President Obama (AP Photo/Stephan Savoia) President Obama (AP Photo/Stephan Savoia)

Everyone is talking about Barack Obama’s falling numbers this week, with a new Pew Poll out today showing him at 41 percent approval, and a slide in Gallup to a post-election low earlier this week. Any careful look should go to the aggregators; the latest HuffPollster average has Obama’s approval at 43%, with a steady slide since a post-election peak back in late December.

Of course, Obama is never going to appear on a ballot again. But his popularity still matters. In some ways, and not in others.

What’s hurting the president? A week of high-coverage negative publicity on the Obamacare rollout can’t help, and in fact his approval rating from Pew on health care is down to 37 percent. On the other hand, that’s only down eight points since January, compared with an 11 point overall drop.

It’s not likely Obama is getting dragged down just by health care. It’s probably a combination of things — a large part of it likely from the dramatic drop in Gallup’s economic confidence index since May, which is about when sequestration started kicking in and the deficit started shrinking rapidly.

Here’s where Obama’s approval numbers do not matter: The future of the health law.

There will be a lot of talk that Dems are running away from the health law because of his drop. But on health care, Democrats pretty much are all in the same boat: They want the law to work. After all, what other choice is there? Dems are not going to support repeal — they voted for the bill. So they’re stuck hoping the administration can get everything running as painlessly as possible. Democrats probably are open to small fixes to immediate problems, but even there it’s not clear how much can be done legislatively — and no fixes that would help Obamacare run better can pass the GOP-controlled House.

Here’s where Obama’s approval does matter.

Presidential approval has real effects on midterm elections. Right now, what matters is perceptions among elites, and potential candidates, about those elections. There’s some evidence Dems benefited from the shutdown, with a small wave of successful recruitment. If the conventional wisdom shifts to a sense that Obama (and Democrats) are doomed, it’s unlikely Democrats could build on those successes. We might see some Republican recruiting coups. Separate from that is the direct effect of presidential popularity; the better Obama is doing in November 2014, the better Dems can expect to do.

The other reason presidential approval matters is that it should, on the margins, help or hurt his ability to influence people, whether Members of Congress or people in executive branch agencies or any of those Washingtonians (in Richard Neustadt’s language) a president seeks to influence.

Low presidential approval now could also matter to Obama’s long term success. If he remains unpopular, he might have more trouble getting judges confirmed, perhaps leading to more court decisions against his programs. Another area: the more popular Obama is, the more pressure there would probably be on states to join the Medicaid expansion.

What can Obama do to revive his standing? All the talk is about health care, but the best thing he could do to revive, even more than fixing, is to do what he can to make today’s good jobs number the beginning of a solid trend.