They've been doing this since the new XP rule last year. The math is pretty straightforward. the XP is about a 94% proposition. If you think your odds are significantly better than 47% of making the 2-pointer, you should do it (ignoring late-game game-score-based decisions). If you don't, you should kick the XP.
The Steelers were much much better than 50% when they tried it last year. They bought themselves a lot of points that way. We should be praising the Dallas D for stopping the Steelers at something they're good at, not laughing at them for a "dumb" strategy.
I'm surprised more teams don't go for 2 more often. (Ha! I'm not really surprised. Nobody is more conservative than NFL coaches).