Report from one of the sessions at the RSS 2017 Conference. More reports of conference sessions are listed here.
Predicting Olympic and Paralympic medal tables is almost a sport in itself, but once the Games get underway, those tasked with delivering sporting success want to keep a close eye on their team’s performance. During the 2016 Rio Paralympics that job fell to Oliver Summers of UK Sport, who was appointed lead Paralympic analyst for Great Britain.
Speaking at the Royal Statistical Society Conference in Glasgow on Tuesday, he described his work to first calculate medal probabilities for each British competitor in each event, and then – with each recorded win or missed opportunity – to monitor whether the team was on course to hit its target.
ParalympicsGB’s goal was to return from Brazil with 121 medals – one more than it achieved while hosting the London 2012 Games. That might not seem overly ambitious but, as Summers explained, host nations often fail to better their home performance four years’ later.
The 121-medal target was agreed between UK Sport and the British government, which was funding training and preparations for the 2016 Games. Failure to meet this target might well have implications for the funding of future Games, but financial concerns are not the only reason to win lots of medals. Success inspires others, said Summers, and podium finishes give a sport greater prominence in the public eye.
So, the two key questions Summers had to answer were: (1) Is ParalympicsGB on track for 121 medals, and (2) Where would we finish in the medals table?
There were 1600 medals available across 11 days of Paralympic competition. The GB team had 264 athletes competing in 19 out of 22 events, putting a maximum of 349 medals in reach. Summers needed to know how likely it was that each competitor would win a medal in their chosen event(s), so he asked the experts – the coaches who were training the athletes – to report the likelihood of a medal win. Expert knowledge obviously has its advantages: coaches know their sport and their athletes well. But the responses were also subjective, and Summers noted that within the coaching system there were different levels of understanding of probability, which required some work to address.
Summers then had to join up these medal probabilities with the schedule of Paralympic events so the team would know, on any given day, how many medals they were likely to win and – once each day was done – how far they had advanced towards the 121-medal goal.
This information was reported each morning in an email to the team’s leadership to help them answer questions from the media. It was also used by the ParalympicsGB communications team to determine where they needed to be at certain times to stand the best chance of witnessing a medal win.
Ultimately, Great Britain went on to achieve 64 gold medals and 147 medals in total – comfortably above target. Summers was also happy with the performance of his medal predictions, the relative ratings of which were in line with the actual medal total. A similar method and system will be in place for the 2018 Winter Paralympics in South Korea.
- Oliver Summers’ presentation was titled ‘Tracking Medal Success at the Rio 2016 Paralympic Games’.