WY-AL: Spoke Too Soon?

Just as we talked-up Gary Trauner's chances in Wyoming on the heels of two polls giving him the narrowest of edges, a new Survey-USA poll this morning finds him in a hole with two weeks to go.

Surveying for Roll Call, SUSA found Cynthia Lummis leading Trauner 50-44, with the libertarian candidate garnering four percent. For reasons I will explain, I can't say I fully buy these results, but they confirm both Trauner's standing in the mid-40s, as well as some of the fears we expressed for Trauner. Here are some fo the key findings:

Among Republicans, Lummis leads 74-22
Among Democrats, Trauner leads 86-9
Among Indepdents, Trauner leads 46-41

Lummis approval split: 41/35
Trauner approval split: 42/36

Interestingly, this marks the third poll in a week that puts Gary Trauner at 44 percent, not a bad place to be in Wyoming, but still several points short of victory. Therefore, I think we can say that that is a fair statement of where Trauner is today (or at least was last week). As a result, this poll differs from the other two mostly in finding where the Republican candidate, Lummis, stands.

If correct, these numbers well-explain Trauner's dilemma. First, he is very slightly underperforming how he did with Republicans in 2006 when he won one-quarter of them. Among Democrats, he is at about where he was, though a couple points behind. The killer is that independent vote we discussed yesterday. In 2006, he won that group 71-29, but this year he is barely ahead of Lummis. Without another big win with the indies and unaffiliated, he has no shot.

In terms of personal numbers, SUSA seems to find that both of them have the same numbers, but unlike R2K and M-D, they find them much less popular.

I think I can explain why SUSA's findings are much different than the other two polls we've discussed. First, is what SUSA says about Republican attitudes in this race. Survey-USA is well known for having very few undecided voters in their polls. This is likely because they push their leaners harder during the survey to get a harder number closer to 100 percent. Both R2K and Mason-Dixon found a substantial number of Republican voters undecided in this race, around one-fifth of them, but SUSA finds here that just two percent of GOPers are undecided.

Now, as we have talked about, this may well be ultimately right. In the end, any wavering Republicans may just be unable to vote for Trauner because of his party affiliation. Indeed, SUSA peg's Trauner's approval rating with the state's Republicans at 21/55, or -34. This is in stark contrast to the 40/52 (-12) that R2K found last week.

I don't doubt that many undecided Republicans will have a hard time voting for Trauner, and that in the end most of them won't be able to do it. However, given R2K and M-D's findings, as well as what I have followed in this race, I do not get the sense that Trauner is at a -34 with Wyoming's Republicans. If he is, he is obviously cooked, but I just don't buy it.

Second, SUSA brings back the discussion we had on the proper party break-down for a sample of Wyoming's voters. SUSA's sample was made up of 55 percent Republicans, 28 percent Democrats, and 17 percent Independents. This mirrors the make-up of the 2006 voting electorate.

I really do not know what to make of it. On the hand, there is little question that I have doubts with R2K's 48-24-28 sample. I think that while it slightly under-samples Democrats, it heavily over-samples independents. As a result, there is a good chance that R2K's numbers are a bit off. However, on the other hand, this year is likely to be a better Democratic year nationally than in 2006, even in Wyoming, so maybe the 55-28-17 sample is not fully reflective of what we will see at the polls on November 4.

Additionally, the question of Republican undecideds might still apply. Both R2K and M-D showed huge swaths of undecided Republicans, yet SUSA with an even bigger GOP sample found that number to be far smaller. That might be less about the size of the GOP sample, and more about pushing undecided voters. We don't know.

I think that the answer is somewhere in between. I do not think that only two percent of the state's electorate is undecided in this race. Nor do I think that Gary Trauner is at a personal +6. I do not have scientific evidence to back that up, just my gut feeling. As a result, I doubt that Lummis is already at 50 percent, though it is very likely that Trauner is at around 44 -- in fact, all three polls in the last week put his number at 44 percent.