Rise of the stats game

The GAA’s National Games Development Conference in 2007 featured a module on Performance Analysis. It was an early example of the developing importance of the subject and the growing prevalence of research and data as a tool for diagnosing problems in your own team and areas of opportunity in the opposition.

It makes a great deal of sense and has been in operation within the top teams in recent years: for instance the half-time statistical read-outs that give instant indicators of where a match is going wrong.

Of course if problems are abundantly clear no-one needs read-outs but as former Down manager Peter McGrath said in the Performance Analysis module, it's not possible for managers to analyse matches adequately in real time and statistics can provide useful, accelerated diagnosis.

In the same session Christy O’Connor, who has pioneered games analysis journalism, was careful to point out that statistics have their limitations in that they’re not a substitute for talent or coaching or tactical acumen but he also produced some telling examples of how they can be constructively used.

One stood out in the memory: the Armagh experience in the 2002 All-Ireland final. O'Connor drew attention to how in-match data had drawn Armagh's attention to the fact that they had only shaded the breaking ball statistics in the middle third over Kerry – an area in which they would have regarded themselves as strong.

In the second-half, Armagh won 71% of the breaking ball. In his presentation O’Connor also pointed out that the purpose of this analysis is to provide performance goals and outcome goals.

If the performance issue for Armagh had been breaking ball in the middle of the field, the outcome was also starkly apparent.

In the first 40 minutes of that match, Kerry corner forwards Michael Francis Russell and Colm Cooper had the ball in their hands 24 times; in the last 32 minutes, they had it in their hands only three times.

That’s all of 12 years ago. In the meantime the depth of the analysis has intensified and has been driven by advancing technology.

Ray Boyne has been closely associated with establishing and directing this area for the Dublin senior footballers in the past 10 years as their Head of Analysis until stepping down after last year's championship.

He has seen perceptions of the techniques evolve from idiosyncrasy into accepted orthodoxy.

“I would say that all county teams and - based on my dealings with providers of analysis software - an awful lot of club teams are using statistical data. We started using it in 2005 and were one of the first counties – Tyrone and Armagh were probably already there – to make use of proper statistics and analytics.”

Individual critiques
One of the issues for management in introducing games analysis is the impact on players, whose performances are individually critiqued.

Speaking now, O’Connor says that this is one of the great advantages of the process. “The classic value is in a situation where Johnny comes in at half-time and the manager roars him out of it for his performance.

“Johnny goes into a sulk but you can’t argue with statistics detailing missed tackles, failure to make runs etc . . .”

This antipathy to being the subject of analytical focus is the other side of a coin referred to by O’Connor.

“When an individual performance is not properly identified and measured in a team performance, often the team’s performance decreases because there’s a diffusion or abdication of responsibility. They slack off, ease off, socially loaf.

“One sport psychologist called NL Kerr is very blunt about the phenomenon, saying it creates two kinds of competitors – the ‘sucker’ who will do most of the work, and the ‘free riders’ who will do less work for the same approval because they’ll get the same share of the credit.

“If athletes believe their own performance within the team can be identified and that they will be held accountable, then they will not slack.”

According to Boyne this was initially an issue for ordinary players but those at elite levels appreciated the performance indicators as a basis for improvement.

“Ten years ago it was hugely resisted at club level but from the start a hunger for it at county level. Guys were very interested in getting that data and information although as it developed the sessions became a little more brutal.

“You’re identifying defensive plays or positional plays where corrections can be made and there’s no easy way to do that. This could get too brutal at times whereas now individual scrutiny and analysis is associated with a secure environment.”

That is, however, as an ongoing tool for the improvement of performance and eradication of errors.

As can be seen from the 2002 All-Ireland example, one of the most urgent applications is to identify trouble during a match and more importantly come up with the solution.“It’s now so advanced that you’re in constant contact with the line,” says Boyne. “In the past things like shot to score ratios, wides, drop-shorts were the sort of statistics produced but that’s not acceptable any more.

“You need to identify trends and know after a couple of misses or lost kick-outs what’s happening and provide creative analysis. Pitch diagrams and heat maps show where you’re in trouble and allow you work on remedies.

“Last week I was with a team and we were able to identify the target area of opposition movement – how they were moving the ball from back to front – and realise it was because we were leaving gaps.”

With the championship about to get underway in earnest, the same thrill and traditional rivalries will be the same as ever but the operation of cool appraisals by data gatherers with i-Pads is also evidence that beneath the surface all things also change.