I'll take on the "should" part of this. In my view, this is Seligism at it's worst: reforming institutions in order to react to some minor glitch in the current rules or practice. In particular, the offense here -- the slow count in two caucus states -- just doesn't seem to me to be a very big deal. At all. It took two weeks for the recount in Iowa? So what! Recounts take a while. The offense here just doesn't seem like a very big deal to me.
In addition to better counting, Wilson endorses a few other reform goals:
Give candidates a break from holiday-time campaigning; ensure that candidates with modest war chests can truly compete; and produce a nominee battle-tested in every region who isn’t too bruised and bloodied to compete in November.Everyone agrees with the first one, but alas coordination problems, as Josh explains, make it difficult. The third goal seems like wishful thinking to me; there's simply no way to assure that a candidate is both "battle-tested" from a sequential nomination process without risking a nominee "bruised and bloodied." Still, it's an overrated concern. Barack Obama certainly survived as tough a process and one could expect and did just fine, and in general there hasn't been a problem that candidates emerge from the process at a disadvantage in November (the strongest cases were all long in the past -- George McGovern in 1972, Gerald Ford in 1976, and Jimmy Carter in 1980). As for helping candidates who can't raise money compete: it's not clear that's a good thing. I'm open to complaints that the system has suddenly tilted too far in the direction of one or two major donors allowing a candidate to survive, but the truth is that any candidate with serious party support is going to be able to raise enough money to compete. Failure to have more than a "modest warchest" is a sign of a failed candidacy, not, in most cases, something that the parties should want to reward. I do think that money should not be the only resource that matters in nomination politics, but fortunately it isn't.
The other part of this is to think of the nomination process as a complex coordination and competition game: hundreds, or really thousands, of party actors across the nation are trying to come to an agreement while protecting their interests within the party and without too much damage to the eventual nominee. You know what helps make that process work better? Stable rules. Without stable rules, it's a lot easier to get odd, random results; it's easier for candidates to game the system; and it's harder for the party to resist press manipulation. Obviously, there are other, competing values here. But stability in the basic rules is really a highly valuable asset for parties, and in my view at least they should be careful before embarking on major reforms without very, very good reason. Democrats in particular can remember the consequences of the major reform of the 1970s: the disaster (at least for them) that was Jimmy Carter.
All that said...the current system, if retained (as I agree it will be), will never give us complete stability either in the calendar of primaries and caucuses or in the choice of states of how to select their delegates. So we'll always have states switching from caucuses to primaries or vice versa. All that is fine. But I see no reason at all to attempt to impose one or the other.