Gun control advocates and President Obama will be pleased about Wednesday's bipartisan compromise broadening criminal background checks for gun sales. Opponents criticize the move, suggesting it wouldn't have prevented several high profile mass shootings. Indeed, of the 30 incidents since 2003 that we looked at, the Senate deal would quite possibly only have stopped one.

The compromise, between Senators Joe Manchin of West Virginia and Pat Toomey of Pennsylvania, would introduce and expand criminal background checks at gun shows and for online sellers. The NRA expressed its opposition to the bill, indicating that it wouldn't have "prevented the tragedies in Newtown, Aurora, or Tucson." In an overview of typical objections to expanded background checks, ABC News also noted the common argument that criminals don't submit to background checks.

Both of those arguments are true. Both are also very narrow.

We decided to look at what recent history actually shows. Largely relying on Mother Jones's guide to mass shootings, we identified 30 incidents over the past decade, stretching back to July 2003. (The magazine's definition requires a lone shooter who killed three or more people in a public place. We added one incident: the 2012 high school shooting in Chardon, Ohio.) We looked at how the weapons used in each incident were acquired, and, to the extent possible, the psychological and criminal history of the shooters.

The question we wanted to answer: How many of the shootings over the last decade involved weapons that would be subject to the Manchin-Toomey background check law? How many, that is, would have required a federal background check where one doesn't exist now (or didn't at the time), and where the shooter would possibly have failed. 

This one:

In three other instances, there wasn't enough information to determine if the law would have prevented the crime since how the weapons were acquired is unclear.

  • Robert Stewart, Carthage, Mississippi, March 2009.
  • Eduardo Sencion, Carson City, Nevada, September 2011.
  • Jeong Soo Paek, Atlanta, Georgia, February, 2012.

Here's how all of the data broke down.

Few sales would have been prevented.

In 26 cases, the proposed law wouldn't have affected the outcome.

Most weapons were legally purchased from gun stores.

More than half of the weapons used in the crimes were purchased from gun stores. In two cases — Goleta in 2006 and Salt Lake City in 2007 — the purchases took place at a pawn shop. Four times, including Newtown, the weapons were stolen. (In one case, the weapon was acquired through the shooter's employer, a law enforcement agency.)

In fact, more than half of the cases in which it was possible to determine, the shooter underwent a background check prior to accessing his (or, once, her) weapon.

One incident, the 2007 shooting by Sulejman Talovic in Salt Lake City, suggests how tricky this analysis can be, even once the crime has been committed. KSL.com reported at the time:

The suit claims that under federal law, the pawn shop was not allowed to sell that type of weapon to Talovic.

"This is an illegal gun sale. They sold it to a dangerous person who never should have got it," said Daniel Vice, lead attorney for the Brady Center. "Federal law has been clear for four decades."

Mike Hansen, Hill's attorney, argued that under straight definitions, the weapon sold to Talovic was a shotgun and legal. Even though he bought it with a pistol-grip, Hansen said the gun was originally made to have both a pistol grip or could be fired from the shoulder, thus qualifying it as a shotgun.

Most shooters displayed mental issues. Most had no record.

In the majority of cases — as determined by our subjective, after-the-fact, media-sourced analysis — shooters apparently displayed mental issues prior to their attacks. In several cases, they'd been hospitalized prior to the incidents.

Most, however, had no criminal record before their attacks — perhaps bolstering the argument that criminals don't submit to background checks. Or, more likely, the attacks were the first criminal behavior by the shooters, many of whom were young at the time of the shooting.

It's true that the proposed law — which will be voted on by the Senate Thursday — wouldn't have prevented many attacks. But there's one more data point worth noting: 38.

Had all of the maybe-preventable sales and the 2003 sale been prevented, 38 fewer people would have been killed or injured in those attacks.

The president's argument has consistently been exactly that point.

We won’t be able to stop every violent act, but if there is even one thing that we can do to prevent any of these events, we have a deep obligation, all of us, to try.

There are dozens of families — in Atlanta and Carthage and Meridian — who'd no doubt agree.