Business Analytics Blog

Business Analytics Blog

Opinions expressed on this blog reflect the writer’s views and not the position of the Capgemini Group

Forecasting a Successful Olympics

Category : Sport

This is the final in our four-part Figure It Out series showcasing the potential of Advanced Business Analytics. The theme has been next summer’s much awaited London 2012 Olympics, and each week we have also featured a new cartoon focusing on a specific analytic technique.

The earlier FiOs in this series, Olympic Tickets - Were you in the winning segment? and Olympic Tickets Bought - How Do I Get There?, and Mining Olympic Data Gold looked, respectively at the roles of Segmentation, Simulation, and Data Mining in providing deep insight for organisations.

This week we take a look at Predictive Analytics to see how it can help to find out what could happen in the future by predicting future probabilities and trends, and find relationships. We’ll try and answer the big question - who could win the headline event, the 100m final?

Thanks, once again, to our ‘Mistry’ Cartoonist we can show you “what the OR team can do for you” if ever you need a helping OR hand to realise the benefits of predictive analytics.



Predictive Analytics is the branch of analytics concerned with the prediction of future probabilities and trends. Multiple predictors are combined into a predictive model, which can be used to forecast future probabilities with a statistically acceptable level of reliability. This usually involves high-end statistical techniques such as linear and logistic regression, computational mathematics and artificial intelligence in addition to data mining.

 

Predicting who you should place your bets on for the 100m final?

Like millions of other people out there I was not lucky enough to get tickets to the 100m final, but I will be watching it and expecting Usain Bolt to run a world record time. Bolt is most people’s favourite but how likely is to that he will actually retain his title?

Predicting if Bolt can win the final depends on a number of factors including his form, the level of competition he is facing, fatigue from the other events he is likely to be competing in. By looking at the previous modern Olympics 100m finals we can start to predict what the impact of these kind of factors might be.

The winners of each previous modern Olympics final are shown in the table below:



At first glance it looks bad for Bolt; only one person, the great Carl Lewis, has ever retained their title or won the final more than once. However, Bolt’s performances have been so spectacular and he is so far ahead of most the competition, he has to win doesn’t he?

 

So who else could win the 100m final? First let’s see what kind of time we would predict the final being won in. To do this we look at the historical data to identify trends in the winning times and evaluate an estimate that best fits this trend and reduces are chances of being wrong.



The simplest method of forecasting involves drawing a straight line through the historical data and extending this line into the future. If we did this we would get a winning time of 9.59 seconds. However in this case, the graph shows us that the reduction in winning times is clearly not linear as the rate at which the winning time is decreasing is itself getting smaller. So it is more appropriate to fit a curved (logarithmic) line to this data. This revised approach gives us a winning time of 9.62 seconds. Bolt, with his world record time of 9.58, is clearly capable of matching this incredibly quick time; but who else can match him?

 

The chart below shows the how the big names have competed in the Diamond League over the two and a half athletics seasons. The Diamond League is the top athletics competition outside of the Olympics and World Championships so should provide a good indicator of the form of the big names.



Bolt has not made as many appearances as in the Diamond League 100m as his major competitors, Gay or Powell. He has chosen to concentrate on other distances such as the 200m and even the 400m, but he has won on 6 of the 7 occasions he has chosen to compete. However, a closer look at the stats revels he may have some competition on his hands.

 

Tyson Gay has won on 10 of the 12 occasions he has appeared in a Diamond League 100m event, recently beat Bolt at this year’s New York Grand Prix. Gay has also recorded the fastest time of any of the athletes in the Diamond League, at 9.69 seconds.

The final athlete to consider is Asafa Powell who has won his fair share of the Diamond League events. He has won on 8 of the 17 occasions on which he has appeared and also has the second fastest time recorded in the Diamond League at 9.73 seconds.

Between them Bolt, Powell and Gay have won 22 of the 24 races in which one or more of the three of them has competed so it would take a risky gambler to put their money on anyone outside of these three.

Over the past couple of years Gay has been running the fastest times of the three men and he has been performing best around August to September, the period in which the Olympics will be held. So while most people will put their money on Bolt, shortening his odds, Gay might yet turn out to be the better bet.

As this article was being written it was revealed Gay is injured for the rest of this season. This will undoubtedly damage his chances and proves how unpredictable and unfair the world of sport is. However, one thing you will be able to count on is that the Olympic Stadium will be packed of people waiting to see who is the fastest man on Earth.

About the author

Nigel Lewis
Nigel Lewis
Nigel leads the Capgemini Consulting’s 35 strong Business Analytics team, which delivers analytical, operational and strategic modelling solutions to clients. He has 18 years consultancy experience as well as 8 years experience in the UK gas industry. Nigel has successfully managed complex projects in both the public and private sector, including capacity modelling, simulating supply chain operations, strategic business modelling to support future policy decisions, and implementing complex demand forecasting systems. Nigel is currently focussing on the development of Capgemini’s customer analytics and analytics advisory services.

Leave a comment

Your email address will not be published. Required fields are marked *.