PwC used big data and machine learning to pick a Melbourne Cup winner — here’s what it said

‘I’d like a box trifecta please’. Photo by Frazer Harrison/Getty Images (file)

Artificial intelligence is hot right now. The data boffins at PwC, naturally, applied it to picking some Melbourne Cup winners.

PwC explains: “Algorithms, including decision trees and their ensemble counter-parts use historical race results to ‘learn’ the complex relationship between a horse’s characteristics (weight, age, trainer etc.) and its most-likely finishing place or probability of winning.”

Who knows what decision trees and the ensemble counter-parts are of an algorithm, but basically a robot is doing the picking.

And the result: Fame Game, followed by Preferment, Trip to Paris, Almoonqith, and Who Shot Thebarman. Here’s the chart:


They’ve run some other models and you can see them all here.

If a robot can pick this it will either ruin all the fun or make a small number of secretive people insanely rich.

See also: