I was unaware this was a competition.
Thats the problem with conventional wisdom and perceptions like this. Even though something is formed mostly on a hunch with little evidence to back it up, it makes you believe anything to the contrary is asinine. Sure, the article may not be a huge sample size, but its still bigger than any which show closers struggle in non-save situations. The closest thing to suggest that is a 3.5 year split from 2002 to mid-2005 that showed closers had a 2.51ERA in save situations and a 3.26ERA in non-save situations. That is clearly a difference, but a 3.26ERA is hardly indicative of a closer struggling, and certainly not indicative of a closer being incapable of pitching as a setup man, which was really the issue at hand.
I never disagreed with the notion that some closers pitch better in save situations, but the idea that a majority of closers (or even close to a majority) struggle in non-save situations is baseless.