I spent the better part of 2024 treating every ride like a time trial. My morning commute had fourteen segments, and I attacked each one without exception. By October, I had racked up twenty-three new PRs and four course records on the local climbs. My functional threshold power — the number that actually matters — had quietly dropped eight watts.
The dopamine math of micro-competitions
Segments are brilliantly engineered feedback loops. You push hard, you get a number, you compare it against hundreds of strangers who rode the same stretch of road. The problem is that this reward structure optimizes for arbitrary maximal efforts: sixty-second sprints scattered across four-hour rides, no recovery protocol, no long-term progression. By November, my log showed the pattern immediately: too much time above threshold, too little actual fitness.
The leaderboard rewards the rider who goes hardest right now. Structured training rewards the rider who can repeat it next month.