Winter attrition: why club attendance quietly halves every January

A Tuesday night run that had eighty-five people on it in October has forty by the last week of January. No single event caused it — no bad run, no falling-out, no announcement. It happened in increments of two or three fewer people each week, small enough that nobody flagged any one Tuesday as the problem, until the cumulative number was impossible to ignore.

The drop is gradual until it isn't

The seasonal reasons are real and unsurprising: it's dark at 6 p.m., it's cold, and the habit that held together from March through the holidays takes a two-week break and doesn't automatically restart. But the reason the drop feels sudden in February is that almost no club is tracking week-over-week attendance closely enough to see it coming — a spreadsheet updated occasionally, or no record at all beyond someone's general sense of 'it felt quieter tonight.'

The members who don't come back never announce it

Nobody cancels their club membership over text. They just stop showing up — first one Tuesday, then two, then it's been eight weeks and nobody noticed because there was never a moment that looked like a decision. The only real signal is attendance history, and if that history isn't being tracked in a way anyone actually looks at, the club finds out the same way it finds out about anything it isn't measuring: too late to do much about it.

• A member who ran three times a week for eight straight weeks in the fall, then zero times in the six weeks after New Year's, with no message either way.
• A weekly headcount that drops from 85 to 78 to 74 to 68 — each week a small enough dip that no single Tuesday triggers a conversation.
• An organizer's only signal being that Strava kudos and comments from a member have quietly dried up.
• A "we'll catch up in spring" assumption that's true for some members and false for others, with no way to tell which is which until spring arrives.

Catching drift while it's still one missed week, not eight

The alternative isn't guilt-tripping members back — it's catching the pattern in week two instead of week eight, when a personal 'haven't seen you, everything okay?' message still has a real chance of working, before the gap has calcified into a habit of its own. That requires attendance history a club can actually see trends in, and ideally a system flagging the drift automatically instead of relying on an organizer's memory of who used to show up.

By the time a decline is visible on the whiteboard headcount, it isn't a trend anymore — it already happened.

This is exactly the kind of pattern-in-the-data problem a club running on scattered spreadsheets can't catch and a club with real attendance history can.