Thanks, everyone, for your questions. Here's the first round of answers. Feel free to keep asking, and I'll tackle more questions (including, probably, a couple I've gotten via email) later this week.
Impliedi: If Mark Melancon goes, is Tony Watson definitely the Pirates' closer, or do they look elsewhere?
There's probably about a 90 percent chance it's Watson. Neal Huntington has said that he'd be very confident in Watson as a closer. Also, Watson's good, and he's close enough to free agency (with two years remaining) that the Pirates don't have to worry much about Watson's saves totals increasing his salaries in future seasons. If the Bucs traded Melancon, and Watson had, say, less than two years of service time, they might well just go get someone else to close and continue to use Watson in a setup role, since saves are a key component of salaries in arbitration seasons. But with four-plus years, they likely won't worry about that.
If the Bucs did trade Melancon and get someone else to close, I don't know who it would be. There are obviously some pretty compelling options on the trade market in Aroldis Chapman, Andrew Miller and Brad Boxberger, but it's hard to see them trading Melancon for younger talent and then using the same type of young talent to acquire a player who plays the same position. Of the three, I suppose Boxberger makes the most sense, since he's still cheap and under control for four more seasons. Then there's the free agent market, but there isn't a ton out there on the the closer market beyond Darren O'Day and Joakim Soria, and even Soria is supposedly asking for three years and $27 million at this point.
Then again, perhaps there's also the possibility that the Pirates could trade Melancon and Watson. That seems highly, highly unlikely, but it might at least be worth considering. See below.
IsIt2015Yet: What about keeping Melancon and trading Watson for a Craig Kimbrel-like haul (due to more control)?
Watson wouldn't command a Kimbrel-like haul, but I'm sure you know that, and the question is a good one. Part of the idea of trading Melancon, rather than Watson, is that saves are overvalued in the market, but that's less true than it once was. (The Kimbrel trade seems, on the surface, to suggest that the saves market is stronger than ever, but as Red Sox blogger Matt Collins has repeatedly pointed out, that deal wasn't really a "saves LOL" trade, it was a trade for a player who's brilliant at what he does. I think they did overvalue him, just not really for reasons that can be dismissed with jokes about Proven Closers.) And maybe you could say that Watson should have plenty of trade value too, and that if Melancon is replaceable, then so is Watson. Some of the same qualities that could make Melancon a good trade candidate (like the fact that reliever performance varies widely by season, or that the Pirates don't have many years of control left) apply to Watson as well. I'm not sure there will be many Watson-related rumors this offseason, but I don't see why the Pirates shouldn't listen if a team makes a nice offer for him. They would just need to have a good plan in place to replace his production.
Iowasteel: I believe that the Bucs have outperformed the projected number of wins in each of the last three seasons (use any system you want) by at least 10. Do you think there is a specific reason for this, or do you attribute it to chance? If the former, what exactly are the projections missing?
You could read Travis Sawchik's Big Data Baseball as a 300-page response to this question. Without getting that deeply into it, I think what projection systems miss (and what I, too, have undervalued in my own pre-season estimations of the team) is that the Pirates' program of coaching, framing, shifts, ground balls and a good ballpark makes their pitching staff more effective than it looks on paper. ERA estimators like xFIP have consistently missed the boat on the Pirates' pitching staff in the past few seasons, and I think projection systems undervalue them for the same reasons. Check out FanGraphs' comments about the Pirates' pitching staff from last February to see what I mean. xFIP and projection systems are great tools in general, but the Pirates' pitchers are outliers in many respects.
Projection systems have missed on some hitters, too, like Jung-Ho Kang and Francisco Cervelli last year (which I think was understandable in both cases, given those guys' complicated track records). And this past year, as readers have pointed out, Melancon and Watson were brilliant when it mattered most, in a way that I would argue was unforeseeable and probably also unsustainable.
Biff Bigelow: It doesn’t seem like the Pirates will get much for Pedro Alvarez in a trade. Do you sense that Neal Huntington is willing to finally deal this guy even if it means he doesn't get much in return?
I would be, but I don't know what Huntington might think. At this point, nearly everything about Alvarez's value hinges on whether he can play defense. It would seem that there's nowhere to go but up in that department, but Steamer, for example, projects that Alvarez's defense will improve significantly next season, and even then, it only projects 0.8 wins above replacement. (Of course, we've already established that projection systems might be imperfect, but I have more confidence in them for Pirates hitters than I do for their pitchers.) That still isn't enough to justify the salary he'll likely receive. I'm sticking with what I said in September, although I can't say how well that aligns with the Pirates' thinking.