Page:United States Reports, Volume 542.djvu/126

Rh to the consistency and rationality of judicial decisionmaking. That is especially so when they are applied to the hypothesizing of events that never in fact occurred. Such an enterprise is not factﬁnding, but closer to divination.

For purposes of estimating what would have happened, it seems to me that the only serviceable standards are the traditional "beyond a reasonable doubt" and "more likely than not." We should not pretend to a higher degree of precision. I would not, therefore, extend our "reasonable probability" standard to the plain error context. I would hold that, where a defendant has failed to object at trial, and thus has the burden of proving that a mistake he failed to prevent had an effect on his substantial rights, he must show that effect to be probable, that is, more likely than not.