Since there’s been some relatively good polling news for Democrats here (Kansas) and there (Georgia, Arkansas) of late, I suppose it’s time for a dose of skepticism about the likelihood that a Democratic presidential party (bad) in a sixth-year (bad) midterm (bad) with an adverse Senate landscape (bad) and low presidential approval ratings (bad) can avoid a pretty bad spanking.

Carefully as always, RealClearPolitics’ Sean Trende offers a “theory” of how things might deteriorate for Democrats between now and November 4, and it’s ingeniously based on the idea that the supposedly warring “fundamentals” (indicating a big GOP landslide) and polls (much more ambivalent) will eventually converge, as they did in previous congressional elections since 2002.

From 2002 to 2012, candidates of the president’s party have tended to converge on the president’s job approval. It isn’t an absolute tendency, but it is nevertheless real.

The Democrats’ problem is that they seemingly find themselves in a position similar to that of Republicans in 2006: They are in tight races. But so far, they seem unable to move past where the fundamentals suggest they should be able to go: Recall again that their maximum showing has generally been bounded at 47 percent.

This is true even though Democrats have generally dominated the air wars recently. They’ve succeeded in driving down Republican numbers, or holding them in check. But they haven’t improved their own. Here, the Peters-Land race in Michigan is instructive. In early September, Gary Peters’ lead was 3.8 points. Today it is 5.4 points. But his numbers are largely unmoved: 45.3 percent at the beginning of the month, 44.3 percent today. Terri Lynn Land’s numbers, however, have tanked: from 41.5 percent to 38.9 percent.

But we know that this race will not have a 44.3 percent-38.9 percent outcome — sooner or later the undecided voters will begin to decide. And given that the Democrats are winning the votes of almost everyone who approves of the president’s job, they will have an uphill — though hardly insurmountable — battle with undecided voters.

Now aficianadoes of polling may note right away that with respect to key Senate races in AK, AR, CO, LA, and NC, Trende’s theory has much the same effect as the increasingly-discredited “incumbent theory” which held that late-breaking undecided voters generally go heavily for challengers (it sure didn’t happen in the 2004 presidential election, did it?). In his case, though, the controlling factor isn’t incumbency but the approval rating of the president–and it would hurt Democrats, of course, just as much in states with low Obama approval ratings where their candidate is not an incumbent (e.g., GA, IA, KY).

It would obviously make a big difference (under Trende’s theory) or at least some difference (in virtually everyone’s estimation) if Obama’s approval ratings generally improved between now and November 4. It’s almost universally thought to be too late for some economic development–e.g., a boffo September Jobs Report–to make a difference. A possible “rally-round-the-flag” effect if the military campaign in Syria and Iraq goes well and blots out the sky is another thing, though not a whole lot of experts think that’s likely to happen. Theories, of course, are just theories, and even if they seem valid there are exceptions to every “rule,” so no one candidate or his or her supporters should be unduly encouraged or discouraged by the possibility of late national influences on any individual race. Trende (and others) have not factored in the possibility of a significant change in turnout patterns that would make past midterm results less predictive. But all in all, it’s looking like if Democrats do hold onto the Senate, some serious congratulations are in store for those who won in such adverse circumstances.

Ed Kilgore

Ed Kilgore is a political columnist for New York and managing editor at the Democratic Strategist website. He was a contributing writer at the Washington Monthly from January 2012 until November 2015, and was the principal contributor to the Political Animal blog.