It was the GOTV All Along

What made Mitt Romney lose the election?

Well, lets consider: The incumbent, faced with an opponent from Massachusetts who wrapped up his nomination around April, went to work fairly quickly in defining him; he did some swing-state pandering to give him an edge in particular in Ohio, the crucial battleground. He knew it would be a close race, so he determined that he would have to win by getting his base out, and worked hard to do that with GOTV and with ‘retail’ appeals to specific subgroups. GOTV was key, as was telling his base what they wanted to hear so they would be excited to vote for him; war on terror successes as well as social and economic issues. His opponent never escaped the definition the incumbent gave him, and although the challenger made valid critiques of his record, somehow the people trusted that he was doing well enough to deserve more time. The incumbent won, even though he lost the independent vote.

That’s what happened … in 2004. Obama just won the election using the Bush 2004 playbook. Change Massachusetts to New York and you get the Truman 1948 playbook. Same playbook: Work your base, paint your opponent, retail politics to pick off enough supporters for a winning coalition. (And no, Virginia, Obama’s coalition is NOT permanent, and we are NOT going fullbore socialist as Obama didnt even run on that, despite the fact that the next 4 years may make us look like the next Spain or Greece.

We now know how incumbents with shaky popularity can win.  Despite Obama’s media advantage, money advantage (oh, you didnt know that? Romney had a lot of SuperPAC money on his side but it didnt really help him much, did it? campaign fund head to head had Obama with more. Obama spent almost a billion dollars to win the race, hence his year long fundraising marathon.), and the power of the White House to shell out money and create media attention, he had a sour economy wrapped around his neck like an anchor. That’s why he was losing this spring to ‘generic Republican’. Obama knew this and so his team knew they had to run the Truman/Bush style incumbent race, a scrappy, tough,  base-driven campaign.

But that’s not my question: Why did Mitt Romney lose?Why didnt  Romney see that it was this type of race and respond accordingly. Romney needed to (a) not let Obama define him; (b) defeat Obama’s panders by calling them out, rebutting them or deflating them;  (c) win over the disillusioned voters in the middle; and (d) win the base turnout competition.   Romney failed to (a) in the summer, but he made up for it in the debate #1; Romney did (b) in his campaign and he also focussed on winning over the independents; it worked. Romney’s biggest failure was (d).

One interesting change in 2012 has been the rise in the ethnic minority share of the vote:


For Republicans, that despair now comes from an electorate that seems to have undergone a sea change. In the 2008 final exit polls (unavailable online), the electorate was 75 percent white, 12.2 percent African-American, 8.4 percent Latino, with 4.5 percent distributed to other ethnicities. We’ll have to wait for this year’s absolute final exit polls to come in to know the exact estimate of the composition this time, but right now it appears to be pegged at about 72 percent white, 13 percent black, 10 percent Latino and 5 percent “other.”

With this base line, and armed with the exit-poll data, we can get a pretty good estimate of how many whites, blacks, and Latinos cast ballots in both 2008 and 2012. Assuming the 72/13/10/5 percentage split described above for 2012, that would equate to about 91.6 million votes cast by whites, 16.6 million by blacks, 12.7 million by Latinos, with the balance of 6.3 million votes spread among other groups.

Compare this with 2008, when the numbers were 98.6 million whites, 16.3 million blacks, 11 million Latinos, and 5.9 million from other groups.

In other words, if our underlying assumption — that there are 7 million votes outstanding — is correct, then the African-American vote only increased by about 300,000 votes, or 0.2 percent, from 2008 to 2012. The Latino vote increased by a healthier 1.7 million votes, while the “other” category increased by about 470,000 votes.

This is nothing to sneeze at, but in terms of the effect on the electorate, it is dwarfed by the decline in the number of whites. Again, if our assumption about the total number of votes cast is correct, almost 7 million fewer whites voted in 2012 than in 2008.

Now – why did white American voters stay home? Now, one possibility is that white Obama voters stayed home in droves. A fly in the ointment to that – look at this, the union voters stuck with Obama:

On Tuesday’s election, labor unions, as usual, played a crucial role in electing Democratic candidates. Union members voted for Barack Obama at a rate of 65% to 33%, according to AFL-CIO polling; according to official exit polls, Obama’s support in union households was 58%, which is 1 percentage point less than 2008 and pretty much par for the course for recent Democratic presidential candidates.


Consider this: 62 million people voted for George W Bush in 2004, Kerry 59 million. (that same year, 59 million voted for Kerry.) In Ohio, Bush got 2,858,727 votes.  In 2008, the electorate expanded, Obama winning 69 million to McCain’s 59 million. All the way back in 1984, Ronald Reagan won 54 million votes in his landslide win. Bush won 50 million votes in 2000.

The result for 2012 – Obama 61 million, Romney 58 million:

Popular vote 61,112,143[2] 58,122,514[2]

What do all these numbers tell us? First, the telling point – Romney had fewer votes not only than McCain but Bush in 2004 as well. How did we lose those voters? 

Obama dropped from 69 million to 61 million. 88% of his 2008 support showed up. Romney had a 2% dropoff. In a race where the incumbent lost 12% of his support in a race,  Romney not only failed to capture those voters but lost ground. This was a failure to win more votes than the previous candidate.

We believed wrongly that 2008 was the anomaly … BIG MISTAKE. The 2012 electorate included enough 2008 and not-2010 voters who came back because they like and want Obama re-elected. Not all of Obama’s voters but enough of the black/urban/young/liberal base to make it like 2008, or at least 2004. Indeed, Obama’s vote tally and Romney’s vote tally are both just below the Bush / Kerry tallies as well. A Bizarro 2004 election indeed! 

Now looking at raw numbers, THE MINORITY TURNOUT WAS AS GOOD IN 2012 AS 2008. The white vote went way down. Why? Nobody has yet to crack open why Mitt Romney had a million fewer votes than McCain … but there is something you always need to do to win an election. Now, Sandy might be a factor:

The biggest plunge by far, according to the American University analysis, came in Eastern Seaboard states still reeling from the devastation from Superstorm Sandy, which wiped out power for millions and disrupted usual voting routines.

But that doesnt explain Texas:

In Texas, turnout for the presidential race dropped almost 11 percent from 2008. Vermont and South Carolina saw declines that were almost as large. The drop-off was more than 7 percent in Maryland, where voters approved a ballot measure allowing gay marriage.

Now, going back to the prior analysis that Obama lost 12% of his voters. If this was purely a Democrat dropoff issue, then we would see about 6 percent drop, and a corresponding gain in the Republican polls of 6% reduced margin. In Maryland, it went from Obama / McCain =1.6m 62% / 959K 36.5%:

1,629,467 61.92 959,862 36.47

To Obama/Romney in 2012:

Now there is one more ‘tell’ – Romney won the largest share of the white vote for Republicans … ever. This ‘tell’ tells me that it may well be that (a) minority voters stuck with Obama but white voters abandoned him but did NOT vote for Romney. Back to Sean Trende and his “The case of the missing white voter” where he looks a this extensively:

So instead, I looked at my current home state of Ohio, which has counted almost all of its votes (absentees are counted first here). The following map shows how turnout presently stands relative to 2008. The brightest red counties met or exceeded 2008 turnout. Each gradation of lighter red represents a 1 percent drop in the percentage of votes cast from 2008. Blue counties are at less than 90 percent of the 2008 vote.

We can see that the counties clustered around Columbus in the center of the state turned out in full force, as did the suburban counties near Cincinnati in the southwest. These heavily Republican counties are the growing areas of the state, filled with white-collar workers.

Where things drop off are in the rural portions of Ohio, especially in the southeast. These represent areas still hard-hit by the recession. Unemployment is high there, and the area has seen almost no growth in recent years.

My sense is these voters were unhappy with Obama. But his negative ad campaign relentlessly emphasizing Romney’s wealth and tenure at Bain Capital may have turned them off to the Republican nominee as well. The Romney campaign exacerbated this through the challenger’s failure to articulate a clear, positive agenda to address these voters’ fears, and self-inflicted wounds like the “47 percent” gaffe. Given a choice between two unpalatable options, these voters simply stayed home.

This may tell us two or three things: First, defining Romney as out-of-touch rich capitalist helped Obama; negative campaigning sometimes works by creating stay-at-home voters. This was intentional by Obama. Romney should have known this and been defeating it with GOTV effort.

Second:  Romney, in failing to reach out and actually convert the disappointed Obama voters, ended up losing. And he failed to have the GOTV to get those who were willing to support him to actually got out and vote.

The last ‘tell’ on what happened however has been the turnout models. 2008 was a D +7 election; 2010 was an R +1 Had this been a case of just Democrats staying home, we would have had a repeat of 2010. As we saw before, union voters supported Obama; minority voters DID turn out (so it may well be that minority voters came to support Obama, but might have stayed home had it been another election).

Conservatives argued with polls because we assumed the turnout models would match the 2010 result. But now if we calibrate for the fact that the 13% black, 10% latino and asian vote did not stay home, and a good portion (9 million) of the white vote DID stay home.  Then it begs the question, what percentage of the dropoff was D and what was R?  Then what percentage of the voters would be D or R?

Here is another tell: White women supported Romney 56% to 42%!

2012 voter support breakdown

But the money quote is this:

Across the board, among women, young voters and even Catholics, the pro-Obama advantage was due to higher turnout among, and support from, racial minority groups. These very basic points call into question what the 2012 election was really about: Was it about the economy, or was it really about the “type” of leadership desired by a new coalition of American voters largely consisting of progressive but not necessarily “liberal” thinking and acting racial minorities?

So here is what happened. Obama kept the minority vote because they were attached to him through thick and thin. This was not going to be a matter of whether they voted Romney or not, it would be a matter of whether personal disappointment would keep them home. On the flipside, the personal disappointment for white voters meant they were not voting for Obama. But could Romney gain some support?

In some cases yes on a percentage basis – see the Washington Post exit poll:

Women chose President Obama over Mitt Romney in Tuesday’s election, 55-44 percent. That’s a slightly wider gap than the 56-43 margin Obama had with over Sen. John McCain with women in 2008.

But white women favored the Republican candidate in even bigger numbers this year (56-42 for Romney) than they did in 2008 (53-45 for McCain).

[Note: Romney also won the married women’s vote by 53% to 46% but lost the single women’s vote nearly 2 to 1.]


The WashPost exit poll has a few other gems, like the fact that Asian vote shifted more against Romney vs McCain than any other group.  Romney actually gained 5% vs McCain on the 18-30 YO, but only stayed even on 30-44 group.

And Romney won the independent voters by 5 points, 50-45.

So what went wrong?  The Washington Post exit poll had a D +6 result.

Here’s how the GOTV and turnout would relate – effective GOTV would increase your party’s share of the total number of voters:

D +0 / R+1: If GOP had great turnout and Dems did not.
D +3: If GOP and Dems had equally good turnout.
D +6: If GOP had lesser turnout and Dems did a great job.

CONCLUSION: Democrats had a great GOTV Operation and the GOP did not!

He failed to close the deal. Romney campaign had a lot of enthusiasm at rallies but they failed to do the GOTV and do the real ground game that would have made this an R+1 election and a victory for our side.

Update: Now, after the fact, we are learning more about the problems in the Romney campaign that caused GOTV to be weak. One was the disaster known as Orca:

So, the end result was that 30,000+ of the most active and fired-up volunteers were wandering around confused and frustrated when they could have been doing anything else to help. Like driving people to the polls, phone-banking, walking door-to-door, etc. We lost by fairly small margins in Florida, Virginia, Ohio and Colorado. If this had worked could it have closed the gap? I sure hope not for my sanity’s sake.


Lots of post-mortems on this system that are quite disheartening and explain in part