A communication and deployment strategy for introducing analytics into a sports organization: start with simple descriptive statistics that accurately name what happened in the game, make sure they are well-defined and discrete (A is A, not B), then build toward complex models by stacking understood building blocks. The key insight from linguistics: language has six properties — arbitrariness, cultural transmission, discreteness, displacement, duality, and productivity. Analytics metrics must be culturally transmitted (taught, not assumed), discrete (clearly separated from similar metrics), and productive (combinable into larger meanings). The more complex something becomes, the harder it is for non-specialists to follow — so build the shared vocabulary first.
(1) Start with descriptive stats: define well-named, well-bounded counting stats that describe what happened in a game in terms a coach would recognize. These are the atoms. (2) Ensure discreteness: each metric must mean one thing, not two. If "superstar" can mean top-5 or top-15, the word is useless — define it precisely. (3) Name things well: put the calculation in the name when possible (xG = expected goals; WHIP = walks + hits per inning pitched). Avoid opaque names (Corsi, Fenwick, PDO) and proprietary "roll-up grades" that promise to solve everything. (4) Build incrementally: once the basic vocabulary is shared, combine building blocks into more complex models. The underlying complexity doesn't change, but the apparent complexity to the audience decreases because they understand the components. (5) Budget 3+ years: this is a multi-year project. Each building block takes months to define, validate, and get buy-in.
Analytics adoption follows a linguistic framework: metrics must be culturally transmitted, discrete (clearly bounded), and productive (combinable). Skipping straight to complex models before establishing shared vocabulary for basic descriptive stats is the #1 reason analytics departments fail. When a coach and analyst argue about a finding, they're usually arguing about definitions, not data. Building the shared vocabulary takes 3+ years.
The most common failure mode for analytics departments is not analytical quality but problem selection. Analysts build sophisticated models that answer questions coaches never had, then are surprised when their work is ignored. The root cause: analysts optimize for analytical impressiveness rather than coach/decision-maker pain points. The fix is not better models but better problem discovery — sit in tactical meetings, listen to what coaches argue about, and solve THOSE problems.