Well, you can't reconsider your approach when you don't like the results.

If anything, the fact that this is what he arrived at, even when starting with the opposite position, is proof of the validity of this result.