NY-26: County Benchmarks (and Predictions Contest)

Longtime fans of Swing State Project/Daily Kos Elections are probably familiar with the game-day staple of county benchmarks, a tool to help enjoyably make sense of the torrent of information (concerning a bunch of counties you probably haven't heard of without the context of their usual lean and how big they are) that comes at you while watching election returns. The method is very simple (probably too simple): use the last election's results as a jumping-off point for recalculating what's needed, county-by-county, to get the Democrat over the 50% mark.

There are a couple caveats, though. One is good news: unlike many small-scale elections, where provincialism becomes a factor and candidates have particular strengths in certain counties (think back to the NY-20 special election in 2009), this one is easier, in that all three candidates are from Buffalo's outer suburbs in Erie County. So there's not as much need to assume counties will perform differently from the more generic presidential results. The other is bad news, and is pretty obvious: this is a three-person race, and how the heck do you map a basic head-to-head onto a three-way?

Since we're getting a pretty harmonious view of the race from the two pollsters who polled this weekend (Siena and PPP), and there haven't been any ground-breaking implosions in the last few days (as was the case in NY-23), I feel fairly confident in expecting Jack Davis to retain the 12% he's polling at. So, I just used a two-step process of extrapolating 2008 presidential results to a scenario where the Dem just hits 50%, and then shrinking that to fit into a space where the Dem + GOP votes are only 88% of the total. I have no idea if that'll prove useful in reality, but it's a starting point; if Davis starts to vary widely from that 12%, I'd say that's good news for Hochul if he's going higher and bad news for Hochul if he's dwindling into single digits.

(Note that I also included the 2008 House race results, between Democrat Alice Kryzan and the recently-resigned Chris Lee, as a reference point. I didn't feel the need to do another model based on those numbers, because Kryzan very consistently underperformed Barack Obama by a uniform 5-6 points in each county of the district.)

The Siena poll from this weekend (where Hochul led 42-38-12 overall) has helpful breakdowns by groups of counties. Hochul leads significantly in Erie County 50-38-9, while Corwin is up slightly in Monroe and Livingston Counties (the Rochester market), 40-36-12, suggesting Corwin is overperforming my baselines there by a few points while Hochul is slightly overperforming in Erie. The rurals may turn out to the swing area that we have to watch (Genesee, Niagara, Orleans, and Wyoming, although that's a weird grouping because the part of Niagara County that's in the district is very much a part of the Buffalo suburbs): Hochul leads 37-36-17.

For some more context, here's the map of the district:

Well, now that you're armed with all that information, what are you going to do about it? Join in our predictions contest, that's what! As an added incentive, we're returning to a tradition we haven't done in a while, and offering a delicious Green's babka to the winner. All you need to do is predict the vote share of each of the four major candidates... and be specific to one-tenths of a percent. Who ever, cumulatively, gets the closest wins babka. The contest is open only to those with dKos user names: here's the form. (Although, of course, feel free to post your predictions in comments too.)