The Relationship Between Predicted Outcomes and Observed Efficacy of Placebo Treatment

"We can prove that the estimate that you get for how effective the treatment is still unbiased. So it's not an overestimate or underestimate on average, correct? We can prove that if you compute a p-value from the analysis, like you would typically do, that it has exactly the right properties as it does out of a regular RCT," he says. "If you compute an error bar, the error bar you get from our analysis ... they all have exactly identical statistics."