Dear Magicians,

Oceanic phytoplankton generate roughly half the Earth’s oxygen. The exact percentage has been revised upward several times because we kept not counting things we didn’t know to look for.

In 2022, researchers discovered that krill and copepods—long assumed to be passive consumers in the carbon cycle—actively pump significant carbon to the deep ocean through their daily vertical migrations. The “biological pump” was substantially underestimated because the creatures running it were too small and too numerous to census properly. Some estimates now attribute 15-45% of carbon sequestration to organisms we weren’t counting a decade ago. → Source​

Ecology keeps finding that the things we didn’t measure were doing the most work.

The most important variables in any system are often the ones you don’t have a spreadsheet for yet.

For most of my career I believed the number that mattered was the one on my CV. Citations. Grants. The h-index that a committee could read in four seconds and pretend it had understood a life’s work.

That number is easy to count. That is its entire appeal.

Here is what nobody warns you about in graduate school. A metric survives not because it captures what matters, but because it is cheap to measure. The citation is countable. The quiet insight you gave a student at 11pm is not. So the institution optimizes for the first and calls the second a rounding error.

Ecology keeps embarrassing us on exactly this point. We assumed the ocean’s oxygen came from the plants we could see. Then we learned that microscopic drifters we never bothered to census were running half the planet’s respiration. The creatures too small to count turned out to be doing the most work.

Science is supposed to be the discipline that resists this. Feynman’s whole posture was a refusal to be impressed by a number you could not trace back to an experiment. Yet the fields that most loudly claim his lineage now grade themselves on proxies he would have laughed out of the room.

The uncomfortable move is to ask what your own scoreboard leaves out. Not the metrics you are losing on. The ones you never thought to keep.

Measure what is convenient and you will slowly become good at the wrong thing. Slowly enough that it feels like progress the entire way down.

The most important variable in your life this year is almost certainly one you do not yet have a column for.

Have a MAGIC week!

Brian