Are Latino voters finally “souring” on the Democratic Party?

That’s the sweeping conclusion that some national political observers are coming to in the final days before the 2014 midterm elections.

For years demographers have predicted that America’s surging Latino population would be a boon to Democrats in future elections. They’re the largest minority group in the country (and usually the fastest-growing)! Seventy-one percent of them voted for Barack Obama in 2012 (versus only 27 percent for Mitt Romney)! Pretty soon Latinos will be turning Georgia and Texas and Arizona blue!

But now? Never mind.

“Burned on immigration reform, Latinos start to pull away from Democrats,” writes Greg Giroux of Bloomberg Politics, referring to Obama’s decision last month to postpone executive action on immigration until after Nov. 4. “Latino voting tends to drop in midterm elections,” adds Mark Z. Barabak of the Los Angeles Times — and “that may prove all the more so next month, given deep frustration with the president.” And Michael Barone of the Washington Examiner sees “evidence” in “the crosstabs of polls and in party strategists’ moves” that “one group Democrats have been counting on is moving away from them: Hispanics.”

There’s only one problem with this conclusion — and it’s the same as the problem with pretty much every other sweeping conclusion that national political observers come to from the comfort of their cubicles in the heady days right before an election, when content must be produced on a near-constant basis.

It’s so premature that it’s practically meaningless.

To see why, look no farther than Colorado — the only truly competitive state this cycle where the Latino share of the eligible electorate (14.2 percent) exceeds the national average (10.7 percent).

At first glance, the Centennial State seems like the perfect place to search for evidence to support Barone & Co.‘s thesis. There are a lot of Latino voters (unlike pretty much everywhere else this year). The big Democratic candidates are struggling—especially incumbent Sen. Mark Udall, who trails his Republican rival, Rep. Cory Gardner, by an average of 2.8 percentage points. Ergo, Latinos are allergic to Dems all of a sudden. Or so the conventional wisdom goes.

That’s certainly the case Barone & Co. would like to make. In Barone’s column, he notes that Udall isn’t “close to Obama’s 75 percent” share of Colorado Latino voters in the latest polls and that Gardner is “running at or above Romney’s level.” “One poll,” he adds, “even has the Latino vote evenly split.”

Unfortunately, the reality in Colorado is a lot more complicated. It’s true that Udall isn’t matching Obama’s whopping vote share among Latinos. But why are we using Obama’s 2012 presidential blowout as a benchmark? Wouldn’t a previous Senate election make more sense — like, say, the one that Udall won in 2008? That year, Udall captured 63 percent of the Latino vote. His Republican opponent got 30 percent. And Udall wound up winning the race by more than 10 percentage points.

So how is Udall doing with Latino voters this year? It’s tough to say for sure, because Colorado Latinos are notoriously difficult to poll. A recent Denver Post survey shows Gardner leading him among Latinos by 14 points (a result that no one in Colorado believes); the latest Suffolk sounding gives Udall a 73-point advantage (which no one in Colorado believes, either).

The truth is somewhere in the middle. The firm Latino Decisions focuses exclusively on polling hundreds of Latino voters at a time — Spanish speakers and English speakers, cellphones and landlines, young and old — so its readings tend to be more accurate than ones extrapolated from public polling crosstabs, where the sample sizes are too small to be representative. In the latest Latino Decisions survey, 66 percent of Latinos say they will or are likely to vote for Udall, while only 17 percent say the same about Gardner. Another 17 percent are undecided.

Meanwhile, “the past five public polls with a Latino crosstab found Udall leading by 28 percentage points among Latinos, on average” — which mirrors what the Udall campaign has been seeing in “months and months of internal polling,” according to a Democrat familiar with those numbers. “We’ll win Latinos by 30 points,” the source tells Yahoo News. “There are a lot of bad polls out there.” In short, Udall appears to be on track to perform almost exactly as well with Latino voters as he did in 2008 — which suggests the group isn’t abandoning the Democratic Party just yet.

But what about turnout? Is it possible that Latinos are so unenthusiastic about the Democrats at this point that they’ll just stay home on Election Day?

Sure, it’s possible. Among Latino voters surveyed by Latino Decisions, the Democratic Party’s approval rating on immigration fell by 12 points (from 49 to 37 percent) between June and October, thanks in large part to Obama’s decision to delay executive action. And on the national level, more Latinos say they were enthusiastic about voting in 2012 (40 percent) than say they are enthusiastic about voting in 2014 (32 percent).

But of course voters no one’s paying attention to are going to be disengaged; the true test is how they act when they’re being courted. And in Colorado — again, unlike pretty much anywhere else this year — Latino voters are actually playing a central role in the election, just as they would in a presidential contest. Udall’s massive field operation, which is three times as large as the last Democratic senatorial effort in Colorado, has been targeting Latinos from the start. The Republican National Committee, meanwhile, has been “identifying and engaging community leaders in Hispanic areas, to create a network of support for Gardner and other Republican[s].” And the candidates have been sparring over immigration for months.

As a result, 43 percent of Latinos tell Latino Decisions that they’re more enthusiastic about voting in 2014 than they were in 2012; only 34 percent say the opposite. Convincing low-propensity voters like Latinos to participate in midterm elections is always a challenge for Democrats. But voters everywhere are apathetic this cycle, and so far there’s little evidence that Colorado Latinos are particularly disinclined to show up at the polls on Nov. 4.The opposite may even be true. Udall’s support among Latinos is steady, and there’s no reason to think they’ll swing the election to Gardner by staying home.

None of which is to say that Colorado Democrats will win on Nov. 4 — just that they aren’t likely to lose because Latinos sat on their hands or flocked to the GOP. “Republicans have done years of work to alienate Latino voters,” says David Damore, an associate professor of political science at the University of Nevada, Las Vegas, and a senior analyst at Latino Decisions. “So Democrats are still getting the benefit of the doubt, especially in places where they’ve really attempted to engage the community.”

The bottom line is that if you can’t argue that Latinos are “souring” on Democrats in Colorado — the only state where they’re a big part of the electorate this year — it seems hasty to argue that they’re souring on the party in general. Sometimes the tea leaves have a good story to tell. And sometimes they don’t say anything at all.