Every year, I issue Cook Political Report–esque race ratings for several downballot races that get shafted in the attention department. In 2015, two (separately-elected-from-the-governor) lieutenant governors, three attorneys general, three secretaries of state, three treasurers, two auditors, three agriculture commissioners, and two insurance commissioners were elected in three states: Kentucky, Louisiana, and Mississippi. My ratings, originally issued in September and updated just before the election), gave Republicans a big edge, mostly based on incumbency and the fact that these are three rather conservative states. I predicted that the GOP would hold onto the 13 constitutional offices they occupied, while Democrats would retain three of their five. The other two, both in Kentucky, were Tossups:
What ended up happening in reality? Like last year, the elections across the board all leaned more Republican than I expected. But unlike last year, none of my ratings were overtly wrong:
- Democrats won Mississippi AG, the only race I rated as Likely Democratic.
- Democrats won two out of two races I rated as Leans Democratic.
- Republicans won both Tossup races—a function of a Kentucky electorate that was a lot more conservative than everyone was expecting.
- Republicans won Kentucky agriculture commissioner, the one race I rated as Leans Republican.
- Republicans won Louisiana LG, the one race I rated as Likely Republican.
- Republicans won 11 of 11 races I rated as Solid Republican, including four races where no Democrat ran.
There's a little hiccup in the middle of the spectrum, where my picks from Tossup through Likely Republican are inverted from what they should be. Louisiana lieutenant governor makes sense as Likely Republican, but Kentucky treasurer and agriculture commissioner clearly were imprecise calls. They ended up being closer to Solid Republican, although there would have been no evidence to support such a rating before the election. Yet again, these results reinforce how unpredictable downballot races can be; compared to Senate or governor elections, there is precious little data available about them from which to make a prognostication. Because of their low profile, they're also sensitive to huge fluctuations caused by variations in turnout, the late movement of an above-average number of undecided voters, and/or wrong assumptions about the top-of-the-ticket race, like in Kentucky. Despite the wide deviation, paradoxically I think these results actually confirm my instinct to err toward Tossups. If an election has a wide range of possible outcomes, it's more uncertain than ever, and isn't that what the Tossup rating is meant to convey?
Overall, 2015 was a step forward for my constitutional office ratings—they were both more accurate and more comprehensive, covering every downballot statewide race up this year. Stay tuned in 2016, when there will be many more such elections and I'll strive to rate them all.