Business Analytics Blog

Business Analytics Blog

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

Developments in Big Data and Predictive Analytics helps to win oldest sports prize

One of the greatest sporting come backs of all time took place last month on the San Francisco Bay and it is no coincidence that big data and predictive analytics were at the heart of it. As the prestige and popularity in sport increases so too does the competition to win. Higher stakes has led to more money, technological progression and increasingly big data developments to help gain an advantage over the competition. Businesses should take note.
 

Pre-Big Data
The America’s Cup or “Auld Mug” is the oldest sporting competition in the world. The sailing competition started in Britain in 1851 and involves a home nation (defender) racing a challenging nation for the cup. Initially it was the Americans that challenged Britain for the trophy around the Isle of White but since they won it so often the cup was named the America’s Cup. America’s dominance continued for the next 130 years where it only left America’s shores once. Then in 1995 it was won firstly by New Zealand and then Switzerland. As the competition has become more and more expensive for teams, in more recent times nations have been financed by wealthy team owners. In 2010 it was won again by the Americans with the backing of Larry Ellison, founder and owner of Oracle.

First Ever America's Cup boat


http://www.klaus-kramer.de/Schiff/America_Cup/AMERICA_Cup_1851.html


Super High Tech
Three years on and the most recent challenge last month saw the Americans defending and New Zealand challenging in one of the most anticipated America’s Cups for years. The added intrigue was down to a technological shift. One of the rules of the sport is that both teams use the same type of sailing boat but the defender can specify what type of boat will be raced. Larry Ellison has endless piles of cash and he wanted to go all out. So what can you spend $100m on? In short, speed. Think of it as F1 motorsport on water (minus the engine). In the same way that F1 uses downforce to help cars fly around corners, the boat design uses aerodynamics to propel them 2-3 times faster than the speed of wind. In fact they go faster than the wind speed sailing into the wind! As the picture below shows, take a main sail 131 ft tall; attach it to a 72 ft long foiling catamaran and use aerospace technology to maximise performance. The result sees the boats fly across the water reaching speeds of up to 88 km/h.

http://www.americascup.com/ac72

Oracle Team USA flying across the water


Oracle Team USA website


Big Data / Predictive Analytics
The other similarity to F1 relates to the development of the boat over the course of the competition. In F1, teams put out a car at the start of the season and then constantly develop new parts based on learnings from testing and race analysis over the season. Oracle Team USA took this to a new level during this America’s Cup. The catamarans have more than 300 sensors onboard collecting 3,000 variables over 10 times a second. They measure everything from the speed of the boat and wind to the strain on the mast, angle of the sail, and the many adjustments made by the skipper. They also capture and analyse videos and photos to provide further data points from a visual perspective. All data is stored in a server at the back of the race catamaran and sent real time to an Oracle database on their support boat for analysis. The data is processed using a variety of different programs that use complex algorithms to mine combinations of variables over time. Performance (boat speed) is assessed against predictions under similar external conditions (eg: wind direction, speed and tidal flow) with the performance gap assessed against internal (changeable) inputs to determine how to improve performance from these inputs. Machine learning is at the heart of the approach. The data is constantly refreshed and new predictions created off the back of further performance gap changes. Although it would be possible to use the insights whilst the race is in play by relaying technical decisions to the skipper, race rules stipulated that data insights could be not be used during the race. Instead Oracle Team USA improved the speed of their boat in between races and during post-race over-night sessions prior to the following day’s racing.

http://www.oracle.com/technetwork/issue-archive/2013/13-sep/o53oracle-teamusa-1996101.html


Failure to keep up with the competition
New Zealand hit the ground (or water) running. In order to challenge for the America’s Cup the challenger has to race against all other prospective challengers first in order to win the right to compete in the America’s Cup. The benefit of this is that the challenger gets additional race time to practice (as well as gain more data insights to develop the boat). At the start of the America’s Cup, New Zealand weren’t considered favourites but quickly sailed to an impressive 8-1 lead thanks partly to their learnings from the pre-race competition. The 2013 America’s Cup was a best-of-17 race so Team NZ only needed one to win at this point (since the first team to 9 wins the cup). New Zealand certainly had the momentum at this point and it seemed a matter of ‘when not if’ they would take it out. The Americans needed to make developments to their boat and fast. They were helped by having a few days without racing due to extreme weather conditions and also played the joker card to postpone a race. This was the analysis time they needed in order to make a better boat. Crunching the numbers to develop a faster boat (particularly in the upwind leg where they had struggled) made them more competitive with Team NZ. They won a couple of close races, developed further and won a couple more. By the end of the competition they were practically the same distance ahead of Team NZ at the finish line as Team NZ had been from them during the initial races.

Whilst Team New Zealand created a formidable boat, their inability to assess performance data as effectively as Oracle during the competition in order to improve the boat cost them dearly in the end. Whilst this wasn’t the only reason for the turnaround in fortunes (other factors such as the introduction of Sir Ben Ainslie and growing confidence from initial race wins definitely helped), the effective use of data throughout the competition helped them to foil better and make the necessary adjustments to win. There is no coincidence between Team USA’s successful developments and having the 3rd largest software company backing your effort. Alas, that’s the opinion of at least one Team NZ fan!

In the same way that developments in F1 led to the introduction of steel disk brakes and active suspension in road cars, developments in the use of big data in sports such as The America’s Cup can help businesses understand how to compete more effectively in the future. And just because you’re 8-1 up or eight times bigger than your competitors doesn’t mean that the business may not need to change tack.

About the author

Jonathan Chadwick
Jonathan Chadwick
Jon has worked for 18 years as an analytical consultant in the UK, USA and Europe for a diverse range of sectors, most recently Financial, Oil & Gas and Government. Jon has extensive experience in benefits realisation, modelling, business analytics, portfolio management and change management. Jon devised and created Figure It Out.

Connect with me

Leave a comment

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