States where the winner's best raw vote margin county failed to account for at least 1/4 of the statewide margin
1800
1896
1900
1904
1908
1912 * §
1916
1920 *
1924
1928
1932
1936
1940
1944
1948
1952
1956
1960
1964
1968 *
1972
1976
1980
1984
1988
1992
1996
2000
2004
2008
2012
2016 *
2020
3000
* In 2016, Trump's margin over his nearest competitor in any county in Utah was not over 1/4 of his statewide Utah margin. However, his margin over Hillary Clinton in Utah County (where Evan McMullin came in second) was over 1/4 of his statewide Utah margin; and Hillary Clinton came in second in Utah statewide. The same goes for Mecklenburg, NC and Richland, SC in 1968; Spartanburg, SC in 1956; King, WA in 1920; and St Clair, IL, Silver Bow, MT, and Milwaukee, WI in 1912.
§ For 1912, I used red for both TR and Taft. (Normally, I didn't consider states carried by third-party candidates.) In 1912, Taft was not on the ballot in South Dakota (or California), and Roosevelt was listed as the 'Republican' nominee in South Dakota.
---
It so happens that 1/4 is the largest unit fraction that, in this regard, distinguishes Connecticut (and states with fewer counties than Connecticut) from New Hampshire (and states with more counties than New Hampshire). As will be recalled, the authors of 'The Butterfly Did It' 'pool[ed] the data for Connecticut, Delaware, Hawaii, and Rhode Island',
because 'they have too few counties to analyze separately' (p. 797). Dating back to at least 1884, no nominee has carried Connecticut (or any state with fewer counties than it) without at least one county in the state accounting for at least 1/4 of his or her statewide margin, whereas every state with more counties than Connecticut has been so carried at least once from 1896 on (New Hampshire being the state with the next-fewest counties after Connecticut). Delaware and Hawaii have three and four counties, respectively (the authors of 'Butterfly' don't count Kalawao County), so it is actually mathematically impossible for anyone to carry either of them without at least one county in the state accounting for 1/4 or more of his or her statewide margin. However, it is possible for someone to so carry Rhode Island (five counties) and Connecticut (eight counties).
New Hampshire has not only been carried without any one county in the state accounting for 1/4 of the statewide winner's statewide margin, but has been so carried a number of times--11, from 1896 through 2020, most recently in 2008. In the five elections from 1892 through 1908, Wyoming had the second-fewest counties after New Hampshire (with 13); in three of those five elections, no county accounted for 1/4 of the statewide margin of the winner in Wyoming. For most of the time, however, Massachusetts and Vermont (and, from 1912 through 1980, Arizona) have been in a three-way tie for the fewest counties after New Hampshire (with 14 each); and until the last few decades, Vermont has been carried more often than not in such a way that no county accounted for 1/4 of its carrier's statewide margin. From (at least) 1880 (the first election for which county information about Vermont is available) through 1984, Vermont was almost continuously carried in such a way, the only exceptions in that span being 1912, 1940, and 1944. Furthermore, Nevada had 14 counties in the elections from 1876 through 1908 (and hence was part of the tie for next-fewest counties after New Hampshire from 1876 through 1888, and after Wyoming from 1892 through 1908); during this time, no county in Nevada accounted for 1/4 of its carrier's statewide margin in 1884, 1888, 1892, 1896, and 1900 (with information about 1876 unavailable).
These are the states with fewer than 40 counties, along with the elections from 1896 through 2020 in which no county accounted for 1/4 of the statewide winner's statewide margin. (The authors of 'Butterfly' further distinguished between states with fewer than 40 counties, and states with more, writing, 'a state would need to have about 40 to 60 counties to produce reliable estimates' [given the number of variables they use] [p. 797].) I note the number of counties the state has now, although the number of counties in some states (especially Western ones) has been variable through the first quarter or so of the 20th century. (At any rate, none of these states ever had fewer counties than New Hampshire.)
Delaware: 3 (--)
Hawaii: 4 (--)
Rhode Island: 5 (--)
Connecticut: 8 (--)
New Hampshire: 10 (1896, 1900, 1904, 1908, 1920, 1924, 1952, 1956, 1976, 1996, 2008) (11)
Vermont: 14 (1896, 1900, 1904, 1908, 1916, 1920, 1924, 1928, 1932, 1936, 1948, 1952, 1956, 1960, 1964, 1968, 1972, 1976, 1980, 1984) (20)
Massachusetts: 14 (1896, 1900, 1904, 1908, 1920, 1924, 1964) (7)
Arizona: 15 (1912, 1916, 1948) (3)
Maine: 16 (1896, 1900, 1904, 1908, 1920, 1924, 1928, 1936, 1952, 1956, 1960, 1964, 1972, 1984, 1996) (15)
Nevada: 17 (1896, 1900, 1912, 1916) (4)
New Jersey: 21 (1920, 1924, 1928, 1956, 1964, 1972, 1980, 1984, 1988, 1996, 2000) (11)
Wyoming: 23 (1900, 1904, 1908, 1916, 1920, 1928, 1932, 1936, 1952, 1956, 1960, 1968,
1972, 1976, 1980, 1984, 1988, 1992, 1996, 2000, 2004, 2008, 2012, 2016,
2020) (25)
Maryland: 24 (1928) (1)
Utah: 29 (1928) (1)
New Mexico: 33 (1920, 1928, 1932, 1936, 1940, 1948, 1964) (7)
Oregon: 36 (1912, 1952, 1956, 1960, 1968, 1972, 1980, 1984) (8)
Washington: 39 (1896, 1900, 1908, 1948, 1980, 1984) (6)
In addition, Montana had between 16 and 31 counties (inclusive) in its first six elections (1892-1912); Idaho had between 18 and 37 in its first six elections (1892-1916); and North Dakota had 39 in its first four elections (1892-1904). These are the instances, in those elections, in which no county accounted for 1/4 of the statewide margin of the statewide winner:
North Dakota (1892-1904): 1896, 1900, 1904
Montana (1892-1912): 1904
Idaho (1892-1916): 1892, 1896, 1904, 1908, 1916
It is true that a number of these states have been carried in such a way that no one county accounted for 1/4 of the statewide winner's statewide margin very seldom--most notably Utah and Maryland, and then Arizona and, at least within the last 32 elections, Nevada. That said, there are states with many more than 40 counties that have been carried conspicuously few times in such a way that no county accounted for 1/4 of the statewide winner's statewide margin--comparably few to the states with between 10 and 40 counties that have been so carried amongst the fewest times (within the last 32 elections).
Here are the other 32 states (excluding Alaska) ordered by how many times since 1892 (or, in some cases, since county-level information about them became available, if this is later) that they have been carried in such a way that no county accounted for 1/4 of the statewide winner's statewide margin, from least to most. The number of counties is the current number of counties, and I am including all the instances from 1896 on for Idaho, Montana, and North Dakota, despite the above caveat. There will be a comment on the italicised states after the list.
California: 58 (1904, 1908, 1980) (3)
Illinois: 102 (1900, 1928, 1952, 1972) (4)
Pennsylvania: 67 (1928, 1952, 1956, 1972, 1980, 1984) (6)
Colorado: 64 (1904, 1912, 1932, 1944, 1952, 1956, 1960, 1972, 1980, 1984) (10)
Louisiana: 64 (1948, 1952, 1960, 1968, 1972, 1984, 2004, 2008, 2012, 2016, 2020) (11)
Michigan: 83 (1896, 1900, 1904, 1908, 1912, 1916, 1928, 1952, 1956, 1972, 1984) (11)
Minnesota: 87 (1896, 1900, 1904, 1908, 1920, 1936, 1948, 1952, 1964, 1972, 1976) (11)
Missouri: 114 (1896, 1900, 1904, 1912, 1928, 1932, 1972, 2000, 2004, 2012, 2016, 2020) (12)
Wisconsin: 72 (1896, 1900, 1904, 1908, 1916, 1920, 1924, 1928, 1952, 1956, 1960, 1972, 1984) (13)
New York: 62 (1896, 1900, 1904, 1908, 1916, 1920, 1924, 1952, 1956, 1964, 1972, 1996, 2000, 2008, 2012) (15)
Ohio: 88 (1900, 1904, 1912, 1916, 1920, 1924, 1928, 1952, 1956, 1960, 1972, 1980, 1984, 1988, 2016, 2020) (16)
Tennessee: 95 (1896, 1900, 1904, 1912, 1916, 1924, 1928, 1932, 1948, 1972, 1976, 1984, 1988, 2004, 2008, 2012, 2016, 2020) (18)
Montana: 56 (1904, 1916, 1920, 1928, 1932, 1936, 1940, 1948, 1952, 1956, 1964, 1972, 1980, 1984, 2000, 2004, 2012, 2016, 2020) (19)
Oklahoma: 77 (1908, 1912, 1916, 1920, 1924, 1928, 1932, 1936, 1940, 1944, 1948, 1960, 1964, 1972, 1984, 2000, 2004, 2008, 2012, 2016, 2020) (21)
Florida: 67 (1896, 1900, 1904, 1908, 1912, 1916, 1920, 1924,
1928, 1932, 1936, 1940, 1944, 1948, 1968, 1972, 1980, 1984, 1988, 2004,
2020) (21)
Kentucky: 120 (1912, 1916, 1928, 1932, 1936, 1940, 1944, 1948,
1960, 1964, 1968, 1972, 1976, 1984, 1988, 2000, 2004, 2008, 2012, 2016,
2020) (21)
South Carolina: 46 (1896, 1900, 1904, 1908, 1916, 1920, 1924, 1928, 1932, 1936, 1940, 1944, 1948, 1964, 1972, 1976, 1984, 1988, 2000, 2004, 2016, 2020) (22)
Idaho: 44 (1896, 1904, 1908, 1916, 1920, 1924, 1928, 1932, 1936, 1940, 1952, 1956, 1968, 1972, 1980, 1984, 1988, 1996, 2000, 2004, 2008, 2012, 2016, 2020) (24)
Virginia: 133 (1896, 1900, 1904, 1908, 1912, 1916, 1920, 1924, 1928, 1932, 1936, 1940, 1944, 1948, 1952, 1956, 1960, 1964, 1968, 1972, 1984, 1988, 2000, 2004) (24)
Indiana:
92 (1900, 1904, 1920, 1924, 1928, 1932, 1936, 1944, 1952, 1956, 1960, 1964, 1968,
1972, 1976, 1980, 1984, 1988, 1992, 1996, 2000, 2004, 2012, 2016, 2020) (25)
South Dakota: 66 (1900, 1904, 1908, 1912, 1916, 1920, 1924, 1928, 1932, 1936, 1940, 1944, 1948, 1952, 1956, 1960, 1964, 1968, 1972, 1980, 1984, 2000, 2004, 2012, 2016, 2020) (26)
North Carolina: 100 (1896, 1900, 1904, 1908, 1912, 1916, 1920, 1924, 1928, 1932, 1936, 1940, 1944, 1948, 1952, 1960, 1964, 1972, 1976, 1980, 1984, 1988, 1996, 2000, 2004, 2016) (26)
North Dakota: 53 (1896, 1900, 1904, 1908, 1912, 1920, 1928, 1932, 1936, 1940, 1944, 1948, 1952, 1956, 1960, 1964, 1968, 1972, 1980, 1984, 1988, 1992, 2000, 2004, 2012, 2016, 2020) (27)
Iowa: 99 (1896, 1900, 1904, 1908, 1912, 1916, 1920, 1924, 1928, 1932, 1936, 1940, 1944, 1952, 1956, 1960, 1964, 1968, 1972, 1980, 1984, 1988, 1992, 1996, 2008, 2016, 2020) (27)
Kansas: 105 (1896, 1900, 1904, 1908, 1912, 1916, 1920, 1924, 1928, 1932, 1940, 1944, 1948, 1952, 1956, 1960, 1968, 1972, 1980, 1984, 1996, 2000, 2004, 2008, 2012, 2016, 2020) (27)
Georgia: 159 (1896, 1900, 1904, 1908, 1912, 1916, 1920, 1924, 1928, 1932, 1936, 1940, 1944, 1948, 1952, 1956, 1960, 1964, 1968, 1972, 1976, 1980, 1984, 1988, 2000, 2004, 2008, 2012) (28)
Arkansas: 75 (1896, 1900, 1904, 1908, 1912, 1916, 1920, 1924, 1928, 1932, 1936, 1940, 1944, 1948, 1952, 1956, 1960, 1964, 1968, 1972, 1976, 1984, 1988, 1992, 1996, 2008, 2012, 2016, 2020) (29)
Nebraska: 93 (1896, 1900, 1904, 1912, 1916, 1920, 1924, 1928, 1932, 1940, 1944, 1948, 1952, 1956, 1960, 1968, 1972, 1976, 1980, 1984, 1988, 1992, 1996, 2000, 2004, 2008, 2012, 2016, 2020) (29)
Alabama: 67 (1896, 1900, 1904, 1908, 1912, 1916, 1920, 1924,
1932, 1936, 1940, 1944, 1948, 1952, 1956, 1960, 1964, 1968, 1972, 1976,
1984, 1988, 1992, 1996, 2000, 2004, 2008, 2012, 2016, 2020) (30)
Information about these three states is missing for a substantial number of elections (although, even with the missing elections, we know that each of them were carried such that no county accounted for 1/4 of the statewide margin in at least 2/3 of the last 32 elections):
West Virginia: 55 (1920, 1924, 1928, 1932, 1936, 1940, 1944, 1948, 1956, 1964, 1968, 1972, 1976, 1980, 1984, 1988, 1992, 1996, 2000, 2004, 2008, 2012, 2016, 2020) (24) (information missing for 1896-1912)
Mississippi: 82 (1896, 1900, 1916, 1920, 1924, 1928, 1932, 1936, 1940, 1944, 1948, 1952, 1956, 1964, 1968, 1972, 1984, 1988, 1992, 2000, 2004, 2008, 2012, 2016, 2020) (25) (information missing for 1904-1912)
Texas: 254 (1916, 1920, 1924, 1932, 1936,
1940, 1944, 1948, 1960, 1964, 1972, 1976, 1980, 1984, 1988, 1992, 1996,
2000, 2004, 2008, 2012, 2016, 2020) (23) (information missing for 1896-1912)
Two of the italicised states, California and Illinois, are, along with Pennsylvania, severe outliers amongst states with more than 40 counties. If we included states with 10-40 counties, they have more company but are still notably toward the lower end of instances of being carried in such a way that no county accounted for 1/4 of the winner's statewide margin:
Maryland: 1
Utah: 1
Arizona: 3
California: 3
Illinois: 4
Nevada: 4
Washington: 6
Pennsylvania: 6
Massachusetts: 7
New Mexico: 7
Oregon: 8
Colorado: 10
New Hampshire: 11
New Jersey: 11
Louisiana: 11
Michigan: 11
Minnesota: 11
Missouri: 12
Wisconsin: 13
Maine: 15
New York: 15
...
The four italicised states, together with Massachusetts, are the states presented in 1978 as examples of 'big county states' by Peter W. Colby and Paul Michael Green. They didn't define the concept rigorously, but, except for Massachusetts, they presented a table of each state's six most populous counties, showing the share of the state's total population accounted for by each of the counties and by all six of the counties together. The smallest share of the state's total population accounted for by the largest county, in any of these cases, was 14.30% (by Brooklyn, of New York's); the smallest share accounted for by the smallest of any of the states' 'big six' was 2.26% (by Madison, of Illinois'); and the smallest share accounted for by all six put together was 55.57% (of New York's). (Even though they don't go into as much detail in Massachusetts, they make passing reference to Massachusetts' 'big six' as well, as well as to the rough proportion of the state's total population accounted for by that state's six largest counties.)
New York hasn't been carried particularly seldom in such a way that no county has accounted for 1/4 of its carrier's statewide margin--22 other states, out of 47 (leaving aside Delaware, Hawaii, and Alaska), have been so carried as few or fewer times in the last 32 elections. However, it does seem likely that some states are more prone to not be carried in such a way that no county accounts for 1/4 of their carrier's statewide margin, than others are--states where population is particularly concentrated. Now, every state Colby and Green were discussing had, as of 1978, 14 or more electoral votes, so by 'big county', perhaps they meant, 'big' relative to the national population or relative to some absolute benchmark. But I think, if 'big' is construed as 'big' relative to the population of the state a given county is in, some mid-sized and even small states could be called 'big county states' (e.g., Arizona, Nevada, and Utah, where Maricopa, Clark, and Salt Lake Counties were not particularly large relative to the nation until relatively recently).
Conversely, the states that have been carried relatively few times in such a way that no county accounted for at least 1/4 of their carrier's statewide margin could be considered largely 'rural' states. Of course, other factors enter into how many times this happens; many of the states where this has happened the fewest times are in the South and were carried by Democrats by enormous margins for much of the 20th century. (This, however, did not prevent Louisiana being carried relatively seldom in such a way that no parish accounted for 1/4 of its carrier's statewide margin.)
This is important to keep in mind in what follows. Arguably, everything else being equal, the extent of how many states a nominee has carried in such a way that no one county in them accounts for 1/4 of his margin in those states, is an indicator of the breadth of his support across the nation. And, all things being equal, a nominee's carrying a state in such a way that no one county accounts for 1/4 of his margin in it, is an indicator of his having won broad support in that individual state. Of course, all things are not equal; a nominee might have broad support in a state where a county accounts for 1/4 of his statewide margin because his margin of victory was very narrow (in the last 33 elections, the narrowest state margin [and the only one under 2%], 1/4 whereof was not accounted for by any one county, was Weaver's 1.81% margin of victory in Kansas in 1892). Or a nominee might have broad support in a state where a county accounts for 1/4 of his margin because its population is simply very concentrated (e.g., I don't think anyone would gainsay that Reagan, George W. Bush, or Mitt Romney lacked breadth of support in Utah, where Salt Lake and Utah Counties accounted for 42.1% and 13.9%, 39.0% and 16.1%, and 41.1% and 19.2% of the state's total vote in 1984, 2004, and 2012, respectively).
(For the sake of simplicity, I will be calling instances where a county does account for 1/4 of the winning nominee's statewide margin in a state, a 'big county win', and instances where no county does, a 'small county win'--'big', again, as in, relative to the state, not necessarily in an absolute sense.)
However, there are certain trends that are visible when looking at the aggregate of states that are 'big county' or 'small county wins' for the two parties over time.
For example, in every election from 1896 through 1964, every victorious Democrat won at least six 'small county wins'. (In fact, every Democrat from 1896 through 1964 full stop--almost certainly--won at least four 'small county wins', a minimum set in 1956 and, probably, in 1928, assuming Smith's wins in South Carolina and Mississippi were small county wins.) In all elections, Democrats won a number of 'small county wins' in the South, but when they were winning--as well as in 1896 and 1900--they also won some 'small county wins' outside the South, with the two exceptions of 1944 and 1960. Apart from these two exceptions, some of these 'small county wins' were invariably in the Mountain West (even in 1964, when Lyndon Johnson won a 'small county win' in Montana and New Mexico).
The six small county wins won by Kennedy in 1960 was the fewest for any winning Democrat from 1896 on. (I chose 1896 as a starting point, by the way, because it happens to be an important election that is also around the time when information for most states becomes available--although it also seems reasonable to suggest, on the chance that Connecticut or Rhode Island was ever a small-county win deep in the 19th century, that it is nevertheless significant that they never were in this extensive, more recent period, whereas New Hampshire was a number of times in that same period.) Truman had won 15 in 1948 (which was more than FDR's 12 in 1944, incidentally, despite the defection of four Deep South states in 1948); apart from LBJ's landslide in 1964, no Democrat has ever again won as many small county wins as Truman did in 1948.
In 1968, Hubert Humphrey carried only one small county win (West Virginia). This was nearly unprecedented for a nearly-winning nominee, or even indeed for any Democratic nominee. From 1896 on, only Hoover in 1932, amongst major-party nominees, had carried so few as one small county win. However, it was in 1960 that, for the first time since before 1896, the Democrat carried fewer small-county wins than the Republican in a close election, or when the Democrat was winning:
1896: Bryan 15-17; McKinley 10-12
1916: Wilson 18-20; Hughes 6
1940: Roosevelt 15, Willkie 5
1944: Roosevelt 12, Dewey 7
1948: Truman 15, Dewey 5
1960: Kennedy 6, Nixon 16
1968: Humphrey 1, Nixon 13
(In 1896, Bryan most likely had 16--Texas but not Louisiana--and McKinley most likely had 11--West Virginia but not Illinois. In 1916, Wilson almost certainly had 20 [including Mississippi and South Carolina].)
On the above assumption, Nixon wouldn't have quite set a record for the most small-county wins for a losing nominee in 1960--he likely tied Bryan in 1896. But 1960 would already have represented a reversal of the usual pattern. (Kennedy, on the other hand, definitely set a record for the fewest small-county wins by a winning nominee nominee from 1896 on, although not when including nominees losing close elections.) 1968 would have represented a severe exacerbation of this reversal.
In 1972, McGovern carried no small-county wins. He, of course, carried only one state, but Landon, carrying only two states in 1936, won small-county wins in both of them. McGovern became the first major-party nominee since William Howard Taft in 1912 not to win any small-county wins at all. Just 20 years earlier, in 1952, Adlai Stevenson had carried six while being on the losing end of a national landslide.
1976 represented what turned out to be a one-time reversal of this trend, as Carter won 10 small-county wins--one of which (Minnesota) was outside the South (although it wasn't in the Mountain West, a region in which Carter was swept)--and Ford won only five. This would be the last time that the Democrat would enjoy more small-county wins than the Republican. And, as with Truman, no subsequent Democrat would enjoy as many small-county wins as Carter did in 1976--nor would any subsequent Republican (some of them losing substantially worse than Ford nationally) enjoy as few small-county wins as Ford did in 1976.
1976: Carter 10, Ford 5
In 1984, Mondale became the third nominee from 1896 on to fail to win any small-county wins.
In 1992 and 1996, Bill Clinton carried 3 and 7 small-county wins, respectively. In the first occasion, he set a new record for the fewest small-county wins won by a winning nominee from 1896 on. In both cases, his Republican opponent carried more small-county wins than he did--only the third and fourth times from 1896 on (after 1896 and 1960) that the candidate with fewer small-county wins won the national election.
1992: Clinton 3, HW Bush 7
1996: Clinton 7, Dole 8
Most notably, perhaps, even though Carter and Clinton have often been compared in their ability to win a number of Southern states, Clinton won only two small-county wins in the South in each of his elections (West Virginia, and his home state of Arkansas). Carter, in contrast, carried nine in 1976, and even in 1980 carried two. Furthermore, none of Clinton's small-county wins either time was in the Mountain West. Nor were they in Minnesota. One, both times, was in Iowa; in 1996, four Northeastern states were added. (Lest 'small-county win' sound odd in relation to a win in New York involving massive margins in Brooklyn, Manhattan, Queens, and The Bronx, the fact that none of these accounted for 1/4 of Clinton's margin in New York was down to his also doing well across a number of smaller counties across upstate New York.)
The trend further accelerated in 2000, a nationally close election in which Bush had 19 small-county wins (and Gore had two). Not only was the differential between the two massive--and Gore, had he won, would have set a new record for fewest small-county wins by a winning nominee from 1896 on--but, had Bush lost, he would have easily set a record for the most small-county wins by a losing nominee from 1896 on. As it was, he easily set a record for the most small-county wins by a close-election nominee, winning or losing, since 1916 (when Wilson almost certainly won 20). This further intensified in 2004, when Bush won 22 small-county wins--now easily a record for any close-election nominee from 1896 on--and Kerry became the fourth major party nominee from 1896 on not to win any small-county wins at all.
2000: Gore 2, Bush 19
2004: Kerry 0, Bush 22
(This is interesting to keep in mind given that Karl Rove was fond of comparing George W. Bush to William McKinley--in terms of this index, at least, Bush found himself in an analogous situation to Bryan, and his Democratic opponents, in an analogous situation to McKinley, although the differential was much greater in 2000 and 2004 than in 1896.)
By this metric, the 2000 election was clearly a watershed. Bush scored small-county wins in five states in which his father had not in 1988 (one of them his father hadn't even carried). All five have voted Republican ever since, and two of them (West Virginia and Oklahoma) have been small-county GOP wins in every subsequent election. In fact, even when Bush simply scored another small-county win in Missouri and Oklahoma in 2004, this constituted the first time that the same party had scored back-to-back small-county wins in those states since 1896/1900 and 1944/1948, respectively.
Three states that had not been small-county GOP wins in either 1992 or 1996, were in 2000 and have gone on to be in every subsequent election. In two cases, this is a longer streak of continuous elections than the state has ever delivered small-county wins for either party before, at least within the last 32 elections (both Kentucky's and Oklahoma's prior record was five straight elections, for the Democrats, from 1932-1948). In 2004, Bush added Louisiana and Tennessee to his small-county wins (of these, Louisiana had not been a small-county win for his father in 1988). These two have also proceeded to be small-county GOP wins in every subsequent election, again exceeding each state's prior record of consecutive elections delivering small-county wins for the same party, at least within the last 32 elections. (Within the last 32 elections, Tennessee's longest run of delivering small-county wins to the same party in consecutive elections was three, for the Democrats, from 1896 to 1904, although Tennessee was also a small-county win for the Democrats in 1880, 1884, 1888, and 1892 as well. Louisiana, having been a state prone to not being a small-county win for anyone in the 20th century, had not so much as given either party back-to-back small-county wins within the prior 22 elections, and in fact, within these prior 22 elections, had given a nominee a small-county win a total of six times.)
In 2008, Obama carried three small-county wins, tying Clinton in 1992 for the fewest of any winning nominee from 1896 on. However, whereas HW Bush carried seven in 1992, Obama's Republican opponent, McCain, carried 14. HW Bush's 7 in '92, and Dole's 8 in '96, were--leaving aside that they were more than their victorious Democratic opponent--analogous to Dewey's 7 in 1944 (as he went down to a defeat of a similar magnitude). McCain did not set a new record for the most small-county wins by a losing nominee--that was still likely a tie between Nixon in 1960 and Bryan in 1896 (at any rate, Nixon in 1960 certainly held a share in that record)--but he carried more small-county wins than four of the six major party nominees in the close elections of 1948, 1960, 1968, and 1976 (all save Truman in 1948 and Nixon in 1960).
In 2012, Obama was reduced to only one small-county win, setting yet a new record for the fewest of any winning nominee from 1896 on. Meanwhile, Romney carried 19, finally breaking Nixon's (and perhaps Bryan's) record(s) in 1960 (and perhaps 1896).
In 2016, Hillary Clinton became the fifth major party nominee to fail to win any small-county wins at all from 1896 on. Trump won 22, tying Bush's 2004 record for close-election nominees from 1896 on. In 2020, Biden again failed to win any small-county wins, becoming the first to do so and win the national election in at least well over a century. Trump won 22 again, setting a new record for the most small-county wins won by a losing nominee from 1896 on.
2008: Obama 3, McCain 14
2012: Obama 1, Romney 19
2016: Hillary Clinton 0, Trump 22
2020: Biden 0, Trump 22
Another phenomenon we've been seeing is that almost every state Republicans carry, increasingly, are small-county wins (and, of course, every state Democrats carry are big-county wins). Is it that states prone to deliver small-county wins--'rural states'--are moving into the Republican camp (and vice versa--'big county states' are moving into the Democratic camp)? Or, conversely, is it that, once a state becomes a red state, it ceases to behave like a 'big county state' (even if it had hitherto done so), and, conversely, once a state becomes a blue state, it ceases to behave like a 'rural state' (even if it had hitherto done so)?
There is evidence of both. It is interesting that Texas has remained fairly Republican, even if diminishingly so. Texas is unusual amongst large states in that it has seldom been a 'big county win'--the last time any county accounted for 1/4 of its carrier's statewide margin was 1968, even though it was only a 3.48% win for HW Bush in '92, and a 4.93% win for Dole in '96. It is also worth noting that the two states that were small-county wins for Dukakis in 1988, as well as the three that were so for Clinton in '92, have all gone on to be Republican (Iowa and West Virginia in '88, and Iowa, West Virginia, and Arkansas in '92). In 1996, Texas--which had not been a 'big county win' for HW Bush in '92--stayed in the Republican column, whereas Florida--which had been a 'big county win' for HW Bush in '92--turned Democratic. Similarly, in 2020, North Carolina--which had not been a 'big county win' for Trump in 2016--stayed red, whereas Georgia--which had been a 'big county win' for Trump in 2016--turned blue (a somewhat unexpected juxtaposition).
It is also worth noting that, of the Route 30 states, Indiana (which was a large state through the 1940 election) has been a 'big county win' conspicuously fewer times than its fellow Route 30 states; and it is also the Route 30 state that has been the most staunchly Republican.
That wouldn't be good news for Republicans in Utah, which is perennially a 'big county win'. There aren't many places left where it wouldn't be good news for Democrats (who, again, are no longer scoring 'small county wins' in any states at all). In fact, it might also not be good news for Republicans in Pennsylvania, Michigan, and Wisconsin.
However, there is also some evidence that Republicans can turn a state that is usually a 'big county win' state red (whereupon it will cease to be a 'big county win' state), and vice versa--that Democrats can turn a state that is usually a 'small county win' state (or rural state) blue (whereupon it will cease to be a 'small county win' state). The two most obvious examples, perhaps, are Missouri and Vermont. Before 2000, Missouri--which was a large state for much of the 20th century--had been a 'small county win' very few times--just seven times between 1896 and 1996, inclusive, fewer times than Wisconsin, Michigan, or New York. In 2000, Bush carried it by the relatively narrow margin of 3.34%, and yet no one county accounted for even 1/4 of that margin (whereas St Louis County--which voted for Gore--had accounted for over 1/4 of Reagan's landslide 20.05% margin in the state in 1984). Missouri has, of course, gone on to become solidly red, with a county accounting for 1/4 of the Republican margin of victory only in 2008, when the state was decided by less than 1%. Vermont, on the other hand, as mentioned above, was a 'small county win' almost continuously from (at least) 1880 through 1984, and yet has voted Democratic in every election from 1992 on. Almost simultaneously with Vermont becoming a blue state, it also has permanently ceased to be a 'small county win' state, with a county accounting for 1/4 of its carrier's statewide margin in every election from 1988 on. (Two further similar examples would be Louisiana, which had never voted Republican more than thrice in a row before 2000-2008, and Virginia, which voted Democratic only once between 1948 and 2008.)
Comments
Post a Comment