Unless of course the horse is Mr. Ed

Revised and expanded, from a comment over at Gordon's place:

Obviously if Ed Thompson decides to seek the Libertarian Party's presidential nomination, he'll be the odds-on favorite to win that nomination.

And, I think he'd do well (by historical LP standards) in the general election -- at least 600,000 votes even if the election is tight as a drum, perhaps much better if it looks like a blowout for the Democrats (or, much less likely, for the Republicans).

BUT (and yes, I know, I have a stake in this as I support a presidential candidate with whom he’d be in competition, so feel free to take it with as much salt as you like) ...

I think he'd be wasting his time and his potential.

He could win a state legislative seat with one hand tied behind his back -- something a Libertarian Party candidate has never done in a large/populous state.

That's not intended to be disrespectful of those Libertarians who have previously won legislative seats in New Hampshire and Alaska, so please don't take it that way. But those are both small (population-wise) states with legislative elections that are really a lot more like small municipal elections in scale. New Hampshire's lower legislative districts each represent a population of about 3,500; Alaska's, 15,000; Wisconsin's, more than 50,000. Wisconsin has about three times the population of Alaska and New Hampshire combined, and is located in the heartland. A state legislative victory in Alaska or New Hampshire is great. A state legislative victory in Wisconsin would be huge.

Thompson could also be very competitive, possibly victorious, in a US House race.

It's even just possible, if everything fell out just right, that he could compete -- this time with a chance of winning -- for the governorship.

I'd rather have a great chance at a Libertarian state legislator in a reasonably large midwestern state, a good chance at the first Libertarian US Representative, or a fair chance at the first Libertarian governor, than a so-so chance at attracting a little temporary attention by pulling a very marginal, rather than a very, very marginal, vote total in one presidential race.

We're a long way from becoming competitive in presidential politics. The presidential race serves some important functions (helping generate national publicity, etc.), but Thompson's credentials are too good, and just as importantly too Wisconsin-centric, to waste on a race he won't win when there are important races he very well might win.

In my opinion, Steve Kubby is at least as sound a pick as Thompson for the LP's presidential nomination, and probably a better pick -- where Thompson would be a much sounder candidate for Congress, the state legislature or the governorship in Wisconsin than Kubby would be in California (33 million population, 400,000 citizens per lower-house legislative district).

I'll be the last one to claim that Kubby's name is a household word, but his name recognition is probably more national than Thompson's, and he has proven appeal to a national constituency (drug policy reform advocates), where Thompson's appeal has been based on local and state issues (he's been mayor of, and currently sits on the city council of, Tomah, and he led a statewide effort to get tavern owners some parity with the Indian casinos by allowing them to have video poker machines in their establishments).

In terms of overall name recognition, it's not easy to tell who has the edge. "Ed Thompson" is a fairly common name, and Google returns 145,000 results on it, as compared to the much less common name "Steve Kubby" (68,500). The problem is sorting out which results refer to this Ed Thompson as opposed to some other Ed Thompson -- on the first page of search results, four of ten results appear to refer to "other Ed Thompsons," where "the" Steve Kubby goes 10 for 10 on that test. If the word "libertarian" is added to the search, Kubby returns 40,200 results, compared to only 12,200 for Thompson. Adding the word "Wisconsin" to the phrase "Ed Thompson" returns 18,700 results.

If I had to guess, I'd guess that a random sample name recognition poll excluding Wisconsin (and, to be fair if you wish, California) would say that more people recognize the name "Ed Thompson" than recognize the name "Steve Kubby" -- but that more respondents could actually identify "the" Steve Kubby than could identify "the" Ed Thompson in any meaningful way.

I guess we'll see what happens.