Big data is (very) old news for sportsmen

nz haka2

As the momentum grows for big data in businesses, it is sometimes useful to look outside to other markets to get a sense of what is happening or where the future lies. Sports is often held up as a useful metaphor for business as the results are very visible and therefore the players and coaches have nowhere to hide.

The most obvious examples are Formula1 where the cars stream gigabytes of data every race. More recently the power of big data came to the America’s Cup to both measure the performance of the boats but also to make is easier for spectators to understand what the hell was going on. With complex machinery – Formula1 cars and America’s Cup yachts – you can understand how big data can make a difference and why it has grown in the last year or two.

So it is slightly surprising to learn that sports with just a team of players on the pitch are able to reap the benefits of big data. As the video below shows, the ability to keep the team performing at peak levels during training is amazing. And what they have discovered is that the players gain from the immediate feedback. They have a better understanding of what is expected of them so they can self-train – in the moment.

What is most surprising is that this video was shot in 2010 – over 3 years ago.   Businesses are nowhere near this use of big data. The idea of measuring individuals’ performance real time and helping them improve and perform at their best, or monitoring customer behaviour and adjusting a website dynamically, is a pipe dream. Big data for business has a long way to go, and sports can lead the way.

About these ads

4 thoughts on “Big data is (very) old news for sportsmen

  1. Ian, if I understand what you are suggesting correctly then you are falling for one of the big fallacies of the Big Data movement. Sports won’t lead the way, just like manufacturing processes do not lead the way for human interaction. One can’t translate what you do in real-time to business predictions. The data that are collected in F1 and in other sports are interesting and help to understand what is going on NOW and maybe allow to adjust accordingly in terms of pit stops and such. Yes, also training is better if you analyze performance and blood samples. I train with a Garmin computer that measures heart rate and watts and it helps me a lot to train correctly.

    But none of that is predictive and thats where the huge fallacy lies. For example with SAP trying to predict a Superbowl on data collected from previous games and trends:

    SAP predicted Denver Broncos 26 – Seattle Seahawks 23
    The reality Denver Broncos 8 – Seattle Seahawks 43.

    http://blogs.sap.com/analytics/2014/01/29/super-bowl-analysis-and-game-predictions-with-analytics-from-sap/

    Big Data is only viable if you are spying in individuals and even that does not tell you what this individual will do at certain times. Big data does not allow you classify people to the point that they will behave in a certain way in certain situations. A Gauss distribution is all you get for any value collected and it is an observation only and not a predictor who in what segment will act how.

    Big Data will at some time be renamed to Big Fallacy, but only after business have spent billions in trying to achieve what millions in marketing dollars promised them. There is no proof whatsoever that Big Data prediction works.

    1. Max

      I agree with you view about big data prediction being a big fallacy. But I think the future lies in better instrumentation to help people make real time decisions now. Most people in business make decisions with only a sub-set of the information. I am not suggesting you need ALL the information, but you do need the right sub-set. And companies currently measure what is most easily measured, not what should be measured.

      1. Ian, then we are not too far apart in perspective. But I still caution in Big Data especially when it is supposedly providing more information about customers. A customer is not a data set. It is a person. He does not produce data but a data collection assumption and model is being made that supposedly represents all customers, which is rather ridiculous to not say stupid and ignorant.

        Treat customers as data and all you get is numbers. Treat customers as people and what you get is emotional interaction that tells you exactly what is right and what not. There is actually no measure to manage, there is only measure to verify. Decisions should never be based on data but on gut and experience. Data produce false certainty that can be deadly. Many plane crashes happended due to reliance on false data. When your gut tells you, you can look at data additionally. When the data say something and you gut something else, ignore the data. They are in fact an illusion. The few samples where for example sales data point to the perfect shopping consumer profile for a holiday weekend are sometimes good and as often bad. This long-term averaging out is never properly considered.

        That is particularly funny because Big Data marketing uses mostly anectodal evidence and no long-term Big Data analysis of Big Data use …

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s