Robert Sarvis, Libertarian Senate Candidate from Virginia, Denies "Spoiler" Charges from the GOP

I blogged on election night about some NBC exit polls that I thought lent some credence to the idea that, without Robert Sarvis in the Virginia Senate race for the L.P., that Republican Ed Gillespie might have won.

The heart of my analysis:

Sarvis drew equally from liberals, moderates, and conservatives according to this poll—3 percent of each.

But when it gets to party identification, he drew statistically nothing from Democrats, 3 percent from Republicans, and 7 percent from Independents. Independents were otherwise split evenly 47-47 between Warner and Gillespie. So, there is indeed some cause for GOPers to think that Sarvis' presence in the race was bad for them.

Sarvis wrote me last night with a contrary analysis, hooked to a fact I neglected to account for: that not all Republicans, if the exit poll is correct, actually voted Republican in this race–7 percent went Warner:

One can't assume the 3 percent Rs would be voting [Gillespie] in my absence—it's quite likely these R voters would have joined the 7 percent of Rs voting for Warner. Polls throughout the race showed Warner enjoying double-digit support among Rs, and a fair number of Rs told us they can't stomach voting for [Gillespie]. A lot of business-type Republicans consider Warner acceptable, so probably many Rs who really disliked [Gillespie] voted for me because I was preferable to Warner, but would otherwise have voted Warner not Gillespie. So those R Sarvis voters were "taken" from Warner not Gillespie.

Similar thing happened last year, with pretty high certainty. A poll in September showed that *among Sarvis supporters*, 60+ percent had a favorable opinion of Gov. McDonnell, but 70+ percent had an UN-favorable view of Cuccinelli. So I was a vessel for moderate, R-leaning, anti-Cuccinelli voters who preferred voting for me to voting for MacAuliffe, i.e., I "took" moderate R votes from MacAuliffe.

Moreover, my share of the Independent vote clearly skewed younger, so from voters not inclined to vote D than R.

Reason on the whole "spoiler" thing with the Libertarian Party.