Six weeks looking for a signal, and what was left when there wasn't one

Six weeks on the Coinbase tokenized equity pools on Base went looking for a directional signal and did not find one. The largest reproducible effect was 2 to 3 basis points, against a round trip that costs ten. That gap is not a tuning problem. It is the answer.

The same data settled two other questions. Swap fees and adverse selection cancel almost exactly, so the fee side of providing liquidity is close to a wash. AERO emissions run about 2.6 times fee income. The return in these pools is the emissions, and the rest is noise around it.

So the thing worth building is not a trader. It is a daemon that holds a concentrated range, stakes it in a gauge, harvests, sells all of it, and re-centres when price leaves the range.

The premise goes in writing because it is easy to drift off. Providing liquidity is not a long position: it means selling into rallies and buying into declines. The bet is emissions against divergence loss and nothing else. Direction is an unwanted side effect, and any rule that conditions an action on where the price goes next is out of scope, whatever it is called.