Why the win for Republicans in Arizona 8 is still good for Democrats

(CNN)CNN has projected that Republicans won the special election in Arizona's 8th Congressional District, but once again Democrats have overperformed in a deep red district, in another sign the national environment is in the their favor.

The vote count is not final, but Republican Debbie Lesko is currently ahead of Democrat Hiral Tipirneni by about a 5-point margin.
In a neutral environment, the margin should be much wider. President Donald Trump won the district by 21 percentage points in 2016 and Mitt Romney won it by 25 percentage points in 2012. Combining those outcomes and controlling for how well Democrats did nationally in each of those contests, we can say that Arizona 8 is 25 points more Republican than the nation. Lesko looks like she's going to do about 20 percentage points worse than that.
The result in Arizona 8 fits a pattern so far in special congressional elections this cycle. In every one of the nine so far, Democrats have outperformed the partisan baseline based on the prior two presidential elections.
    Including Arizona 8, the average improvement for the Democrats has been 17 percentage points versus the partisan baseline. That's better than any party out of power has done in the lead-up to a midterm cycle since at least 1994.
    You'll note that there is a clear correlation between the average overperformance of Democrats in special elections and the midterm outcome. The only cycle that looks like this one so far is 2006, when Democrats had a net gain of 30 seats and took back the House.
    But while Arizona 8 looks pretty much like the average special election so far, there are three reasons why we might have expected Republicans to do significantly better than average.
    First, this race plus the one in Pennsylvania's 18th District last month occurred when the Democrats' position on the generic congressional ballot has weakened significantly from the end of the last year. The continuing Democratic surge in returns in two districts on the opposite sides of the country would suggest that the shift in polling results may be statistical noise more than an actual change in preferences.
    Second, Republicans turned out in this election. The relative difference between Democrats and Republicans in registration among those who voted was about equal to overall registration figures. The number of people who voted in the special is fairly close to the number who voted in the the district during the last midterm election, in 2014. That's not surprising because it is easy to vote early and by mail in Arizona. This allowed Republicans, who perhaps might have be been uninspired, to cast ballots without too much hassle.
    It also means, however, that poor turnout is not an excuse for Republicans in this race. One common reason to be cautious of the special election results so far has been low turnout. Yet this election, like Pennsylvania 18 last month, saw turnout close to or exceeding 2014 levels, and Republicans trailed greatly behind the partisan baseline of these districts.
    Finally, Republicans had a good candidate in Lesko. She had no major scandals and raised plenty of money. One of the excuses in previous elections that Republicans lost like Alabama US Senate (with Republican Roy Moore) and Pennsylvania 18 (with Republican Rick Saccone) was that the Republican was either scandal plagued or didn't know how to raise funds. Lesko wasn't either of those, and there was still a significant shift to the left.