The player who never plays is the one who breaks

Squad workload arguments obsess over the overworked starter and ignore the reserve whose body has forgotten what a match feels like.

Everybody worries about the man who played eight games in twenty-four days. Fair enough. But the squad member I would actually worry about is the one who came on for eleven minutes in October and then sat there until February, because when he finally starts he is being asked to do something his tissue has not done in months.

Training cannot fake a match. You can run a squad player into the ground on a Wednesday and it still will not reproduce what happens to a hamstring when a winger drags him into a full-speed turn at minute seventy with a crowd behind him. The high-speed running you get in a controlled session is bounded by the session. Matches are not bounded. That is the whole problem.

So the load curve for a fringe player looks like a flat line with a cliff at the end. Weeks of moderate, predictable output, then one Saturday of maximum output because somebody rolled an ankle in the warm-up. If you plotted recent load against long-term load for that man you would see a spike no medical department would ever sign off on deliberately, and yet it happens constantly, and we call it bad luck.

I think clubs know this and quietly accept it, because the alternative is expensive. The alternative is giving your fourth-choice centre back real minutes in games you want to win, or arranging genuine behind-closed-doors fixtures at competitive intensity, which requires an opponent, a referee, a schedule, and a manager willing to lose something to protect something else. Loan moves solve it for the young. They do not solve it for the thirty-one-year-old on good wages who is too useful to release and not good enough to start.

There is a version of this that annoys me more. A manager rotates heavily in a cup tie, six changes, and the pundits say he is protecting his players. He is protecting eleven of them and exposing six others to exactly the spike I just described, and if one of those six pulls up in the second half it gets filed as misfortune rather than as the predictable outcome of the plan.

Minutes are a terrible unit for all of this. What matters is the shape of the distribution, not the total. Two hundred minutes spread evenly is a different body to two hundred minutes delivered in two frantic bursts either side of a ten-week gap.

Count the gaps, not the games.