It's true that in most cases if one thing beats another, you can find some dimension on which the winner was superior. But it isn't necessarily true in all case

It's true that in most cases if one thing beats another, you can find some dimension on which the winner was superior. But it isn't necessarily true in all cases. If it is the case that sometimes things win only due to entrenchment or random network effects, then there may be some cases where the winner is strictly worse than the loser, though these cases will be rare. If you found one, though, it would be the strongest possible evidence that sometimes markets/evolution/whatever don't pick the best thing. The OP is asking for those examples.