Data Analysis – it’s just not cricket: or is it?

Ian Welch argues that England’s obsession with data has proved costly.

It has not been a good week to be an England cricket fan living in Sydney. Our dismal exit from the World Cup being played on these shores achieved the seemingly impossible – failing to meet a bar that had already been set depressingly low. Its not been this bad since. . . oh just over a year ago when we were thumped 5-0 in the Ashes.That disaster saw a clean sweep of the England hierarchy and a new coach Peter Moores was brought in – all bright-eyed and bushy tailed and having the aura of one of those PE teachers you dreaded at school, forever beaming while calling for another lap of the field and gazing at his watch timing you.

England fans’ expectations were low going into this competition. Despite ruining the last Ashes schedule by forcing two series together in order to give us more time to prepare for this World Cup, results were poor and team selection confused. Our new, inexperienced captain Morgan couldn’t buy a run and our one decent hitter Buttler was wasted at no 7.

But worse than these selection mistakes was the overpowering sense that the new set-up had been enslaved by the cult of the laptop. No interview with player or coach appeared without some reference to what ‘the data’ had found.

Defeat after defeat were brushed off as we became entrapped in analysis and stats about the pitch, the opposition, the par score at that ground etc etc. Field placings and lines of attack were being seemingly directed on the basis not of instinct and experience but on pre-match plans directed by what the mystical data had found.

Respected commentators suggested this over-reliance on analysis was detrimental and restricting independence of thought and freedom of action. The set-up was becoming overwhelmed, not assisted, by the streams of data being channeled at players who seemed scared to try anything original. What was intended as a helpful tool had become a crutch to lean on.

The situation reached its nadir after Monday night’s defeat by minnows Bangladesh, which saw England exit without even reaching the quarter-finals, the minimum expected of any major nation. Facing the press afterwards, Moores lamely stated that a target of 275 had seemed gettable but – wait for it – he’d have to see what the data said.

At this point England fans lost it. As social media exploded at Moores’ inane comments one fan summed up my feelings exactly, ‘Stone me, he really did say that. I thought it was a spoof!’

Now as KPMG constantly shows, effective capturing and use of data is crucial in business. Data mining and application of data and analytics techniques, whether in the finance function, auditing or GST work is transforming companies’ operations, and improving value on an unprecedented scale. And evidence-based research is crucial to effective policy making.

But when used poorly or in inappropriate environments the search for data can just end up confusing everyone and being counterproductive. And while professional team sport may benefit from some performance data it should not be enslaved by it. There is such a thing as gut feel, inspirational leadership and players being left to back their own instincts and judgement.

Ahead of yet another Ashes series in July, England will no doubt undertake a new round of deck chair-rearranging and internal reviews. We are certainly world champions at that.

Anthony Mason replies that data is instrumental to modern sport.

For all of the discussion about England’s frailties in the 50 over format leading into the Cricket World Cup, not even the most prescient pundit would have actually professed a genuine belief that England would lose to fringe nation Bangladesh and be carted out of the tournament before the quarter-finals.

It is no secret that the current coaching regime, headed by Peter Moores, uses data analytics to learn more about their teams and their opponents, select playing Xis, and to dictate on-field strategies. This does not make England unique, however, as professional teams of all sports are now sophisticated users of data – the only differences were the unabashed publicity of it, and potentially a misapplication of data which relied on poor and stubborn architecture.

In a sporting context, where the success or failure is contingent on the performance of a handful of individuals, data can sap the sense of personal responsibility.

If the data says a side should win on probabilities from a certain point, will that not seed fear and nervous caution? If the data says the side should lose, will that not excuse mediocrity?

Data can also add a layer to the already complex equations running through a player’s mind as they prepare for a fast bowler to send in a thunderbolt at their toes or throat. Sports psychologists talk about ‘being in the moment’, and complex mental equations can cloud the ‘see ball, hit ball’ mental state.

Worse still, instructions based on data can be counterintuitive if that data cannot move with the match situation. We all know that 300 is a good score, but is it what a batting team should aim for if they are 3 for 240 after 40 overs?

But data is instrumental to contemporary sport, the demise of tennis’ serve and volley archetype, the rise of football’s counter-attack and rugby league’s dubious ‘wrestling coaches’ all have a basis in data analysis. Even the great Shaquille O’Neal was often deliberately fouled by NBA opposition because of his ineptitude from the free throw line; averaging a petty 52.7 percent across his career. The NBA average is about 75 percent since attempted free-throw data was first recorded in 1958. Shaq missed 5,300 free throws in his career.

Brad Pitt’s 2011 film Moneyball is an incisive look into substance over style in the sporting arena. As General Manager of the struggling and underfunded Oakland Athletics he, before their 2002 season, enlisted the support of a young maths wiz, portrayed by the bumbling and awkward Jonah Hill, to dictate new signings, trades and line-up selections. All the while, the belligerent coach, played by the late Phillip Seymour Hoffman, was the data sceptic who believed in ‘momentum and feel’, and would argue that games are played in the real world, even muttering “you’re killing this team” to his superior. Needless to say, the clever and sometimes callous data-driven approach took the side to a record 20 match winning streak.

The root of the question is to what extent we let data guide us. Data is useful only if the right questions are asked of it – and no excel spreadsheet can tell the difference. Moores’ comments point to him looking to data as if it were an oracle or clairvoyant, and not as a spotlight or reporting mechanism on certain strategies or actions, which is its proper application.

Data is dangerous, as England found out – but it is still the evidence upon which change can and should be made.


One thought on “Data Analysis – it’s just not cricket: or is it?

Add a comment