Now that Texas Senator Ted Cruz has chosen former Hewlitt-Packard CEO Carly Fiorina, a former rival, as his running mate, it's worth looking at history to see whether such "unity tickets" (linking ex-opponents) have been successful in the past.
To determine this, my college students and I looked at this very question earlier in the semester. Each chose a different presidential candidate to see if he chose a VP nominee who had run against him in primary season. Then we looked at whether such "unity tickets" did better than pairings between non-rivals.
Democratic Party cases
Barack Obama and Joe Biden, rivals for the 2008 Democratic nomination, were able to prevail in November. Of course, this unity ticket was reelected in 2012. But it was a different story when rivals John Kerry and John Edwards teamed up in 2004, as that unity ticket fell short. It was the first unity ticket since 1960, when JFK and LBJ went all the way to the White House. Additionally, Adlai Stevenson and Estes Kefauver, former rivals, lost badly to Eisenhower in 1956. That's a 60 percent success rate.
Among Democrats, they've had 12 races involving a President and VP that did not face off against each other in the primaries. Of these tickets, only five were successful (Clinton-Gore in 1992 and 1996, Carter-Mondale in 1976, the Johnson-Humphrey ticket of 1964, and Truman-Barkley in 1948). The other seven (Gore-Lieberman in 2000, Dukakis-Bentsen in 1988, Mondale-Ferraro in 1984, Carter-Mondale in 1980, McGovern-Shriver in 1972, Humphrey-Muskie in 1968, and Stevenson-Sparkman in 1952) didn't win.
Republican Party cases
Most remember Ronald Reagan picking rival George H. W. Bush as his running mate in 1980, winning that contest and reelection four years later. But in only one other occasion (1948) did Republicans try a unity ticket, where Thomas Dewey and Earl Warren were unsuccessful candidates.
On the other hand, Republican Presidential nominees have picked a non-rival 14 times. In these cases, the nominee has won seven times (Bush-Cheney in 2000 and 2004, Bush Sr. and Quayle in 1988, Nixon-Agnew in 1968 and 1972, and Eisenhower-Nixon in 1968 and 1972 as well). The eight failed cases of non-rivals pairing up include Romney-Ryan (2012), McCain-Palin (2008), Dole-Kemp (1996), Bush-Quayle (1992), Ford-Dole (1976), Goldwater-Miller (1964), and Nixon-Lodge (1960).
Overall success rate
If you're keeping score, that means that overall, unity tickets have prevailed in five of eight cases (62.5%), while President-VP tickets with non-rivals have worked a little less than half of the time (12 of 26 or 46.2% successful). That would tell you that Senator Cruz's pick was a good idea. But when you look at the chi-square analysis, unity tickets are not significantly more likely to prevail in the fall election than their non-rival counterparts. Unity tickets are only slightly more likely to do better than expected, given all 34 cases of post-World War II elections.
The other factor to consider is that many of these tickets were formed after a candidate cinched the nomination. Picking a rival or non-rival before the primary season was over is pretty rare, except for the case of incumbents. Had Cruz made this pick after Iowa or Wisconsin, when his chances looked better, it might have made a difference. But now, as he has fallen far behind Trump, it looks more like a desperation move.
John A. Tures is a professor of political science at LaGrange College in LaGrange, Ga. He can be reached at email@example.com. His students Michael J. Buckley, Alexandra H. Butson, Keaton W. Coates, Anisa S. Cole, Brandon S. Collins, James R. (Reid) Emery, Jacob D. Gassert, Seth T. Golden, Helon H. Hammonds, Richard C. Howell, Alexander O. Hughes, Chandler E. Joyner, Mary E. Loftus, Miguel Martinez, Erin Missroon, Duncan M. Parker, Benjamin J. Puckett, Nicholas J. Rawls, Christopher A. Smith, Tressea K. Stovall, Brooke N. Turner, Stephen P. Wagner and Lindsey G. Weathers contributed to the research for this column.