That's not what the OP was saying. The argument is if somebody can take Silver's model's predictions over time and produce a better next estimate than it, then

That's not what the OP was saying. The argument is if somebody can take Silver's model's predictions over time and produce a better next estimate than it, then Silver's model is incoherent about its own beliefs.