Many industry professionals in sport assert that there are few players left to be discovered. Some argue that commoditized access to data has eliminated the information moat of identifying-and-recruiting a talent that the market has not yet priced in. Oftentimes, this view is expressed to justify spending above market-rates for a talent considered to be elite; otherwise, it attempts to discount the value of investing in scouting infrastructure (data, information, and personnel).
This is a dangerous posture to hold in the era of Machine Intelligence (MI).
MI promises to make perfect, or nigh perfect, any existing processes for combining data towards an end. Moats that existed in the machinery, the engineering, or the system itself will be greatly diminished. Ultimately, there will be no gap between the data being collected from reality and the outcomes from it.
The concept of an analytics department that processes data will be foregone, just as computer vision increasingly diminishes human-counters who tally actions during a game.
What will remain in this era are relationship-builders (scouts and club executives who demonstrate interest and convince talents to join their team), and information-conduits (multi-hyphenate individuals who can extract and suggest new kinds of records for MI to collect and integrate into its math).
This future will have no human Data Analysts in sport since self-improving, intelligent automations will comprehensively cover this. It will likely have no Data Scientists nor Data Engineers – both roles will eventually merge into a hybrid Data Architect that will co-design the system with MI.
Because MI yields nigh infinite engineering capacity, data about the world and the human body (biomechanics), which could previously not be collected at scale, will now be.
The scouting moats that remain will be access to frontier MI, deep-integrations into an club's workflow, and the personnel (relationship-builders and information-conduits) they employ.