Most placebo effects are modest and the ones that aren't are usually mixed in with regression to the mean effects(which are usually pretty modest). So OP is implying that if an effect is especially large it's less likely to be placebo.