Gallup has released President Obama's approval numbers for 2013 for the 50 states and the District of Columbia. I thought it would be interesting to compare those percentages with Obama's percentage of the 2012 popular vote in each state and D.C. His approval ratings seem to track closely with his percentage of the vote: His job approval nationally was 50 percent in the first week of November 2012, and he received 51 percent of the popular vote. Gallup conducts many interviews over the course of a year, and so its figures for each state are statistically meaningful, although I think that the error margin in small states is probably pretty large.

With that caveat in mind, let’s look at where Obama’s job approval fell significantly from 2012 to 2013 and where it remained relatively steady — or remained identical (in one state) or improved (in three). Nationally, Obama’s Gallup job approval in 2013 (46.5 percent) was down 4.5 percentage points from his 2012 popular vote percentage of 51.

Let's start off with the 2012 target states, including the quasi-target states of Michigan, New Mexico and Pennsylvania. I have rounded off the changes to integers, since tenths of a percentage point tend to be distracting and are without statistical significance.

Colorado: -9 percentage points

Florida: -3 percentage points

Iowa: -10 percentage points

Michigan: -6 percentage points

Nevada: -7 percentage points

New Hampshire: -8 percentage points

New Mexico: -8 percentage points

North Carolina: -8 percentage points

Ohio: -8 percentage points

Pennsylvania: -9 percentage points

Virginia: -4 percentage points

Wisconsin: -7 percentage points

The reader will note that Obama approval has dropped farther (percentage point-wise) than the national average in every one of these states except Florida. That seems plausible: The Obama campaign spent great effort boosting his job approval there in 2012, and that effort has not been sustained in the nonelection year of 2013. This is good news for the Republicans, especially because there are Democratic Senate seats up in six of these states (Colorado, Iowa, Michigan, New Hampshire, North Carolina and Virginia), which Republicans seem to have some chance of winning. There are also nationally significant and probably seriously contested governor races in Colorado, Florida, Michigan, New Mexico, Ohio, Pennsylvania and Wisconsin.

Is there any offsetting good news for Democrats? Yes, probably. Let’s look at the states where Obama's approval is even with or ahead of his 2012 percentage.

Arizona: Same

Georgia: Same

Idaho: +1 percentage point

Nebraska: Same

Texas: +4 percentage points

Utah: +2 percentage points

The potential good news for Democrats is that Obama's steadiness (or gains) in the polls may be due to increasing percentages of blacks and/or Hispanics in the largest of these states: Arizona, Georgia and Texas. That would suggest better prospects for Democrats in state and presidential contests in the future, in states with 55 electoral votes currently and almost certainly more after the 2020 census. I have a simpler explanation for Idaho and Utah, the two states with the largest Mormon percentages: Their percentages for Obama were unusually low in 2012 because of enthusiasm for their co-religionist Mitt Romney. Now that Romney has left the stage, they are reverting, slightly, to normalcy, which still leaves them very heavily Republican. I'll let readers come up with their own explanations for Nebraska. Warren Buffett?

Obama's numbers don't seem to have changed much in the South, except in West Virginia (-10 percentage points), where resentment over his “war on coal” is raging, and in the two target states of Virginia and North Carolina. This is predictable if we suppose that black voters' approval of Obama remains high, and white voters' approval was always low.

Obama's numbers seem to have slumped significantly in wet and cold country: Oregon (-9 percentage points) and Washington (-9 percentage points) in the wet Pacific Northwest, and Maine (-11 percentage points) and Vermont (-10 percentage points) in the cold north of New England. Does this mean their electoral votes are now in play? Obama's approval is well below 50 percent in three of them, which currently have 23 electoral votes; it remains high in one of his bastions, Vermont, as it does in two others bastions, despite some slump from 2012: his birthstate of Hawaii (-9 percentage points) and his current residence, the District of Columbia (-10 percentage points). In the District, his approval remains near-unanimous at 81 percent.

In 2012, Obama carried 26 states and D.C. with 332 electoral votes. Gallup's 2013 numbers currently show him above 49.9 percent (his lowest winning percentage in any state in 2012) in only 11 states and D.C., which would give him 163 electoral votes. They're all coastal states, if you count the coasts of Lake Michigan, Lake Champlain (classified as a Great Lake by one act of Congress) and the Potomac River: Vermont, Massachusetts, Rhode Island and Connecticut in New England; New York, New Jersey, Delaware, Maryland and D.C. just a little farther south; Illinois in the Midwest; and California and Hawaii on the Pacific. Obama approval rounds off to 47 percent or 48 percent in four more states with 67 electoral votes, states that thus seem well within Democratic reach -- Florida, Michigan, Minnesota and Washington. Add those states to those where his approval is above 50 percent and you have 230 electoral votes --competitive in a presidential race, but 40 votes short of a majority.

What’s my conclusion? To me, these numbers cast doubt on the analyses that assume that the next Democratic nominee is favored to carry just about every state that Obama did. Obama carried all 12 of the target and quasi-target states listed above in 2008 and in 2012 carried all but North Carolina, where he lost 50 percent-48 percent. The apparent drop in Obama approval between 2012 and 2013 suggests that, if the presidential election were held at some point in 2013, all those states would be up for grabs — and a Republican nominee might be leading in most or all of them. But the numbers also raise the possibility that Arizona, Georgia and Texas may become genuine target states. It would be quite a change if Michigan and Pennsylvania (Democratic states presidentially starting in 1992) leaned Republican and Texas (a Republican state presidentially starting in 1980) leaned Democratic.

Remember also that Obama’s current job approval is 43 percent, 3.5 percentage points below 46.5 percent, the Gallup national average for the whole of 2013. The electoral map has been pretty static in the last two presidential campaigns and has not changed a lot over the last four. These numbers suggest that more changes are possible, and that the advantage that Democrats have held in the Electoral College over the last two decades may not turn out to be eternal.

Note: This story was originally published at 6:50 p.m. on Jan. 29. It was updated at 3:05 p.m. on Jan. 30 to clarify percentage versus percentage points.