A Lesson In Analytics From The World Of Dance
An Article With Great Transition But Likely Not Much Musicality
For all my avid readers, it is likely no shock that I can find an analytic lesson in the World of Dance. It may come as a shock that I watch World of Dance. It sometimes shocks me… enough to give me the “goosies”. Cringe.
World of Dance is in its third season and as the father of a daughter who loves dance, I have been watching all along. It is a great show for daddy-daughter bonding which is otherwise hard to come by in their teenage years. That said, I won’t pretend I don’t enjoy it.
The third season just wrapped it’s Qualifier round. Various dance acts from around the world competed within three groups and were scored along five categories. In the qualifiers, the divisions really aren’t very important as acts are judged separately.
Each act can receive up to 20 points in each of the five categories. This makes the maximum possible score 100 points. Much was made this season that the minimum score to qualify was raised from 80 to 85. But the beat has slowed — so let me transition back to my viewing habits.
Once again, for the avid reader, it will be no shock that I add to my viewing enjoyment by predicting the contestants scores. Well, actually, while I enjoy the performances — I hardly qualify as a dance expert.
I actually predict the judges scores. I listen to their comments and then while they are ‘locking in’ — I announce my prediction to my daughter. I am rarely off by more than a single point. Predicting their scores is easy. Although it has changed a bit this season… but perhaps not how you might have expected.
In prior seasons, we enjoyed a second game. In college, it would have been a drinking game. In the office, they would call it boardroom or buzzword bingo. Yes — World Of Dance has it, too. For the past two seasons, it was headlined by the word musicality. This season it has been much more about transition. These are certainly not the only terms used by the panel, just the most popular.
But let’s transition back to my claim. I can predict the judges scores within a rounding error. You may or may not believe me. You may or may not be impressed. What matters — is how.
It is a simple process really. I start with 85. I then simply listen to the judges tone. I weight criticism, praise, and buzzwords. It is my own three category subjective scoring system and it works every time. If I could remove all personal bias, it might be flawless.
But we have arrived at our lesson. I give credit to WoD, using a five category point system is more objective than a simple ten point scale or an up-and-down vote. However, it is still a subjective system.
What’s more — all this hype about ‘raising the bar’ by requiring an 85 this season to qualify, is complete silliness. You can back that with science and cognitive bias. But you can also confirm it, if by chance you have spent three seasons forecasting scores and realize that all you did to adjust was change your starting point from 80 to 85.
This lesson applies even more clearly in the next round of the competition — the duels. Here those categories become less meaningful and the scoring is much more easily seen as comparative. Essentially, all objectivity is tossed aside in the next round. Here a simple up-or-down vote would suffice — though it would make for a very rough transition. I only hope that the producers don’t see the need to hype some new and more irrelevant scoring change.
If you are seeking more proof, go back and look at the scores given to each contestant. They are not easy to find, but they are out there. Note the clustering of scores in prior years between 77–79.7 vs this seasons 82–84.7. Or just note the minimum score for each season. In the area of science known as cognitive bias — this is called anchoring. It is a form of subconscious benchmark. For more on those:
Bringing this routine to an end. The World of Dance is very entertaining, but it can also teach us a lesson. Categorized scoring is more objective than other systems because of the benchmarks it provides, but raising the minimum overall score to qualify is a highly subjective farce. One that provides us easy evidence of the cognitive bias of anchoring. Like most things in analytics, benchmarks work in both directions.
There is probably a much deeper set of lessons as well. We could dig into the nature of art vs science. We could dig into the strong connections between music, dance, and mathematics. But all of that would require a much longer article. Or a different one entirely:
Thanks for reading!
