Football tacticians bowled over by quick-fix data risk of being knocked for six | Football tactics

EEngland won the Second Test against South Africa comfortably, but there was a disappointment before tea on the first day as Kagiso Rabada and Anrich Nortje added 35 for the ninth wicket. With the bowling a bit full earlier, England switched to a short attack to no great effect. Interestingly it was a full-pitched ball from Ollie Robinson after tea that delivered the breakthrough as Nortje was lbw.

So why did England change strategy? Perhaps they were influenced by the Test against India at Lord’s when they successfully got the tail out, or perhaps it was a reaction to the nature of the Dukes’ cricket balls this season which lost their danger quicker than usual, which demands something else from the bowler. But there is also, it seems, data that South Africa’s tail is susceptible to short sound bowling. The problem is that if every ball is pitched short, the batters expect it and can set for it; far more dangerous is the surprise short-pitched ball.

As CricViz analyst Ben Jones said: “You can’t just look at dismissals” – Jimmy Anderson’s inswinger is more dangerous for following a series of outswingers. CricViz’s predicted wickets model shows that good balls tend to take wickets, but Jones acknowledges that context is important and sees that as one of the areas where the use of data needs to improve sport.

Or take the yorker, which no one doubts is the most effective ball in one-day cricket. The problem is that there is a small margin for error: too full and it’s a low full-toss, too short and it’s a half-volley, both very hittable. A batter waiting for a yorker can step forward or step back to change the length.

As taught by Tim Wigmore and Freddie Wilde Cricket 2.0, that is, allied to the suspicion that Ben Stokes would try to hit him over a longer boundary on the leg side, which enabled Carlos Brathwaite to hit those four consecutive sixes to win the 2016 T20 World Cup final. Chris Jordan to Jimmy Neesham who went 23 in the 2021 tournament, too, was the result of the predicted yorker.

Similar problems have plagued data analysis in football almost since its inception. Charles Hughes, the FA’s technical director wrote the book in 1990 The Winning Formula confirmed direct football as official doctrine, drawing his conclusions from the evidence of 109 matches involving “successful sides” – Liverpool, England Under-16s and Under-21s, and World Cup or European matches Championship involving Argentina, Brazil, England, the Netherlands, Italy and West Germany – between 1966 and 1986. He focuses almost entirely on the 202 goals scored in those games – as cricket analysis tends to focus on dismissals – and 87% came from moves of five passes or less. Therefore, he concluded, teams should try to limit moves to five passes or less.

Even leaving aside the shockingly low sample size and the selective nature of the data, there is a lack of nuance. Could it be that what works for England Under-16s in a friendly in the mud and cold of a British winter is not necessarily suitable for Brazil amid the heat and altitude of a World Cup in Mexico?

Hughes added that Brazil are the side most likely to score after long strings of passes, 32% of their goals coming from moves of six passes or more, with West Germany next on 25 %. Having won six of the 13 World Cups played, the obvious conclusion is that possession football looks good for you, but Hughes didn’t pursue it.

Neither he, nor Charles Reep, the novice statistician whose ideas Hughes developed, considered that straight balls could be more effective if they were used sparingly. Just as a batter can set themselves up for repeated short-pitched bowling, or prepare themselves for a string of yorkers, so a defense can drop deep and prepare for a aerial bombardment.

Brentford manager Thomas Frank with his tactics board during their game against Manchester United earlier this season. Photo: Mark Greenwood/IPS/Rex/Shutterstock

Just as the danger of the occasional bouncer can be increased by the surprise factor, with a batter trying to advance having to adjust, so the threat of a long ball can be greater if a defense is caught by a team in possession. (And since almost nothing in sport is absolute, there are times when a batter is too intimidated by the short pitch of the bowling or a defense is overwhelmed by a string of long balls, when the most effective tactic is the stifling pressure of a constant that barrage. .)

Hughes and Reep were, to use the most respectful possible term, pioneers and did almost as much to modern data analysis as Pliny the Elder did to modern medicine. But the issue of context is one with which statisticians continue to struggle.

A coach at a Premier League side told me a story about his manager who was convinced by their data department to run a high line against a team with a remarkably quick forward, despite a first -choice centre-back who had to be replaced by a veteran who had just returned from injury and was not quick to turn even in his prime.

They got three in 30 minutes and lost 3-0, but the analysts justified their advice by pointing to their team winning xG. But that was because, as the coach angrily responded, scoring three early chances, the other team didn’t need to attack. They sat back, conserved energy and were less bothered if they conceded a few half chances: the game ended with an hour to spare. That’s not to say that xG isn’t a very useful tool – it is – just that it doesn’t always provide the full picture.

CricViz’s Jones is clear that data analysis is not enough; it only makes sense when used in conjunction with video analysis by people who understand the limitations of what statistics can tell you. There are some absolute rights and some absolute wrongs and the meaning of everything is partly determined by its relation to everything else. Context is important; players are human. Sport is not an algorithm.