The problem with the Stanford Prison Experiment was that the guards were secretly coached by the researchers to get the effect their hypothesis predicted, not that "the effect was too strong": https://www.vox.com/2018/6/13/17449118/stanford-prison-exper...