But you know what happens to you and me when you assume things, so I checked it out. Turns out I was mostly right. I’m going to put together a summary of information that got reported about this race, but it got me wondering about this question: How much information – and what kind – of information do North Carolinians need about downballot statewide primary races? Are they getting? From where? Or why not?

After all, if journalism’s worth saving it’s only because of the impact it has on public life. I’ve long been curious about the connection between information and citizen participation. The presumption – not always right, as Samuel Popkin and Michael Schudson might tell you – is that the more information voters have the “better decisions” they will make.

A little more than 700,000 people voted in those races. Some of them might have wanted more information than others? How many had enough? How many would have changed their votes if they had had different information?

And, if we can figure out who needs this information – and what information they need – is there any business model that gets it to them? Do we need independent reporting on downballot races like this or is informing voters the job of the State Board of Elections and the candidates themselves?

Election returns are commodity information — news that people primarily differentiate by price. In the case of free online news sites, the monetary price is zero. So, really, people only differentiate commodity news by the effort of finding and consuming it.

Everyone has election returns. The AP sells a very good turnkey solution and most state government Web sites — state governments for crying out loud! — post the results faster than anyone.

But results are seen as something that a news Web site just has to have. A few — like The Charlotte Observer on Tuesday night — just punt and link to the state Board of Elections. I’m glad most news sites do see this obligation. Online election returns put food on my table at washingtonpost.com for a few cycles.

So here’s the deal — legit news sites spend a lot of resources on these one-shot collections of commodity news. They see a spike in traffic on election night, but when it comes to bang for your buck, returns probably don’t match many of the other features a site will build during an election season in its attempt to differentiate its coverage from all others.

It is a game you have to play, but it’s a game you can only lose. A small hiccup in the system that’s spitting out numbers on to your site, or a small design flaw or browser incompatibility sends users clicking off to another site. In the North Carolina primary, I started my hunt for numbers on newsobserver.com, but couldn’t find the percentage of precincts reporting. WRAL.com had that small bit of information, so I moved over to them.

One way sites could get more value out of their election night coverage is to build them in to an evergreen database of political information. Build up an easily navigable collection of votes (and exit polls… you ever try to find exit poll data more than three months after an election? Impossible.) and a site can ride the long tail until the next Bush administration.