Learning / Writing
On focus, in climbing and in numbers
A short piece on what eight years of competition climbing taught me about attention, and why it still shows up in how I approach a proof or a forecasting model today.
I stopped competing in 2024. I expected to lose the thing competition had built in me. What I actually lost was the scoreboard — the habit of attention stayed, and it turned out to be the part worth keeping.
The problem with trying harder
Effort isn't the bottleneck past a certain level — direction is. The climbers who plateaued were rarely under-trained; they were mis-focused, spending energy on the wrong five percent of the problem.
You see the same thing in a model that won't converge. The instinct is to add data, add parameters, add hours. Usually the fix is upstream: an assumption nobody restated, a variable doing more work than it should. Trying harder in the wrong direction is still the wrong direction, just more expensive.
Reading the rock, reading the data
Route reading is pattern recognition under time pressure — you scan for the sequence that isn't there yet. You get one look from the ground, and the move you can't see is usually the one that decides the climb.
Debugging a model or reading a regression table rewards the same instinct: resist the first explanation, look again. The first story the data tells is the one you already believed before you opened the file.
What competition took away
Scorecards narrow attention in ways that are useful in a final and corrosive over years. When every session is scored, you stop noticing anything that isn't measured — and most of what makes the work interesting isn't measured.
Stepping back from the national team was, in part, about getting that width of attention back. It took about a year before I stopped grading my own training sessions out of habit.
Where it shows up now
Fewer variables, held for longer, looked at more carefully — in a forecast, a painting, or a boat hull, that's the same move.
The transferable skill was never the fitness. It was learning what to give attention to, and being willing to spend a long time on it before deciding whether it worked.