Should the Mets sign Matt Wieters? If you were to ask Mets fans, you wouldn't have a very difficult time finding a vocal subset of supporters who would scream, "Yes, of course." Just search "Mets + Wieters" on Twitter, and you'll find no shortage of blog posts and tweets in support, and the idea has popped up in multiple Mets Inboxes this offseason. After all, Wieters is a four-time All-Star who has yet to find a home, and incumbent Travis d'Arnaud is coming off a disappointing .247/.307/.323 season, in addition to regular injury issues.
But the Mets almost certainly aren't going to sign Wieters, and while it's partially about money -- they've yet to clear Jay Bruce's contract and still would like to sign a reliever or two -- it's not entirely about money. It's about the fact that despite the huge differences in their reputations, there's enough reason to think that d'Arnaud will be as good or better than Wieters in 2017 that New York is making a reasonable enough choice in allocating its dollars elsewhere.
How could that be possible, after d'Arnaud's down season and Wieters' reputation? Let's start with the fact that "number of All-Star Games made" isn't really relevant -- Eduardo Nunez and Stephen Vogt made it last year, while a star like Justin Turner has never made one -- and investigate.
Because you can't just look at the most recent season to define a hitter
You can't really make d'Arnaud's 2016 season look good, because it wasn't. His slash line comes out to a 74 Weighted Runs Created Plus, a fancy stat that's actually quite simple to understand. It's adjusted for park and league, and it sets 100 as league average -- for context, Yoenis Cespedes had a 134 wRC+, and Curtis Granderson was at 114 -- so you can say d'Arnaud was 26 points below average.
But All-Star appearance aside, Wieters' 2016 was only slightly better. His line of .243/.302/.409 was the same in average and on-base percentage, and it contained a bit of extra power. Wieters' wRC+ of 87 was better than d'Arnaud's, yet still below average. Either way, you can't simply look at last year and expect a repeat. If you look at the past two years, d'Arnaud (102 wRC+) was better than Wieters (93). If you look at the span of d'Arnaud's career, he's at 98, and Wieters is 93. The only way Wieters looks better is if you only look at '16, which no team would do.
Think about it this way: Most projection systems look at the past several years of performance, with more recent numbers weighted more heavily (this helps Wieters), and they include an adjustment for age. The respected Steamer projection system sees d'Arnaud in 2017 as an average hitter (.254/.321/.417, 100 wRC+) and Wieters as slightly below (.245/.302/.408, 88 wRC+). Another system, ZiPS, sees them as identical. It's hard to find a way to prefer Wieters at the plate -- particularly if mechanical flaws in d'Arnaud's swing can be resolved.

Because d'Arnaud is a better framer
The rise of pitch framing as an important way for a catcher to add (or subtract) value has been well documented over the years, and it's a huge part of why weak hitters such as Jason Castro and Jeff Mathis extracted multiyear deals from Minnesota and Arizona, respectively, this offseason. Framing matters, and teams continue to show they care about it.
Wieters isn't bad, but he's not strong here. In 2016, Baseball Prospectus ranked 114 catchers, and Wieters was 89th, at -4.4 runs. In '15, at -3.6 runs, he was 93rd of 117, and he's been below average in each of the past five years. d'Arnaud, meanwhile, was No. 7 in '15, at +12.6 runs, and No. 14 in '16, with a mark of +7.5 runs. It's a pretty large gap.
Maybe you care about the way a catcher steals a strike, and maybe you don't. Either is fine. The point here is that teams care, and that's important. 
Because the differences in preventing steals may be overstated
"But d'Arnaud can't throw out any runners," you may be protesting, and that was true in 2016. He allowed 61 steals and threw out just 17, a 22 percent success rate that was below the Major League average of 28 percent. Wieters, meanwhile, had an above-average 35 percent success rate. It wasn't quite so bad in '15, when the two were nearly even (33 percent to 31 percent), while Wieters was recovering from right elbow surgery.

Part of that, of course, was that d'Arnaud dealt with right shoulder issues in 2016, and part of it is that some Mets pitchers such as Noah Syndergaard didn't do a great job of holding runners on -- and we're learning through Statcast™ data just how important that is in terms of giving catchers a chance to catch basestealers. Plus, after a season without a catching instructor, the Mets have added coach Glenn Sherlock to help d'Arnaud behind the plate.
Ultimately, it's going to come down to health for d'Arnaud, as it always does, because the Mets are betting on upside. With Wieters, who turns 31 this year, you could spend millions and be reasonably sure you'd have a competent veteran catcher, but with little chance of a breakout, as his performance has not kept up with his name value or his salary demands. With d'Arnaud, 28 later this month, there's still the chance he'll have that big season we've seen flashes of. It might not be the popular choice, but it's an easily defensible one.