The Frequency Where You Stop Watching
A number climbs somewhere you can see it. Unread count, pending updates, unmerged pull requests, league standings during a losing streak. The number has nothing to do with how dangerous your situation actually is, and everything to do with how bad you feel about it, and those two facts are not the same fact.
Take the simplest version: a package manager that shows you how many updates are waiting. Watch it a few times a day and it reads like a threat level, climbing toward some vague catastrophe you can't name. Check once a week instead and the same hundred-plus updates arrive as a single batch, you install them, something occasionally needs a small fix, you reboot, done. The actual risk, exposure to an unpatched bug, a broken dependency, didn't change between those two checking schedules. What changed is how many times you sat with the number before you did anything about it.
That's the part worth sitting with: the anxiety a counter produces scales with how often you look at it, not with the rate at which the underlying risk accrues. Most of what these counters track doesn't accrue continuously. It accrues in lumps, at the moment of the reboot, the moment you actually read the inbox, the moment the season ends and the standings lock. Between those moments nothing is really happening to you. But the display doesn't know that, so it ticks up anyway, and your body reads a ticking number as danger whether or not danger is the right word for it.
The instinct, when a number makes you anxious, is to get better at managing what it represents. Build a triage system for the inbox. Read the changelog before deciding whether an update matters. Track your team's underlying stats instead of just the standings, so you can tell yourself the record doesn't reflect the real talent gap. All of that is real work, and none of it touches the actual problem, which is that you're going to keep looking at the number on the same schedule and feeling the same way regardless of how sophisticated your explanation for the number gets. You can become an expert at contextualizing your own dread and still feel it every time you open the tab.
The fix that actually works is dumber than triage. It's finding the interval at which you stop looking enough to feel anything. Not zero, usually, zero just defers the risk until you have no feedback at all and something eventually breaks in a way you can't trace. Not every few minutes either, since at that frequency the counter is updating faster than any real consequence could, so you're reacting to noise and calling it vigilance. Somewhere in between is a frequency where the gap between checks is wide enough that most of what accumulates is harmless by the time you look, and you get to see that for yourself, repeatedly, until the number stops reading as a threat and starts reading as a chore.
This is a calibration you run on yourself, not a policy you reason your way into. You don't know the right interval until you've overshot it a few times, checked too often, felt the dread, checked too rarely, gotten burned once by something that should have been caught sooner, and landed somewhere in between by elimination. The number itself can't tell you where that point is. It just keeps climbing at whatever rate you've set up to watch it climb.
Which is the thing about a visible counter: it was never measuring your risk. It was measuring your schedule. Fix the schedule and the number goes back to being a number.