment of a ball being played—which may stretch across a large fraction of a second as one’s leg/foot swings through the ball—someone can go from a technically onside to a technically offside position, or vice versa. Without the application of freeze-frame video review, I would describe this as being level with the ball, and attentive, well-positioned linesmen of the past presumably would have as well. If the flag nonetheless was raised in such an instance, there would be plenty of moaning by players and fans with the attacking side, to be sure, but this would not be deemed as unfair by anyone but the most partisan of observers. It's a marginal call, not a howler—a coin-flip decision well within the overall spirit of the game.
Freezing action down to whatever the frame rate of the VAR camera allows seems so artificial to me as to introduce its own kind of unfairness. The assistant referee's eyes may be less precise than the camera, but at least the AR's brain is operating at the same speed as the players. I think the spirit of the offside law, and the broader spirit of the game, would be better served by a DRS-style margin of ambiguity that uses technology to eliminate the howlers and leaves the marginal decisions to referees. Of course, as we see in DRS, such decisions often are not seen as fair, particularly in the eyes of partisans. But given the choice, I prefer this traditional sense of unfairness—instant judgment by a presumably neutral human observer—to the hegemony of machines and the arbitrary algorithms they are programmed to apply.