Kevin Drum wondered last night if the conventional wisdom about California in the GOP primaries might be wrong:

One thing to keep in mind if you’re not from California is that our Republicans are not like, say, Maine Republicans: kind of moderate because they live in a basically liberal state. California Republicans are fire-breathing, take-no-prisoners, down-with-the-ship Republicans. I live in Orange County, which most people think of as ground zero for conservatism in the Golden State, and it’s true that we’re pretty conservative here. Our county board of directors routinely turns down federal money if it’s sullied in any way with connections to Obamacare. Still, as near as I can tell, OC Republicans are pussycats compared to Central Valley Republicans. I don’t know if the Central Valley Rs are more conservative than Alabama Republicans, but they’d sure give them a run for their money.

I don’t know very much about California state politics, but I do know that there’s some indirect evidence that might help us assess the last question, regarding the relative conservatism of California Republicans vs. Alabama Republicans. We can look at the voting records of their respective Representatives in the House.

Below is a plot of the average DW-Nominate scores of various partisan congressional delegations from the 111th Congress. The number next to each state is the state’s ranking among co-partisan delegations as to its tendency toward moderation (i.e. NY was the 5th most liberal GOP delegation; Idaho was the most conservative Dem delegation). Aside from California and Alabama, the states on the plot were not chosen by any algorithm; they are just there to give you the lay of the land.

You can see that California’s GOP delegation in the 111th Congress was, on average, more conservative than the average GOP Representative. The delegation ranked 31st most liberal out of the 41 states that had GOP Members in the 111th. Note that Alabama ranked 13th most liberal of GOP delegations. So based on this evidence (which is obviously indirect; we’ll get to caveats in a minute), Kevin’s most narrow point is well-taken: from a purely ideological point-of-view, it may be the case that Santorum or Gingrich might have more appeal in California than is typically assumed right now. Or at least the converse is not true: we don’t have any evidence that the Alabama GOP electorate is more conservative than the California GOP electorate.

More generally, I think the narrow takeaway point here is that there are a lot of surprises when you try to deduce the leanings of the primary electorate of a state based on that states general liberal-conservative reputations. For instance, it’s almost certainly wrong to assume that because California is a generally liberal state, it’s conservatives will be less conservative than those in a state that is generally more conservative. Another good example of this is Wisconsin, which I think we can fairly say is thought of as a liberal or liberal-moderate state overall. Indeed, its Dem delegation in the House was the 41st most conservative (i.e. 6th most liberal), on average, in the 11th Congress. But it’s GOP delegation was the 39th most liberal (i.e. the 3rd most conservative).

Now, for the caveats. Let’s start with the methodological ones. This is indirect evidence, at best. All sorts of disconnects exist between the DW-Nominate scores of Representatives from the 111th Congress and the ideology of the GOP primary voters in a state. To name just five: Reps are a small sample; Reps come from gerrymandered districts, not states, and those districts may be more/less polarized across states; conversely, GOP primary voters also exist in Dem districts; this was two years ago; and Member-district ideologies can diverge. The point is that you shouldn’t take this is some sort of smokin’ gun about primary voter preferences in California. It’s a lot closer to meaningless than it is to that. Just to highlight what I mean, Romney cleaned up in Arizona’s closed primary, which was the most conservative GOP delegation in the House in the 111th Congress.

Of course, this only begs a more important caveat: does the DW-Nominate scores of the House delegation sort along the same measure of ideology that is separating GOP primary voters between Romney and not-Romney? I haven’t looked into this enough to say, but my instinct is that it may very well not. Romney, Santorum, and Gingrich are all laying down pretty conservative policy positions, so the differences between them may have as much to do with style or messaging as they have to do with the actual policy differences of the type that could be picked up by something like a DW-Nominate score in the minds of voters. And so even if it’s true that California Republicans are more conservative than Alabama Republicans, it may also be the case that certain frames (such as an anti-Massachusetts attitude) are much more prevalent in Alabama than in California.

And yet, none of this is the main caveat, which undoubtedly remains the same as it has been for months: voter ideology, or voters in general for that matter, may have little left to independently say in the primary. This is because a fair amount of GOP primary voter belief (or any primary voter belief ) is really a coordination game being driven by party actors and party media outlets, who seem less and less likely to show any enthusiasm for the toppling of Romney in favor of Santorum. Which, of course, simply reiterates the fundamental truth of what is going on: California may matter this year, in the sense that Romney may not have the delegates to put him over the top by then, but the nomination in my view has been sewn up (save for Act of God type external shocks) for quite some time.