Skip Navigation

Close and don't show again.

Your browser is out of date.

You may not get the full experience here on National Journal.

Please upgrade your browser to any of the following supported browsers:

The Statistical Significance of Sandy Could Alter Electoral, Popular-Vote Math The Statistical Significance of Sandy Could Alter Electoral, Popular-V...

This ad will end in seconds
Close X

Want access to this content? Learn More »

Forget Your Password?

Don't have an account? Register »

Reveal Navigation



The Statistical Significance of Sandy Could Alter Electoral, Popular-Vote Math


After canceling a morning campaign rally in Orlando, Fla., President Barack Obama walks into the White House in a driving rain after returning to Washington to monitor preparations for early response to Hurricane Sandy, Monday, Oct. 29, 2012.(AP Photo/Jacquelyn Martin)

One of more absurd notions to crop up in the latter stages of the presidential campaign is that GOP challenger Mitt Romney could  win the popular vote and lose the Electoral College vote.

Until Hurricane Sandy, this was a cable TV notion in search of historical and mathematical mooring. That it bobbed aimlessly through the occasionally mindless waters of talking heads made it no different from any other might-this-happen adventure in televised banality.


But think about it for just two seconds.

The concept is built on the theory that Romney could run up the vote in Southern and high Plains states and get close enough to Obama in the swing states to win the popular vote but fall just short of the magic 270 Electoral College number for victory.

I know from where this myopia springs. It would be hard in this campaign to remember that California, Illinois, Maryland, New Jersey, and New York are still part of the Union and their votes count. We remember Massachusetts only because Romney's headquarters is there. But we forget about it entirely in the context of this Romney-wins-the-popular-vote-only lunacy.


Consider this simple demonstration of Clintonian arithmetic. Here are the margins of victory in actual votes in 2000, 2004, and 2008 for the Democratic candidate in the following big-population blue states: 
California: Al Gore over George W. Bush by 1.3 million; John Kerry over Bush by 1.2 million; Barack Obama over John McCain by 3.2 million.

New York: Gore over Bush by 1.7 million; Kerry over Bush by 1.3 million; Obama over McCain by 2 million.

New Jersey: Gore over Bush by 504,000; Kerry over Bush by 241,000; Obama over McCain by 602,000.

Maryland: Gore over Bush by 331,000; Kerry over Bush 308,000; Obama over McCain by 669,000.


Massachusetts: Gore over Bush by 737,000; Kerry over Bush by 732,000; Obama over McCain by 795,000.

Illinois: Gore over Bush by 569,000; Kerry over Bush by 545,000; Obama over McCain by 1.3 million.

What about Texas, you might say. Wouldn't Romney roll up huge numbers there, partially off-setting these big blue states?

See for yourself. Texas: Bush over Gore by 1.3 million; Bush over Kerry by 1.6 million; McCain over Obama by 950,000. Romney will roll up a big number in Texas. But as the data show, the number fell by 600,000 votes from '04 to '08 and was 300,000 lower than 'the 00 cycle. It is likely to be somewhat smaller this year due to Texas's changing demographics and ideology.

Ask yourself this: Has Texas become more or less hospitable to Democrats since 2008 and does it still offset all the blue states? What about California, Illinois, Maryland, Massachusetts, New Jersey, or New York? Of these, only New Jersey can be argued to have taken a slightly more right-leaning slant since 2008. Even so, it has, on average, given the Democratic nominee 449,000 more votes than the Republican in the last three cycles.

Don't kid yourself that turnout would be lower in these states this time because they aren't swing states and don't receive the media attention or candidate visits. That was just as true in 2000, 2004, and 2008 as it is now. And the popular-vote margins in these states are enormous and will loom large for President Obama. 
That means Obama won't have a popular-vote problem with Romney. He may have an electoral college problem, though.

Imagine a scenario in which Romney edges Obama by 100,000 in Ohio, 30,000 in Iowa, 15,000 in New Hampshire, and 50,000 in Virginia. That's 41 electoral votes with a microscopic edge of 195,000 votes in four states. That 195,000 would be slightly more than a third of the average Democratic margin since 2000 in New Jersey and one-tenth of the average Democratic winning margin in California.

Here's where Sandy comes in and could make a profound difference in terms of the popular vote and electoral vote.

Pennsylvania took a hit from Sandy. It wreaked havoc in Philadelphia and the eastern part of the state. Even before Sandy struck, the Obama campaign announced it was going up with TV ads to protect a statistically small lead. Obama knows he has to run up the vote in Philadelphia and outperform Romney in Bucks, Chester, Delaware, and Montgomery counties in the Philadelphia suburbs to offset Romney's dominance in western Pennsylvania. 
Mobilizing voters in Philadelphia will be crucial and a big post-Sandy challenge. How important are these votes for Obama?

In south-central Philadelphia's 1st Congressional District, Obama beat McCain by 222,664 votes. Kerry beat Bush there by 179,818. In western Philadelphia's 2nd Congressional District, Obama beat McCain by 267,250 votes. Kerry beat Bush there by 228,363.

From two Philly districts, Obama extracted a popular-vote advantage of 489,914. Obama carried Pennsylvania by 620,478. Seventy-nine percent of Obama's statewide margin came from the 1st and 2nd congressional districts. Those votes and the ability of those voters to navigate city streets and deal with the post-Sandy deluge may loom large in the battle for Pennsylvania and its 20 electoral votes. This will also be true in Bucks, Chester, Delaware, and Montgomery for both campaigns. But inner-city Philadelphia is vital to Obama.

Obama might also absorb popular-vote losses disproportionate to his 2008 performance or the Kerry and Gore standards in Massachusetts, New Jersey, and New York. Together, these states produced an Obama victory of 3.4 million votes. They gave Kerry a 2.3 million edge and Gore a 2.7 million margin. The three-cycle average of Democratic nominee victories in these three states combined is 2.8 million votes. 

Storm-diminished turnouts in these three states could cost Obama tens of thousands of popular votes. It could also cost him 20 electoral votes in Pennsylvania. The implications are obvious in Virginia as well, but that state was always going to be close and the margin of victory understood to be narrow. There are ways Obama can win without Virginia but not many without Pennsylvania.

The chance of a Romney popular-vote victory and Obama Electoral College victory were always statistically and mathematically remote. The chances of the opposite occurring were always easier for me to see. And Sandy may alter that terrain in ways that prove more harmful to Obama than Romney.


comments powered by Disqus