I published an article a few days looking at the major party U.S. presidential tickets with the largest age gaps between the presidential and vice presidential nominees. While that was a fun survey, I concluded that there is no reason to believe that age differences between the top of the presidential ticket and the running mate have an obvious effect on much of anything. In the end, most people vote for president, not president and vice president. With that in mind, I had a spin off survey idea: Does the older presidential candidate win more often than not? Join me as we find out.
Rules for My Survey
Below, you will find a chart showing the top two finishers and their birthdays for every U.S. presidential election from 1824-2020. I decided to start with 1824 because that is the first election where the majority of Electoral College electors corresponded to a popular vote (I only consider this significant insofar as considering that ordinary voters may weigh age differently than Electoral College electors). Non-AmerMoreover, some of the pre-1824 elections were peculiar for purposes of this survey. For example, George Washington won uncontested in 1788/89 and 1792 and James Monroe did the same in 1820. Due to some peculiarities with the pre-Twelfth Amendment election system, 1800 ended in an Electoral College tie between Thomas Jefferson and his ostensible running-mate, Aaron Burr, forcing the election to the U.S. House of Representatives. For those reasons and also because 1820 was the last uncontested election, I decided that 1824 is a good place to start.
You will find that I put the names of all candidates who ran as the incumbent in bold. I did not make any distinctions in how a candidate became the incumbent. For example, Richard Nixon was the incumbent in 1972 when he ran for re-election after having won the 1968 election. Gerald Ford was the incumbent in 1976 despite his highest elected office having been a U.S. House seat in Michigan (Ford was appointed by Nixon and confirmed by the Senate as Vice President during Nixon’s second term). For our purposes – both are incumbents all the same. As you will see in my analysis section, I took the time to compile statistics distinguishing elections involving incumbents and open elections where the sitting president was not a candidate.
I made an exception to my top-two finisher rule in one case – 1912. In that election, three candidates won multiple states. Woodrow Wilson won the election, former president Theodore Roosevelt finished second, and the incumbent president William Howard Taft finished third. Because I am distinguishing incumbent elections in our survey, I decided to list both Roosevelt and Taft as the losers. The three-way race does not otherwise affect our survey, however, since Wilson was the oldest of the three candidates, thus meaning that it would have been marked as an election where the older candidate won regardless of whether I made an exception to my ordinary one-candidate rule (had Roosevelt won instead of just dragging his former protege into the abyss, it would have been a younger candidate win since Roosevelt was narrowly the youngest of the trio). There are several other elections where more than two candidates won states, namely 1824, 1836, and 1860, wherein four candidates won states, and 1832, 1856, 1892, 1924, 1948, 1960 (odd case), and 1968, wherein three candidates carried states. However, I did not see a reason to break my ordinary two candidate rule in these elections since they each had a clear;y second strongest candidate. There are fair cases for including three candidates in 1824, 1836, and especially 1860, but (A) all of these elections had a clear second-strongest candidate in terms of electoral college strength, and (B) I decided to refrain from making things unnecessarily complicated. In any event, including the third-place electoral college finisher in 1824 and 1836 would not have changed the fact that the oldest (in 1824) and the youngest (in 1836) candidate won. 1860 is slightly trickier. The winner, Abraham Lincoln, was older than the candidate I did count for finishing second in electors, John C. Breckinridge. Lincoln’s Illinois rival, Stephen A. Douglas, had the second most popular votes but he only carried a single state and thus finished fourth in the electoral college, and he was also younger than Lincoln. The third-place finisher in electoral votes, John Bell, was older than Lincoln, but he was competitive in the smallest number of states. I have no issue with excluding Bell and noting that Lincoln was older than the two most significant candidates – Breckinridge who was strongest in the soon-to-be Confederacy and Douglas, who was competitive in some northern states and the main threat to hold Lincoln below the electoral threshold.
Presidential Head-to-Head Age Chart
(Note: Wikipedia has a helpful page with links to all of the major U.S. presidential election candidates, which I used as a resource for taking down dates of birth.)
Without further ado, I present my chart:
| Election | President-Elect | DOB | Runner-Up | DOB | Winner |
| 1824 | John Quincy Adams | 11-07-1767 | Andrew Jackson | 03-15-1767 | Younger |
| 1828 | Andrew Jackson | 03-15-1767 | John Quincy Adams (I) | 11-07-1767 | Older |
| 1832 | Andrew Jackson (I) | 03-15-1767 | Henry Clay | 04-12-1777 | Older |
| 1836 | Martin Van Buren | 12-05-1782 | William Henry Harrison | 02-09-1773 | Younger |
| 1840 | William Henry Harrison | 02-09-1773 | Martin Van Buren (I) | 12-05-1782 | Older |
| 1844 | James K. Polk | 11-02-1795 | Henry Clay | 04-12-1777 | Younger |
| 1848 | Zachary Taylor | 11-24-1784 | Lewis Cass | 10-09-1782 | Younger |
| 1852 | Franklin Pierce | 11-23-1804 | Winfield Scott | 06-13-1786 | Younger |
| 1856 | James Buchanan | 04-23-1791 | John C. Frémont | 01-21-1813 | Older |
| 1860 | Abraham Lincoln | 02-12-1809 | John C. Breckinridge | 01-16-1821 | Older |
| 1864 | Abraham Lincoln (I) | 02-12-1809 | George B. McClellan | 12-03-1826 | Older |
| 1868 | Ulysses S. Grant | 04-27-1822 | Horatio Seymour | 05-31-1810 | Younger |
| 1872 | Ulysses S. Grant (I) | 04-27-1822 | Horace Greeley | 02-03-1811 | Younger |
| 1876 | Rutherford B. Hayes | 10-04-1822 | Samuel J. Tilden | 02-09-1814 | Younger |
| 1880 | James A. Garfield | 11-19-1831 | Winfield Scott Hancock | 02-14-1824 | Younger |
| 1884 | Grover Cleveland | 03-18-1837 | James G. Blaine | 01-31-1830 | Younger |
| 1888 | Benjamin Harrison | 08-20-1833 | Grover Cleveland (I) | 03-18-1837 | Older |
| 1892 | Grover Cleveland | 03-18-1837 | Benjamin Harrison (I) | 08-20-1833 | Younger |
| 1896 | William McKinley | 01-29-1843 | William Jennings Bryan | 03-19-1860 | Older |
| 1900 | William McKinley (I) | 01-29-1843 | William Jennings Bryan | 03-19-1860 | Older |
| 1904 | Theodore Roosevelt (I) | 10-27-1858 | Alton B. Parker | 03-14-1852 | Younger |
| 1908 | William Howard Taft | 09-15-1857 | William Jennings Bryan | 03-19-1860 | Older |
| 1912 | Woodrow Wilson | 12-28-1856 | Theodore Roosevelt William Howard Taft (I) | 10-27-1858 09-15-1857 | Older |
| 1916 | Woodrow Wilson (I) | 12-28-1856 | Charles Evans Hughes | 04-11-1862 | Older |
| 1920 | Warren G. Harding | 11-02-1865 | James M. Cox | 03-31-1870 | Older |
| 1924 | Calvin Coolidge (I) | 07-04-1872 | John W. Davis | 04-13-1873 | Older |
| 1928 | Herbert Hoover | 08-10-1874 | Al Smith | 12-30-1873 | Younger |
| 1932 | Franklin D. Roosevelt | 01-30-1882 | Herbert Hoover (I) | 08-10-1874 | Younger |
| 1936 | Franklin D. Roosevelt (I) | 01-30-1882 | Alf Landon | 09-09-1887 | Older |
| 1940 | Franklin D. Roosevelt (I) | 01-30-1882 | Wendell Willkie | 02-18-1892 | Older |
| 1944 | Franklin D. Roosevelt (I) | 01-30-1882 | Thomas E. Dewey | 03-24-1902 | Older |
| 1948 | Harry S. Truman (I) | 03-08-1884 | Thomas E. Dewey | 03-24-1902 | Older |
| 1952 | Dwight D. Eisenhower | 10-14-1890 | Adlai Stevenson II | 02-05-1900 | Older |
| 1956 | Dwight D. Eisenhower (I) | 10-14-1890 | Adlai Stevenson II | 02-05-1900 | Older |
| 1960 | John F. Kennedy | 05-29-1917 | Richard Nixon | 01-09-1913 | Younger |
| 1964 | Lyndon B. Johnson (I) | 08-27-1908 | Barry Goldwater | 01-02-1909 | Older |
| 1968 | Richard Nixon | 01-09-1913 | Hubert Humphrey | 05-27-1911 | Younger |
| 1972 | Richard Nixon (I) | 01-09-1913 | George McGovern | 07-19-1922 | Older |
| 1976 | Jimmy Carter | 10-01-1924 | Gerald Ford (I) | 07-14-1913 | Younger |
| 1980 | Ronald Reagan | 02-06-1911 | Jimmy Carter (I) | 10-01-1924 | Older |
| 1984 | Ronald Reagan (I) | 02-06-1911 | Walter Mondale | 01-05-1928 | Older |
| 1988 | George H.W. Bush | 06-12-1924 | Michael Dukakis | 11-03-1933 | Older |
| 1992 | Bill Clinton | 08-19-1946 | George H.W. Bush (I) | 06-12-1924 | Younger |
| 1996 | Bill Clinton (I) | 08-19-1946 | Bob Dole | 07-22-1923 | Younger |
| 2000 | George W. Bush | 07-06-1946 | Al Gore | 03-31-1948 | Older |
| 2004 | George W. Bush (I) | 07-06-1946 | John Kerry | 12-11-1943 | Younger |
| 2008 | Barack Obama | 08-04-1961 | John McCain | 08-29-1936 | Younger |
| 2012 | Barack Obama (I) | 08-04-1961 | Mitt Romney | 03-12-1947 | Younger |
| 2016 | Donald Trump | 06-14-1946 | Hillary Clinton | 10-26-1947 | Older |
| 2020 | Joe Biden | 11-20-1942 | Donald Trump (I) | 06-14-1946 | Older |
Analysis
Let us see what we can take away from the chart, which gives us 50 elections to look at.
Older vs Younger Candidates (all elections)
Without considering incumbency or the size of age differences between candidates, the older candidate won 28 out of the 50 elections in our sample. The older candidates pick up steam as the years progress. Younger candidates won 12 of the 21 elections from 1824 through 1904, but older candidates have since won 19 of 29. The key stretch is 1908 through 1956, when older candidates won 11 out of 13 elections, with 7 of those 11 wins featuring incumbents.
Older vs Younger Candidates (incumbent elections)
26 of our 50 elections included a sitting president. The older candidate won 16 of those 26 elections. Unsurprisingly, the incumbent won more often than not – also 16 out of 26. I was particularly curious to see if there is a trend in elections where the incumbent loses. However – I found no such trend. Of the ten challengers to beat incumbents, five were older than the sitting president and five were younger. Interestingly, while older candidates have tended to do better in the latter half of our sample, younger candidates fared better in incumbent elections as time went on. For all incumbent elections (regardless of whether the incumbents won), older candidates won ten out of the first 14 (through 1944) before splitting the next 12. In elections where the incumbent was defeated, the challenger was older on the first three occasions (1828, 1840, and 1888) but younger in 5 of the last 7.
I suspected going in that incumbent elections would favor older candidates because, granting the numerous cases where the incumbent was elevated from the vice presidency instead of having been elected in his own right, someone running for re-election in the ordinary case is four years older than when he ran for a first term – and thus at least somewhat likely to be older than the average new candidate (of course, many caveats apply to this general thought). My suspicion turned out to be right. Because the older candidate won 16 of 26 incumbent elections, we can deduce that older and younger candidates each won 12 of the 24 elections without an incumbent (that tie will be broken in 2024 since this is now a non-incumbent election).
Significant Age Differences
Next, let us try to isolate the elections featuring a significant age difference. The terms older and younger are relative. Our best examples may be the first two elections in our sample – both of which feature John Quincy Adams and Andrew Jackson. Jackson was about eight months older than Adams, so Adams gets credit for winning as the younger candidate in 1824 (we will set aside the myriad questions about what was perhaps the messiest post-Twelfth Amendment election), and Jackson gets credit for winning as the challenger to an incumbent in 1828 as the older candidate. But assuming arguendo voters were highly concerned with age in 1824 and 1828, I sincerely doubt they would have parsed an 8-month age difference between two accomplished men in the latter-half of their 50s any more than one would today.
To the extent we may be able to glean something valuable from candidate age differences, neither 1824 nor 1828 are likely candidates. 1844, which featured James K. Polk defeating Henry Clay, who was more than 18 years his senior, is more along the lines of what we are looking for. Now this is not perfect either. In the upcoming 2024 election, Donald Trump is more than 18 years older than Kamala Harris, but the age difference is driven more by the fact that Mr. Trump is now the oldest presidential nominee ever (surpassing the record set by Joe Biden in 2020) than Ms. Harris being a young nominee. Ms. Harris, at 59, errs on the older side of things for a presidential candidate in her own right. Is an age difference of 78 vs 60 at election time as compelling as 1844, when the difference was 67 vs 49? Who knows?
Of course – we can only quibble so far. I studied by chart and decided to define the term significant age difference as being at least eight years. Eight was a good cut-off in that it gave me a decent sample size to look at and I think it is thematically significant since only one president has served for more than eight years and, with the two-term limit for the presidency established by the twenty-second amendment in 1951, presidents will ordinarily be limited to eight years of service (there are scenarios where it is possible to serve more than eight years – for example Lyndon Johnson would have been able to had he remained in and won the 1968 election because he had initially served less than half of John Kennedy’s term, but such cases are rare and Franklin Roosevelt, who won four consecutive elections prior to the introduction of term limits, remains the only person to have been president for more than eight years).
The trends for the big age difference elections are interesting. Younger candidates won six of the first 11 through 1876, thanks in large part to three consecutive younger victories by Ulysses S. Grant and Rutherford B. Hayes. Things then took a dramatic turn, with older candidates winning eight of nine through 1972 and achieving a maximum advantage of nine (16-7) over younger candidates with George H.W. Bush’s victory over Michael Dukakis in 1988. But since then, we have seen the younger candidate win four straight – with all four being credited to former presidents Bill Clinton and Barack Obama.
We only have four significant age gap elections where the incumbent fell (apparently it is good to be an incumbent president who is much older than your opponent). The younger candidate won in 1976 (Carter over Ford) and 1992 (Clinton over Bush) while the older candidate won in 1840 (Harrison over Van Buren) and 1980 (Reagan over Carter).
Because we have a significant age gap election coming up with no incumbent – albeit a very odd one since one candidate is the incumbent Vice President and the other candidate is a previous president – I decided to cap off the survey by looking at big age gap elections with no incumbent involvement. 8 of the 11 races had occurred by 1896, with the younger candidate winning 5 of those. The latter three were 1952 and 1988 (Eisenhower and G.H.W. Bush winning as the older candidates) and Mr. Obama winning in 2008 as the younger candidate.
What is the Significance Presidential Candidate Age?
The ages of the presidential candidate has more influence on an election than anything involving the age of the vice presidential candidate, and thus the information in this survey is of more practical interest than the information in my previous survey. But I come away unconvinced that the relative ages of the main presidential candidates in a given election is a consistent factor. While older candidates win more often, my survey showed that the advantage largely disappears if we look at races without incumbents. Open races came out just about even, including when we limited our inquiry to races with age differences of at least eight years.
I will venture that the extent to which candidate age plays a role is very election-specific. For example, one can make a fair case that the Clinton/Gore ticket in 1992 made strong electoral use of youth (Clinton was the second youngest person to win a presidential election and Al Gore was the 7th youngest vice president) against the then-68 year old George H.W. Bush and subsequently against the 73-year old Bob Dole in 1996. But on the flip-side, Ronald Reagan not only won a combined 1,014 out of 1,056 available Electoral College votes in his two election wins in 1980 and 1984, he remains the third oldest person to win a first term (behind Biden 2020 and Trump 2016) and is also the second oldest to win any election (behind Biden 2020), and he easily hit our significant age difference threshold in both races. Reagan did not win because of his absolute age or because he was roughly 13.5 and 17 years older than his opponents (notwithstanding his famous quip in a 1984 debate), but it clearly did not weigh him down in the final tallies.
We have seen a recent notable trend in older candidates (see full list of presidents by age). William Henry Harrison, who won the 1840 election, held the record as the oldest president upon first being elected (68 years and 54 days) until the record was broken by Ronald Reagan (69 years and 349 days) 140 years later in 1980. Since 1992, we have had six elections featuring at least one candidate older than Harrison, with four of those elections featuring a non-incumbent (G.H.W. Bush running for re-election in 1992, Bob Dole in 1996, John McCain in 2008, Donald Trump and Hillary Clinton in 2016, Donald Trump running for-relection in 2020 and Joe Biden running for his first term, and Donald Trump in 2024). Moreover, we have had several candidates who fall short of the Harrison threshold but are nevertheless on the older end. For example, had Mitt Romney won in 2012, he would have been, at the time, the fourth-oldest first term president (at the time excludes Messrs. Trump and Biden). For the current election, Mr. Trump would be the oldest first-term president (depending on how you want to account for non-consecutive terms) and Ms. Harris would be the 11th oldest (note: I am referring to the time of election, meaning for Ms. Harris’ purposes that I am excluding Gerald Ford, who was older than she is now when he assumed the presidency after Richard Nixon’s resignation).
In the end – my survey will probably not help anyone make money gambling on the 2024 election or any future contests, but I think it is nevertheless a fun project.
