Taking the Moneyball approach beyond the touchline

Clubs can now measure almost everything a player does. The rooms where a club's future is actually decided remain unmeasured. That gap is the widest inefficiency left in sport.

More than two decades on from Moneyball, the on-field argument is over. Recruitment runs on models. Expected goals sits in the Saturday night graphics. Tracking data prices players, sets pressing schemes and gets set-piece coaches hired. A Premier League club can tell you a full-back's sprint volumes, pressing triggers and progressive carries across three seasons in two leagues, and no serious club would sign him without that file.

Then look at the decisions that actually move a club's valuation. A transfer negotiation. A stadium planning application. A licensing conversation with a regulator. A league vote on money. A pricing announcement to the fanbase. These are decided by people, in rooms, and most clubs prepare for those rooms the way they prepared for opponents in 1995: relationships, instinct and whatever the building already believes.

The stakes point the wrong way for that to be sensible. Overpay by ten per cent on a £40m transfer and the club has lost £4m in an afternoon, several years of an analytics department's budget. A failed planning application adds years and tens of millions to a stadium project. In England, a mishandled regulator relationship now carries licensing consequences. The best-measured organisations in sport are flying blind precisely where the money is largest.

The gap exists for an understandable reason. On-field analytics works because match events are plentiful and structured: millions of passes, presses and shots, all timestamped. Off-field events are rare and human, so the assumption took hold that they cannot be modelled. The assumption confuses events with people. You cannot build an event model of a planning committee. You can build an evidence model of the people on it. Counterparties, councillors, ministers, league executives and regulators all have records: past deals, votes, statements, interviews and revealed behaviour under pressure. Different evidence, same discipline.

Where it applies

Negotiation is the sharpest case. Clubs spend months modelling the player and no time modelling the person selling him. Yet the other side of the table has a track record too: how they anchor, when they concede, how they behave against a deadline, which arguments have moved them before. A deal team that has rehearsed against an evidence-bound model of the specific counterparty walks in having already seen most versions of the meeting. In a window where one misread costs seven figures, rehearsal is the cheapest thing on the invoice.

Regulation is next. English football now has a statutory regulator, with licensing and ownership tests attached. How named officials interpret written tests is a modellable question, and anticipating an interpretation is worth considerably more than contesting one.

Politics and planning follow. Stadium and training-ground projects live or die in council chambers and in Whitehall. Committee members, local MPs and ministers have voting records, stated priorities and constituency pressures, most of it public and almost none of it systematically used. Clubs tend to discover these dynamics in the consultation responses.

Fans are the audience clubs think they know best and test least. Pricing moves, kit decisions and format changes can be run against synthetic fan populations before release, segmented the way a fanbase actually splits: season-ticket holders, match-going casuals, overseas members, lapsed attendees. The reaction arrives before the announcement rather than in the stands.

And league politics, distribution deals, format votes and franchise processes, including those forming around basketball's expansion into Europe, are decided by a small number of named people whose incentives can be mapped well before the vote.

The same move, twice

This is the identical intellectual move that recruitment analytics made twenty years ago: replace anecdote with evidence and keep humans deciding. Data did not replace scouts. It disciplined them, and the clubs that resisted paid for the education. Stakeholder modelling does not replace the sporting director, the chief executive or the public affairs lead. It means they arrive prepared, with the counterparty rehearsed, the committee mapped and the fan reaction already tested.

Our version of that preparation is built from evidence-bound twins of the people across the table, graded for fidelity, with every forecast logged and scored. It is the same method we apply to ministers and regulators elsewhere, because a sporting director under deadline and a minister under pressure are, analytically, the same problem.

The first clubs to take Moneyball seriously bought wins cheaply for a decade before the market corrected. The off-field market has not corrected. The widest margins left in sport are in the rooms, not on the pitch.