Florida's 4th district

Choose another district

A district number is not always the same place. States redraw their districts after each census, and some redraw at other times. The results below are the history of the number 4 in Florida, and may cover different ground from one decade to the next. The map shows today's boundaries only.

Latest results

U.S. House · 2024

Florida, District 4

Final result

Won

Aaron Bean

Republican

57.3%

222,364 votes

LH

Lashonda “l. J.” Holloway

Democratic

42.7%

165,912 votes

U.S. House · 2022

Florida, District 4

Final result

Won

Aaron Bean

Republican

60.5%

165,696 votes

LH

Lashonda "L.j." Holloway

Democratic

39.5%

108,402 votes

U.S. House · 2020

Florida, District 4

Final result

Won

John H. Rutherford

Republican

61.1%

308,497 votes

Donna Deegan

Democratic

38.9%

196,423 votes

U.S. House · 2018

Florida, District 4

Final result

Won

John H. Rutherford

Republican

65.2%

248,420 votes

GS

George "Ges" Selmont

Democratic

32.4%

123,351 votes

Every result in the file

Other districts in Florida in the latest election: FL-01 · FL-02 · FL-03 · FL-05 · FL-06 · FL-07 · FL-08 · FL-09 · FL-10 · FL-11 · FL-12 · FL-13 · FL-14 · FL-15 · FL-16 · FL-17 · FL-18 · FL-19 · FL-20 · FL-21 · FL-22 · FL-23 · FL-24 · FL-25 · FL-26 · FL-27 · FL-28. All of Florida · The 2026 race for this seat

Download these results as a spreadsheet (CSV)

How this archive is built

• The file holds each candidate's vote total for each race. It has no winner column. The winner shown here is the candidate with the most votes in the race; if two are level at the top, the page says so and names no winner.
• Margin is the winner's votes minus the runner-up's. Points are that margin as a share of all votes counted in the race, including blank and scattered votes where the file has them.
• Two-party share is the Democratic candidate's votes divided by the Democratic and Republican votes together. It is only worked out when both parties had a candidate.
• Where the same person is listed on more than one ballot line (for example a party line and a joint line), the lines are added together.
• A race with one candidate on the ballot, write-ins aside, is marked unopposed. The file records some of these with a total of 1 instead of a vote count; the page shows "not published" for them, not 1.
• When a seat had more than one round in an election, the last round decides the seat: a runoff, then the general election, then a round the file labels "pre". The file's codebook does not define that label, so the page shows it as the file gives it.
• Special elections are listed beside the regular races for their year and are not counted in the seat totals.
• This page is about general elections and the runoffs that follow them. Primaries and primary runoffs are on the primary results page.
• Party names are the party lines the file gives. Candidates who ran on lines other than Democratic or Republican are shown as "Other party or independent" in seat totals. Independents who caucus with a party are not grouped with it.
• Names are written in capital letters in the file. The page changes only the case.
• For the years from the Federal Election Commission, each candidate row is the FEC's own, with general-election votes. The check figure is the total the FEC prints at the end of each race ("District Votes"). Where it prints none, the page says so.
• The FEC lists a candidate who ran on several party lines (New York, Connecticut, South Carolina and others) once for each line, and in some years also on a combined line. The page adds the lines together and drops a combined line so no vote is counted twice. The party shown is the line with the most votes.
• Write-in and scattered votes are shown as one row, "Blank, over and scattered votes", not as a candidate. A candidate the FEC lists as "Unopposed" with no vote count is shown as unopposed with the count "not published".
• The races counted for each year include the seats of the District of Columbia, American Samoa, Guam, the Northern Mariana Islands and the Virgin Islands, and Puerto Rico's resident commissioner, which the FEC lists beside the 435 state districts. Those members cannot vote on most questions on the House floor. The seats won by party count them too.
• In Louisiana (and Georgia, for the Senate) the first round and the runoff are separate rounds of one election. The runoff decides the seat.
• Where the FEC marks as winner a candidate who did not have the most first-choice votes (Maine's ranked-choice voting), that candidate is listed first, the votes shown are first-choice votes and no margin is worked out.
• Some seats were filled twice on one ballot: once for the new term and once for the rest of a term left vacant. The second is listed as a special election beside the regular race. The FEC calls them "full term" and "unexpired term".
• For the years from the Clerk of the House, each candidate row is a line the Clerk prints, with party names as printed. The check figure is the total the Clerk prints for the race: in its state recapitulation table, or on a "Total" line. A state's table is sometimes printed sideways over two pages, and for those races the page confirms only that the lines add up to a total printed in the state's table. Where the Clerk prints a total that does not equal its own lines, both are kept and the race is flagged.
• The Clerk prints "Under Votes" and "Over Votes" lines from 2020 on, and "Blank" or "Scattering" lines in some earlier years. They are shown in the one row "Blank, over and scattered votes". A race where the Clerk prints a footnote mark instead of a vote count (a candidate with no opponent, in the states that do not count such votes) is shown as unopposed with the count "not published".
• For 1976 to 1994, each row is from the MIT lab's file, which the lab compiled from the Clerk of the House's Statistics of the Congressional Election. The Clerk is the original source; this site has not checked these years against the Clerk's printed pages. Party names are as the file gives them, and a candidate on several party lines (for example in New York) appears once, with the lines added together.
• A district number is not the same place in every decade. States redraw their districts after each census, and some redraw at other times. District 7 in 1998 and District 7 in 2024 may cover different ground, so a district page is the history of a number, not always of one place.

Source for 2004 to 2022: Federal Election Commission, Federal Elections ↗, the Excel workbook the FEC publishes for each election (U.S. federal government work; no license text on the FEC pages). Fetched and checked November 5, 2026. The workbooks were read into one flat file; its SHA-256 is b331053c623e430a3acc12ff5daf50b81745b7d003d48254d3d47dfe94ce440e.

Source for 1920, 1922, 1924, 1926, 1928, 1930, 1932, 1934, 1936, 1938, 1940, 1942, 1944, 1946, 1948, 1950, 1952, 1954, 1956, 1958, 1960, 1962, 1964, 1966, 1968, 1970, 1972, 1974, 1996, 1998, 2000, 2002, 2024: Clerk of the House of Representatives, Statistics of the Presidential and Congressional Election ↗, the PDF the Clerk publishes for each election (U.S. federal government work; no license text on the Clerk pages). Fetched and checked November 5, 2026. The PDFs were read into one flat file; its SHA-256 is 890fe7e3ada3a04ec96bde2fd9a2dd0f9e76371eeb5f328c1b4adbf90ca37563. Where the Clerk prints a total that does not equal its own candidate lines, both are kept as printed and the race is flagged.

Source for 1976 to 1994: MIT Election Data and Science Lab, U.S. House of Representatives 1976-2024 (Harvard Dataverse, doi:10.7910/DVN/IG0UN2), compiled from the Clerk of the House's Statistics of the Congressional Election ↗, CC0 1.0, version 15.0, released March 9, 2026. Downloaded and checked November 2, 2026. File SHA-256 76eec42483ee9f0fc5cf6e377442c783d1ae6eefa929729fd6561a69bab47f2a. The lab is the compiler; the original source is the Clerk of the U.S. House of Representatives, Statistics of the Congressional Election ↗.

Check run on all 18,152 races with a total to check against: the candidate rows add up to the source's own total in 18,082. The 70 that do not are flagged where they appear.