Analysis Works – Where is the Proof?

The most recent edition of the International Journal of Performance Analysis has just been published. This is a much needed journal and I have even had a paper published myself. One thing that stands out to me is that the vast majority of the papers look at ‘performance indicators’ rather than looking at the science of Performance Analysis. That may be the remit of the journal, I don’t know, but I do feel there is a need to look at the science of Performance Analysis.

Top Secret

It’s safe to assume that in a lot of cases the most cutting edge of analysis is not going to be published, there is very little upside for a club to allow analysts or sport science staff to publish papers. For minority sports academic papers might offer the only opportunity for good detailed statistical analysis to be performed but that is not necessarily the case in all sports where data companies can employ there own data analysts and create proprietary products & services.


There is a much greater need to educate coaches and analysts on how best to deliver the KPI’s. What is the best format of presentations, how long should they last, should they always contain a positive message? What’s the best way to present data, how important is colour, visuals and the ability to explore the data or be guided to the answer.

These issues need to be explored as much as KPI analysis. Ultimately if you can’t get your analysis into a team talk or onto the training pitch what is the point?

Current Research:

Ryan Groom et al., is someone who has published in this area;

A very nice collection of papers that look at the science of Performance Analysis and feedback within a sports context. Two other papers which I have found very interesting are published in the Journal of Physiology and Behaviour;

Recently I stumbled across the excellent Perception and Action podcast. The podcast is hosted by Professor Rob Gray.

Rob Gray’s work illuminates the dynamics of perception, cognition, and action in skilled performance.

In terms of making science accessible, understandable and ultimately actionable Rob Gray does a tremendous job and the podcast goes above and beyond simply publishing a PDF of his research. The podcast has brought to light research that is very applicable in the world of Performance Analysis. One paper in particular intrigued me;

*The Moneyball Problem: What is the best way to present situational statistics to an athlete.

This paper which is yet to be published looks at the presentation of situational probabilities in baseball. It is very difficult to summarise the entire paper here but a few things that really stood out for me;


1) despite there wide-spread use ‘” review of the literature revealed that there are no studies that have demonstrated that the use of situational probability information has a direct, beneficial effect on performance indicators”. 

2) The paper found that “Providing situational probability information gives a significant advantage to a batter, however, there is a trade-off (between short-term effectiveness and negative transfer) which depends on how the information is presented.”

3) “Providing situational probabilities in an accrued manner led to, through a process of guided discovery, not only information that could be used to generate action plans against the current opponent, but also helped develop their ability to generate their own event profiles when explicit situational information is not available.” This strikes me as give a man a fish feed him for a day teach a man to fish feed him for a lifetime.!

It’s also not dissimilar to this quote from a paper published in 1996

These finding provide empirical support for the Guidance Hypothesis, which suggests that immediate performance is facilitated because the subject is guided towards the target performance by the feedback, but that long-term retention (i.e. learning) is degraded because the athlete will rely on these guidance properties to perform correctly. (More & Franks, 1996).

Yes this paper is from baseball and situation probabilities are more applicable in 1 v 1 sports than invasion games – but the principle remains. What is the best method of feedback of all this data and video we are collecting?

We need more of these papers, whether they come from the Performance Analysis world or others, this type of research is vital, especially as technological changes make data and video visualization more accessible and powerful.

If I have missed any paper do send them my way.

* Proceedings of the Human Factors and Ergonomics Society 2015. In Press. 

Add Comment