He isn't arguing that they do that, he's saying that if we accept that that remains a reasonable methodology to extrapolate the data, then we can use that metho

He isn't arguing that they do that, he's saying that if we accept that that remains a reasonable methodology to extrapolate the data, then we can use that methodology to get an updated estimate. He gets his estimate a couple of different ways, and they end up similar.