Long View : Judging by the Run
You don't trust a season, you trust a run of them. That's not something a study taught me — it's just how farming teaches you to read anything, because a single season lies about as often as it tells the truth. A wet year looks like a wet year whether it's the start of a wet decade or a fluke storm that never repeats, and there's no way to tell the difference from inside it. You need the years stacked up end to end before a single reading turns into something you can actually stand on. This week, without my going looking for it, everything I read landed on the same lesson from a different direction: how long you have to keep watching before one data point earns the right to mean something.
Start with the plainest version. A study out of JAMA Network Open followed more than twenty-five thousand women for up to twenty-five years, tracking how closely they stuck to a Mediterranean pattern — vegetables, fish, legumes, olive oil, not much red meat or processed food. The closest followers had up to a 23 percent lower risk of dying from any cause over the study period, and thirty-three separate blood markers tied to metabolism and heart disease moved in the same direction at the same time, which is usually the sign you're looking at something real. Twenty-five years is a long run — close to the twenty-seven I spent farming, and that's not nothing, because you don't trust a single crop, you trust a run of them. But even a run this long only tells you what happened on average, across thousands of different lives. It can't tell you what happens to any one woman, any one man, eating this way for the next quarter century. That's not a flaw in the study. It's just what a long run of readings can and can't promise you — the pattern gets more trustworthy the longer it runs, and it still never resolves down to a single life.
The same lesson showed up somewhere I wasn't expecting it, in a week spent just noticing how a craving for a drink actually sounded when it showed up. Some days it was a real pull, old rope pulling tight. Other days it was closer to being offered more coffee after I'd already said no twice — background noise, easy to wave off. I used to assume that wanting ran at a constant hum, the same volume every day. It doesn't. It comes and goes like weather, and you can't read a single day of it and know anything true. There's a piece of neuroscience about this that reframed it for me — the brain's dopamine system fires hardest not at the reward itself but at the cue that used to precede it, which is why you can go on wanting something you don't even like anymore. I've watched a root do something close to that: a fig tree on the home place once found a hairline crack in an old irrigation line and drove a root straight at it, past better soil closer by, and kept returning to that exact spot for years after the leak was patched and there was nothing left to find. It took a run of days, not one, to notice that what I was actually chasing most evenings wasn't the drink at all — it was a stopping point, a line between work and not-work, permission to sit down. One day of noticing would have told me nothing. It took the whole week, stacked up, to see the actual shape of what I wanted.
Then there was the smaller experiment of writing a whole piece this week in iA Writer instead of Ulysses — no sidebar, no toolbar, nothing to look at but the sentence in front of me. A flat sentence in iA Writer is still a flat sentence; the app didn't make Tuesday's writing better than Monday's. What it did was let me sit with a half-formed thought longer before reaching for something to fix it, and that only became visible once I'd run the whole piece through it, start to finish, rather than judging it off one paragraph. It's the same thing as walking a field instead of diagnosing a crop from the truck window — you don't get the real read from a glance, you get it from staying in the row long enough for the pattern to show itself.
None of these three are really about diet, or drinking, or software. They're the same instruction showing up in three different coats: don't trust the single reading, trust the run. A practical person doesn't need to become a statistician to use this. It just means treating one lab number, one bad night, one craving, one clumsy paragraph the way I'd treat one dry week in July — worth noting, not worth deciding anything from. Wait for the run. The pattern either holds up or it doesn't, and either way, you'll know a good deal more than you did after a single look.
The full posts go deeper — they're available to subscribers.
Find me on Ghost: https://gardener-of-life.com
Micro.blog: https://micro.blog/gardeneroflife
Substack: https://gardeneroflife.substack.com
Food recall alerts: https://www.foodsafety.gov/recalls-and-outbreaks
If this resonates, buy me a coffee — it keeps the field notes
https://buymeacoffee.com/gardeneroflife
This post is free through December 31, 2026. Starting January 1, 2027, Gardener of Life moves to a paid membership.
Questions or thoughts? Email me at :
contact@gardener-of-life.com
