What Do Golf And Data Science Have In Common?

Posted 2019-05-02 Posted by Deepak

Hello world! My name is Deeps and I am the community manager and a data scientist here at Pivigo. For those of you who know me, you will know I am a huge sports fan and an avid golfer. Having been in the field (of both data science and golf) for near-on two years, I have noticed some surprising similarities between golf and data science.

Most club-level golfers play or are members of one particular course, and these courses consist of 18 different holes, each hole consisting of different difficulties. The pin (where the golfer is trying to get the ball to) is cut into an area called ‘the green’ for each hole. Where the pin is located dictates the strategy for that hole. But how does this relate to data science?

Data science projects are like golf holes where the pin represents the success criteria for a company on that project. You might play each hole many times over, however, your strategy for playing each hole depends on where the pin is cut. This is similar to data science where you may work on several projects of the same kind (e.g. recommendation systems), but the approach will change depending on the success criteria for the company.

The conditions of the day can also affect how you play each hole; just as with data science projects, the conditions within the company you are working for affect your methodology. In golf, playing against high winds requires you to use a less-lofted club to keep the ball under the wind. For data science projects this is akin to getting buy-in from key stakeholders by explaining the benefit to the business.

The difficulty of a hole is measured in two ways. The par, which is the number of strokes expected to get the ball into the hole, and the stroke index, the relative difficulty of the hole to the other holes on the course. Some data science projects are more straightforward than others, so it’s important to manage the expectations of your stakeholders. So, you could think of a par 3, stroke index 15 (relatively easy hole) as a traditional primary classifier on tabular data and a par 5, stroke index 1 as a computer vision project to recognise specific brands, products or people in large volumes of image files.

Most amateur golfers will have a handicap, which is the number of strokes that are taken off your gross score at the end of a round. Beginner golfers have a high handicap, that is that they get many shots taken off their final score. More adept golfers have a low handicap; some so low that they actually need to add shots on to their score. This is similar to data science experience levels. The lower your handicap, the more difficult a golf course you can play, and you may become a touring professional. The more data science experience you have the more complex the projects you can work on.

So, think of golf and data science as parallels; two disciplines that with practice, dedication, and experience, allow you to take on courses and projects of increasing complexity as you progress.

Get priority access to Pivigo news, features, events and networking opportunities

TOP