Thursday is the deadline for presidential campaigns to submit their third quarter fundraising reports to the Federal Election Commission. Many campaigns -- confident in their numbers or hoping to inoculate the response to bad ones -- have already released top-line details. For all of them, though, it's a moment that's not unlike heading to the doctor for a particularly invasive examination.

There's one bit of data, the result of a simple calculation, that outside observers will be paying close attention to: burn rate. A campaign's burn rate is how much it raised versus how much it spent -- in other words, how much of its money it has already "burned." It's helpful here to revert to the campaigns-are-like-businesses simile. If a business is spending much more than it is taking in, it is unlikely to survive for long. If that business sees large expenses on the horizon, it needs to be saving even more.

These presidential campaigns should be seeing large expenses on the horizon -- the actual elections. A look at the recent history of presidential campaigns suggests that the eventual winners had lower burn rates -- and therefore a higher percentage of cash socked away -- through the first three quarters of the year before an election than the eventual losers.

The calculation here is simple: Amount spent divided by amount raised. If a campaign spent one million and raised two million, the burn rate is 0.5, or 50 percent. If it spent two million and raised one million, the burn rate is 2, or 200 percent. Or, looking at it another way, spending was two times the fundraising. Using FEC data for each winner, loser and incumbent since 2000, here's how the quarterly burn rates (and the figure for January of the election year) compare.

Winners save more money -- until the fourth quarter before a presidential year, when they start spending it. This makes sense. After all, better campaigns have more money saved, so by the time the end of the year rolls around (and they need to start investing heavily in TV ads in Iowa and New Hampshire) they have more to spend. The ratio spikes, because what they're spending is money they saved earlier. (Incumbents, who haven't faced real primary challenges this century, spend even less.)

But this is not uniform, even over the past four election cycles.

In 2000, George Bush and Al Gore ran relatively easy primaries, saving early and spending late. In 2004, John Kerry had a tougher time of it. In 2012, Mitt Romney managed to keep a manageable burn rate throughout the prior year.