On the Sudakov Form Factor, and a Factor of Two

I answer a question that Roman Jackiw asked me, and I draw some lessons from the answer. The question is: why is the Sudakov form factor larger by a factor of two, if computed for off-shell fermions, in comparison to the onshell case? The answer sheds some light on the interplay between infrared and collinear singularities — and the importance of factors of two.