The argument about how strong (or weak) the current Republican field of presidential contenders is back, with Nate Silver putting together some evidence of how unpopular the current groups is compared with previous fields, and Brendan Nyhan adding even more data.
I still find this discussion a lot less systematic than it should be. First of all, I agree with Nyhan's previous post on the subject: we can expect candidates who have little stature now to gain that quality as a consequence of winning primaries and caucuses, and eventually having a convention devoted to praising them and then debating the President of the United States of America (or, in other years, at least the other party's nominee). And I'll stick to what I said before: when thinking about the general election, it's not the field that matters; it's the winner. But that just begs the question: what do we mean when we talk about a weak candidate field?
It seems to me that the first thing to distinguish is between strengths as a candidate for nomination and as a candidate in a general election. It's pretty clear that these are not always the same thing at all; to begin with, moderation would usually be a minus in primaries, a plus in November. Some candidates have no chance for the nomination because they fail a key litmus test on an issue vital to an important party group, but would be excellent general election candidates if they could somehow navigate that problem.
So: what makes a "strong" or "weak" candidate? For the nomination, I'd start, before anything else, with those veto issues. Indeed, the 2008 GOP field was (in my opinion correctly) thought of as weak in that sense because each candidate seemed to have a serious issue clash with a veto-wielding portion of the party. Thus the demand for a new, broadly acceptable candidate (alas for Republicans, that only managed to produce the ill-fated Fred Thompson semi-campaign).
What else matters? That's where it gets awful tricky. Paper qualifications count (so Steve Forbes or Alan Keyes could be identified as weak candidates). Ability to run a national campaign, which includes both the public and the management sides -- but identifying the ability to do that is very difficult, and probably calls for subjective judgement by careful observers rather than any objective criteria. Presumably, that is, someone should have known that Fred Thompson was likely to be listless, that Phil Gramm wouldn't appeal outside of his home region, that Bill Clinton would persevere when the going got rough, and so on.
What about the general election? First of all, it's worth noting (as Nyhan and Silver both do) that candidates make less of a difference in November. Much less.
Beyond that, I'd probably say that there's little, but not nothing, that we can guess at from eighteen months out. Generally, as I said above, moderation is better than extremism. It's probably worth looking at some of the "ability to run a campaign" markers mentioned above, but not all that closely -- anyone with very strong negatives in this area isn't going to win the nomination.
Now, what I've omitted so far is the one and only thing that Nate Silver looks at in his post: current popularity. For the general election, I just can't see that it matters. Tim Pawlenty may be virtually unknown now, but if he's the Republican nominee he's going to have universal name recognition and will likely be reasonably popular. Or, perhaps, not -- but I'm not at all convinced that his early lack of name recognition gives us even a hint of how he would fare compared to a generic Republican in the general election. I would agree that the rare candidate who has proved to be unusually unpopular over many years -- Sarah Palin, Newt Gingrich -- would be an exception, but even that comes with a caveat to be careful in how to interpret early negative ratings. For a well-known candidate, it could just be normal partisan polarization kicking in, something that will eventually come to Pawlenty or the other obscure candidates should they get nominated.
Last thing: the strength or weakness of the "field" needs to be thought of top-down for most questions that might interest us (such as: could another candidate enter late and win? Or: will the eventual nominee be stronger or weaker than a generic candidate?). The 2008 Democratic field was unusually deep, with the 4th through 6th candidates (Biden, Dodd, Richardson) unusually strong for either the nomination or the general election, but that was a curiosity, not an indication of anything. On the other extreme, many thought that the 1984 field was unusually weak behind Walter Mondale -- but for the nomination, Mondale was a very strong candidate, so it nothing about Alan Cranston or Fritz Hollings really mattered.
All in all, I think that attempts to measure the strength of the candidate field at this point are interesting, but they should not be based only (or even mainly) on early polling, and it's important to separate out the idea of strength for the nomination and strength for the general election.