I agree with your first comment, but for the second: I think that means that high earnings multiples and elevated risk would normally result in returns lower th

I agree with your first comment, but for the second: I think that means that high earnings multiples and elevated risk would normally result in returns lower than, rather than in line with, the average. So he expected what he thought would be the consequence of overvalued markets, but got averages instead. Outside the context of this bet, there's quite a lot out there on whether unusually high multiples (I think the really popular one for discussions is Shiller's CAPE) do or do not predict returns, or whether they predict anything at all. The opinions usually range from "the market says this is the right price for this risk, so it must be right" to "we're going to crash", but it's not all that clear who's right. (I think Shiller's right about overvaluation—I'm not sure what Buffett thinks.) But it's a pretty old topic, so that one is definitely not something he made up just as an excuse.