The 2014 Midterms.

Oregon results

I'll update this number later, but right now turnout appears to be headed towards a 16 year low.

No surprises here, as Jeff Merkley won convincingly by 17 points, over Monica Wehby.  At the same time, Dennis Richardson halved the gap of the most-recent poll, losing to John Kitzhaber by 3.5 points.  I reiterate, that with the failures of Cover Oregon, it was remarkable that any Republican couldn't beat out the incumbent governor.  Add onto that the recent issues of Kitzhaber's fiance's troubles, but still pulled out the win, it's quite clear that Oregon is blue, except for rural areas.

Now I have no idea why two years ago voters turned down the opportunity to legalize marijuana, but this year, voters chose to legalize it by nearly 10 percentage points.  I suppose it required seeing the massive tax revenues in Colorado and Washington that finally convinced voters in Oregon to support legalization?

GMO labeling, as it stands right now, down by 2.5 points, is going to fail.  But the failure here, it seems, is that the anti-labeling groups (Monsanto and DuPont) overwhelmingly outspent pro-labeling groups to obfuscate what the law would have required.

National results

I have to admit, I really did think that the Democratic ground game would make up the difference in polls, but it's starting to look like Democratic voters were turned off in extremely high numbers, allowing Republicans to retake the Senate and keep much of their seats around the country in all levels of government -- well, except of course in the bluest of blue states.

I would have been right, had Democrats participated in equal levels as Republicans.  Of course, I did not expect this much disengagement.  So for instance, in 2010 (the last midterm election) in Colorado, some 73% of eligible voters participated, but in 2014, that number has dropped to around 51% (numbers are not final, of course, as they're still being counted).  Go back to 2006, and the participation rate was 62%.

In Kansas in 2010, 857K voters cast ballots -- about 50% participation rate.  In 2014, the current total is around 800K, which at this level, is about 46% participation rate.

On why I was optimistic (but wrong)

So, about half a year ago I had been looking at a data set that I hinted at, back then.  Had I actually taken the time to track the data, I would have seen what was going on this year, but at the time, the snapshot showed that Democrats were faring okay.  This data set?  Gallup's tracking of party affiliation.  Their poll goes beyond most others, by asking people which way they lean.

My theory was, when you add Republicans and Republican-leaning, and subtract Democrats and Democratic-leaning, that gap and the direction it was going, would explain which way the general election was going.  The steeper the curve, the bigger the wave, the flatter the curve, the less change.

As I said, had I actually bothered to insert the data into a spreadsheet and put them to charts, I would have seen where 2014 was going.  Let's have a look-see, shall we?

In these charts, an upward slope would mean that the trend was benefiting Republicans.  When either the raw data or trend line goes above the green demarcation (zero differential), it means that there were more people affiliated with GOP and GOP-leaning than Democrats and Democrat-leaning.   In 2008, while the trend line was headed upward, the trend line remained far below the green line -- the Democrats may have lost some support but its support was extremely strong.  The result of the 2008 elections was a complete takeover of Congress and the White House.

In 2010, the trend line started out heavily in support of Democrats, but then steeply moved in favor of Republicans, crossing the green line during the summer of 2010,  This meant that Americans were aligned with Republicans, and as a result, we had a wave election where the US House shifted back to Republicans while the Senate remained in Democratic control but the gap had shrunk significantly.

In 2012, Republicans started out with a slight edge, but then the trend line shifted toward Democrats.  The slope of the trend line wasn't as steep as 2010, but because it had shifted into Democrats' favor, Democrats not only took the White House, but gained seats back in both the House and Senate, though they did not regain control of the House.

Finally, in 2014 you can see roughly three sections, here.  In late 2012 and early 2013, Democrats continued to hold a significant advantage.  From the summer of 2013 to the summer of 2014, things were still favorable for Democrats, but by a smaller margin.  Then, towards the end of summer, the trend shifted towards Republicans.  The last data point in this selection was the middle of October.  Had Gallup conducted another poll closer to the election, I'm certain that the trend line would have crossed the green line, as the raw data had already shifted toward Republicans.  Furthermore, the slope was not as steep as 2010, but closely resembled 2012's.  In turn, Republicans gained enough seats to flip the Senate, as well as gained seats in the House.

Combine the Democratic voter lack of enthusiasm and the trend towards Republicans, it is now clear that the GOP were going to win big.

So there it is.  Had I tracked the data closely, earlier this year, I would have seen this coming.  I had seen the data shift these past two months, but the data seemed noisy.  Once I had put the data down into the spreadsheet and inserted the trend line, it was clear that the GOP would win big.

Mea culpa.