Mini Mike vs. Big Don

Does Height Matter in Presidential Elections?

Do you have to be tall to be president? For most political observers the answer is yes. A study by a Texas Tech Political Scientist, Gregg Murray, concluded that the taller candidate from 1789 to 2012 won 58% of presidential elections and received the majority of the popular vote in 67% of those elections.

The reasoning among some researchers is that taller candidates “look stronger” and are seen as having more leadership and communication skills. But do such perceptions still matter in the modern world since the age of television? Most studies that have looked at height and presidential races starting in 1900, where 20 of the 29 candidates were taller than their opponent (and two were the same height). Statistically, that’s 68% of all elections since 1900.

In a recent article, using these statistics the Independent had this to say about why tall men win presidential elections: “History suggests that beating Donald Trump in 2020 could be a tall order for Democrats – because voters tend to vote for presidential candidates with a height advantage over their opponent. While it may sound superficial, taller presidential candidates have fared better over the years, with the taller of the two candidates winning the popular vote in two-thirds of elections, and in the electoral college more than half of the time.”

All of these height statistics are mostly based on studies from 1900. Now you may ask yourself, how did voters prior to the age of television know which candidate was taller? The most common answer is stereographs, and of course, traditional black and white photographs. I find this explanation somewhat dubious, since the photograph would have to include both candidates side by side. Possible, but hardly the impact of a TV debate.

This height issue has now resurfaced since the entry and early exit of Michael Bloomberg into the race. Both Trump and Bloomberg have a history together in New York and its not a positive one. And true to Trump’s penchant to characterizing his opponents, he has renamed Bloomberg as “Mini Mike.”

Trump claimed that “Mini Mike” is only 5′ 4″, but his actual height is listed at 5′ 8,” still some seven inches short of Trump’s 6′ 3″ claim. In addition, if Bloomberg does become the Democratic nominee, he will have tied Michael Dukakis, as the shortest Presidential candidate since the television era.

So if “Mini Mike” had decided to stay in the race and became the Democratic nominee was he destined to lose the general election. So I finally decided to put this idea that bigger men usually win Presidential races to an empirical test. But this time, the starting date is at the beginning of the television age: 1952.

Since then there have been 16 presidential elections that had at least one televised debate and news coverage galore. Although there are no surveys that asked about height differences, the probability of voters noticing the taller candidate is certainly better than a black white stereograph.

Out of the 16 races (I heave excluded Hilary Clinton’s loss due to gender differences), five winners were the shorter candidate, giving up 4.6 centimeters (1.8 inches) to the taller losers. The average height for all winning candidates was 184 centimeters (72.4 inches); the average height for losers was 182 centimeters (71.6 inches), a difference of less than inch.

It’s worth noting that the average height for an American male during this period was 175 centimeters (69.1 inches). In other words, men who ran for president were on average taller (2.5 inches) than the general male population.

The question, of course, is there a significant statistical difference between the heights of winners and losers. Or, in the alternative, there is no statistical difference and taller men don’t have an electoral advantage when they run for president.

Using Analysis of the Variance (ANOVA) we can test the null hypothesis that there is no statistical difference between taller and shorter candidates when it comes to winning the White House.

ANOVA is a statistical technique that measures whether differences between two or more groups (tall vs. short candidates) and success in wining the election. Significance levels above .05 level, shows there is no difference and consequently, that heights do not contribute to election success.

When we measure the differences between tall and shorter candidates and victory at the ballot box from 1952, the significance level is far greater than the .05 level (.833), as shown under column “Significant” in Table 1 below. In other words, height had no impact on whether a candidate, tall or short, won or lost.

  Sum of Squares df Mean Square F Significant
Between Groups 2.286 1 2.286 .046 .833
Within Groups 691.714 14 49.408    
Total 694.000 15      


On the surface, taller presidential candidates seem to win more elections (11 taller and 5 shorter), but this difference is random or the result of campaign effects and not based on the height differences.

So why do previous studies support the “height wins theory”? There are a couple of possibilities for these differences, foremost is the time difference. Previous studies used arbitrary starting dates for their data, predominantly the year 1900. Others have included even elections back to George Washington.

During the early Republic, partisanship was in its infancy and a candidate’s party label did not have the influence on election choice as it does today. For a lifelong Republican today, a tall Democrat against diminutive Republican would still not get his vote, just ask John Kerry.

Information about the candidates position on issues was often scarce. No CNN all day long, and voters had fewer heuristics to use to determine their choice. Consequently, voters might use height as an alternative for information. The stereotype that taller people are more successful and even, more intelligent, could make a difference absent any other information. Also possible, that the winning and losing differences between tall and short candidates are random.

So what about Mini Mike if had stood on that debate stage some seven inches lower than the Big Don? I have shown that being tall isn’t a requirement to become president in the modern age. That said, having Donald Trump labeling you continuously as “Mini Mike” could have a negative effect over time. Ask “Little Marco” how he feels about it. And he is 5’10” tall.

By Jim Kane

Jim Kane is a pollster and media advisor, and was for fifteen years an Adjunct Professor of Political Science at the University of Florida. Kane is founder of the polling firm USAPoll and served as the Director of the Florida Voter Poll. His political clients have included both Republican and Democratic candidates, including the Republican Party of Florida, and both the Sun-Sentinel and Orlando Sentinel newspapers. At the University of Florida, Professor Kane taught graduate level courses in political science on Survey Research, Lobbying and Special Interest Groups in America, Political Campaigning, and Political Behavior. In addition to his professional and academic career, Jim Kane has been actively involved in local and state policy decisions. He was elected to the Broward County Soil and Water Conservation Board (1978-1982) and the Port Everglades Authority (1988-1994). Kane also served as an appointed member of the Broward County Planning Council (1995-2003), Broward County Management Review Committee (Chair, 1990-1991), Broward County Consumer Protection Board (1976-1982), and the Broward County School Board Consultants Review Committee (1986-1990).


  1. Hello Dr. Kane, I really enjoyed reading this blog mostly because I am French and our winners at the presidential elections of late have been on the tall side of the equation. However, we also have had confounding factors such as gender that you removed from your analysis for the US.
    Regarding your analysis, I am quite glad to see that the height does not appear to be a significant factor! However, do you think that the results are the same if the heights (and why not weight-height ratio) of the candidates are compared using paired datasets (winner size vs looser size for each election as paired data)?

    Thank you so much for your time! Stay safe from COVID-19!

    1. I’m glad you enjoyed the post. As for gender confounding effects, as you are aware the number of women running for president in the US is sparse. In addition, with the exception of Hillary Clinton, women who have run were from minor parties that never would have been competitive even it they were seven feet tall. I excluded Hillary from the analysis, because I felt that voters would not use height as a hueristic when comparing a male vs. a female candidate.
      I never thought of weight-ratio since I was addressing the common impression that “taller candidates always win.” Your suggestion of using paired data-sets is a good one. As time permits, I will try it and let you know. Thanks

Leave a comment

Your email address will not be published. Required fields are marked *