Third parties in 2020

The Wikipedia article for the 2020 election lists only two minor party candidates: Jo Jorgensen (who got 1.18%) and Howie Hawkins (who got 0.26%). I found this somewhat odd, as, in most elections, there are many more candidates listed.

It turns out that most minor candidates did terribly. (That said, Wikipedia has listed candidates doing as terribly as the several candidates just below Howie Hawkins for other elections--but beyond Hawkins there is no obvious cutoff for another seven candidates.) Here is how all the third-party candidates who got at least 0.01% of the national vote did:

Jo Jorgensen (Libertarian): 1,865,724 (1.18%)
Howie Hawkins (Green): 405,035 (0.26%)
Rocky de la Fuente (Alliance): 88,234 (0.06%)
Gloria La Riva (PSL): 85,623 (0.05%)
Kanye West (Birthday): 70,296 (0.04%)
Don Blankenship (Constitution): 60,148 (0.04%)
Brock Pierce (independent): 49,700 (0.03%)
Brian Carroll (American Solidarity): 39,230 (0.02%)

After Carroll, there is a pretty good gap; the next-strongest candidate, Simmons, received 7,138 votes--fewer than a fifth as many as Carroll.

Now, to be fair, some of these candidates were off the ballot in a good number of states. Kanye West--by far the most independently famous of these candidates--was on the ballot in just 12 states, none of them having 14 or more electoral votes. He had write-in status in another 15 states, including New York, Pennsylvania, and New Jersey. Blankenship was on the ballot in 18 states (including Florida, Michigan, and North Carolina), and had write-in status in 12 others, including New York, Pennsylvania, Georgia, and Illinois (and his home state of West Virginia).

West and Blankenship, in particular, were two of the biggest losers in the struggle amongst third party candidates. With Blankenship--'the most well-known presidential candidate the Constitution Party has ever nominated'--the Constitution Party hit an all-time low. Even running against Pat Buchanan, it got 0.09% in 2000, and then 0.12% in 2004, 0.15% in 2008, 0.09% in 2012, and 0.15% again in 2016. In 2004, 2008, and 2012, it consistently came in fifth, after the two major parties, the Libertarian nominee, and the Green nominee (or Ralph Nader, if he was running as an independent), not necessarily in that order. 

It's hard to tell which states were the best for these nominees, as sources seem to list only their raw vote numbers for various states. But in Utah--normally one of the Constitution Party's best states--Blankenship got 0.37% and came in fifth, behind Jorgensen and West (but ahead of Hawkins). In Idaho--another normally strong CP state--Blankenship got 0.22% and came in sixth, behind Jorgensen, West, and Pierce (ahead of Hawkins, who had only write-in status in the state). In Alaska--again, traditionally a strong CP state--Blankenship got 0.31% and came in fifth, behind Jorgensen and Jesse Ventura (who was the Green nominee in Alaska).

By contrast, at the CP's peak in 2008, with the charismatic Pastor Chuck Baldwin as its nominee, the party received 1.25%  (and third place) in Utah, 0.72% (and fourth place--behind Nader but ahead of Libertarian Barr) in Idaho, and 0.51% (and fourth place, again behind Nader but ahead of Barr) in Alaska. (These were Baldwin's three best states; Blankenship wasn't on the ballot in Baldwin's fourth-, fifth-, sixth-, or seventh-best states [South Dakota, Wyoming, Oregon, and North Dakota]. In Baldwin's eighth-best state, Arkansas, Blankenship got 0.17% and eighth place, with one of the candidates ahead of him being Phil Collins, who got about 5,000 votes nationally. Baldwin had gotten 0.37% and fifth place.)

Kanye West was not the first independently famous third-party presidential nominee to crash and burn. However, he did worse than Benjamin Spock in 1972 (who got 0.10%) or Roseanne Barr in 2012 (who got 0.05%)--who may have been on the ballot in states representing more population than he was. (Barr, for example, was on the ballot in California, which alone makes a huge difference.) 

The Green Party was also something of a loser. It did still come in fourth. If Ralph Nader is counted as its de facto nominee in 2004 and 2008, it has come in third or fourth in all of the past six elections. And Hawkins did do better than the de jure Green nominees in 2004 and 2008 (who, unsurprisingly, with Nader in the race, did not do well). However, he did worse, not only than Jill Stein in 2012, but than Jill Stein in 2008 (when she received 0.36%). He also did worse than Nader in all three of his runs this century; if, again, Nader is counted as the de facto Green nominee in 2004 and 2008, then Stein's 0.36% in 2012 was the party's low-water mark this century--until 2020.

The Libertarian Party was the biggest winner. Jo Jorgensen--who was, I believe, relatively unknown before the 2020 election--notched the second-best showing of any Libertarian nominee in the party's history, ahead of Ed Clark's 1.06% in 1980 and former New Mexico Governor Gary Johnson's 0.99% in 2012.

Furthermore, Jo Jorgensen achieved what could be called 'dominance' of the 'other' vote, whereby I mean that she got 1) an outright majority of all the votes not cast for either of the two major-party nominees, and 2) over twice as many votes as her nearest competitor. Defined this way, most of the relatively noteworthy third-party nominees who nevertheless fell short of the 5% mark dominated the 'other' vote: Eugene Debs (although only in 1920), Norman Thomas in 1932, John Schmitz in 1972, and Ralph Nader in 2000 (although, surprisingly, Eugene McCarthy fell just short of capturing a majority of 1976's 'other' vote). Gary Johnson dominated the 'other' vote in both 2012 and 2016, but elections without a 'dominant' third-party candidate included both 2004 and 2008 (despite the presence of Ralph Nader, with his substantial name recognition) and 1988 (when Ron Paul was the Libertarian nominee). Indeed, Jorgensen is only the second Libertarian nominee to achieve 'dominance' of the 'other' vote.

1972
leading 'other' candidate: John Schmitz (1.42%)
% of 'other' vote: 78.5%
proportion of next-closest competitor's vote (Linda Jenness): 1,320%

1976
leading 'other' candidate: Eugene McCarthy (0.91%)
% of 'other' vote: 47.7%
proportion of next-closest competitor's vote (Roger MacBride): 429%

1984
leading 'other' candidate: David Bergland (Libertarian) (0.25%)
% of 'other' vote: 36.8%
proportion of next-closest competitor's vote (Lyndon LaRouche): 289%

1988
leading 'other' candidate: Ron Paul (Libertarian) (0.47%)
% of 'other' vote: 48.0%
proportion of next-closest competitor's vote (Lenora Fulani): 198.8%

2000
leading 'other' candidate: Ralph Nader (2.74%)
% of 'other' vote: 72.7%
proportion of next-closest competitor's vote (Pat Buchanan): 642%

2004
leading 'other' candidate: Ralph Nader (0.38%)
% of 'other' vote: 38.0%
proportion of next-closest competitor's vote (Michael Badnarik): 117%

2008
leading 'other' candidate: Ralph Nader (0.56%)
% of 'other' vote: 39.6%
proportion of next-closest competitor's vote (Bob Barr): 141%

2012
leading 'other' candidate: Gary Johnson (Libertarian) (0.99%)
% of 'other' vote: 57.1%
proportion of next-closest competitor's vote (Jill Stein): 272%

2016
leading 'other' candidate: Gary Johnson (Libertarian) (3.28%)
% of 'other' vote: 57.3%
proportion of next-closest competitor's vote (Jill Stein): 308%

2020
leading 'other' candidate: Jo Jorgensen (Libertarian) (1.18%)
% of 'other' vote: 64.4%
proportion of next-closest competitor's vote (Howie Hawkins): 461%

I don't know how good a metric this is of third-party performance necessarily. (David Bergland, achieving a fairly dismal proportion of the overall national vote, achieved one of the two dimensions of 'dominance'.) It could well be that a quite obscure third-party candidate in the distant past achieved 'dominance' of the 'other' vote, since, technically, this measures third-party candidates only against other third-party candidates. That said, Jorgensen's vote share was the third-highest of any non-major party candidate's this century, after Gary Johnson's in 2016 and Ralph Nader's in 2000 (but ahead of Gary Johnson's in 2012, Nader's in either 2004 or 2008, Pat Buchanan's in 2000, Jill Stein's in 2016, or Evan McMullin's in 2016). 

I would imagine, at any rate, that the fact that Jorgensen achieved 'dominance' of the 'other' vote means that the counties she came in third in predominated on the county map, much as those that Gary Johnson came in third in in 2016 did, as shown by Nick Conway (for what that is worth--again, coming in third in a county is simply a measure of success relative to other minor candidates; since 1996, only Evan McMullin in 2016 has beaten one of the two major party nominees in a county). In fact, it seems, in all likelihood, that the map of Jorgensen-third counties is even more dominant than that of Gary Johnson-third counties in 2016 (which is not surprising, despite that Johnson did much better--both Jorgensen's share of the 'other' vote, and her proportion of her nearest competitor's vote, were the highest of any leading minor candidate since 2000). Howie Hawkins failed to hold Stein's third place in any of the counties in Hawaii, for example; Jorgensen came in third in every county in Hawaii. Other counties where Stein had come in third in 2016, but Jorgensen came in third in 2020, were Mendocino, Humboldt, Alameda, and San Francisco, California; San Juan, WA; Multnomah, OR; and Manhattan, NY (Hawkins retained third-place in Queens, Brooklyn, and The Bronx).

Jorgensen also came in third in every county in Vermont, with the possible exceptions of Grand Isle and Orleans (where she nevertheless beat Howie Hawkins--in some counties, there's not enough information to tell who came in third for sure because the aggregated 'other' vote outnumbers the vote given for any single candidate). In 2016, write-in votes came in third in every county in Vermont, which in practice presumably meant that Bernie Sanders came in third in every county in Vermont. 

One significant county in which Hawkins did retain Stein's third-place position is Prince George's, MD. This county seems to appear green on Conway's map, but he doesn't mention it in his discussion.

Jorgensen also absorbed the third-place position in most of the counties in which McMullin had come in third (or second). She came in third in every county in Utah, and appears likely to have come in third in every county in Idaho. (Wikipedia is still showing Blankenship as having gotten 11%--and third place--in Camas County, but this appears to have been a clerical error that has been corrected on Dave Leip's atlas.)

Were there any places Jorgensen didn't come in third, other than those few places where Hawkins maintained Stein's third-place finish? Well, yes--in a number of counties in Arkansas, someone else (mostly, but not entirely, Kanye West) came in third. (Arkansas, in general, seems to be an unusually good state to find unusual third-place county finishes.) West finished third in Hempstead, Lafayette, and Dallas Counties for sure, although there were a number of other counties in which we know, at a minimum, that he finished ahead of Jorgensen. Brock Pierce finished third in Desha County (an unusual county in which the third-party vote actually increased considerably relative to 2016) and in St Francis County. Gloria La Riva finished third in Lee County, Arkansas. Phil Collins came in third in Conway County. And Blankenship came in third in Little River and Nevada Counties. Even Howie Hawkins notched a third-place county finish in Arkansas (where Stein had had none): in Monroe County. So Arkansas is a state that would appear substantially less 'yellow' than in 2016.

It's hard to tell whether West's (relative) strength in Arkansas extended into other Southern states or was an Arkansas idiosyncrasy. It is harder to figure out who came in third in some states' counties than in others, and two of those states are Mississippi and Tennessee, two states where it seems likely that West may have had some more third-place county-level finishes. (However, Louisiana is a state in which it is fairly easy, and there, Jorgensen came in third in every parish.) West definitely came in third in McNairy, Houston, Decatur, and Lincoln Counties, Tennessee, and finished ahead of Jorgensen in Weakley and Hardin Counties, Tennessee. (None of these lists is necessarily exhaustive.) In Mississippi, West definitely finished third in Tunica and Bolivar Counties, and finished ahead of Jorgensen in Holmes, Leflore, Tallahatchie, Washington, Adams, Panola, Marshall, Jefferson Davis, and--most impressively--Hinds Counties. Hinds County cast 100,208 votes in 2020, and is the most populous county in Mississippi; West got 0.42% of its vote, to Jorgensen's 0.40% and 0.70% for everyone else. If West did come in third in the county, that would be a fairly substantial achievement (amid an otherwise fairly dispiriting performance). At any rate, Mississippi and Tennessee seem likely to have been a good deal less 'yellow' than in 2016 as well.

From what I've gathered, those seem likely to be the only states with significant non-yellow patches in 2020. West did come in third in at least one county in Minnesota (Mahnomen), and Brian Carroll came in a relatively impressive third in Taylor County, Wisconsin (with 1.73% and 185 votes) (making for at least seven distinct candidates who came in third in at least one county). Aside from that--and one other exception, discussed below--those are the only counties I know of in which Jorgensen didn't come in third.

---

Blankenship came in third in two counties in Arkansas. This alone represents quite a fall for the Constitution Party. 

The Constitution Party nominee has come in third in at least one county in every election this century (even in 2000, when it ran its own nominee despite Pat Buchanan being a candidate--Howard Phillips came in third, ahead of Nader and Buchanan, for example, in Columbia County, Arkansas). In 2004, Michael Peroutka came in third in nine counties in Utah (including Utah County, in which he got over 1% and over 1,000 votes), and in 18 counties in Idaho casting at least 1,000 votes. (He also actually did come in third in Camas County, which cast 590 votes that year, and where he got five votes.) He came in a relatively strong third in Pottawatomie County, KS (with 2.27% and 200 votes); this county had been an atypically strong patch of Buchananite support in 2000. He came in third (albeit not very impressively) in Malheur, OR, as well as in Tehama, Lassen, and Mariposa Counties, CA (in the last election that the party would be on the ballot in the state). (He also, a bit surprisingly, came in third in Prince William, VA.)

In 2008, its high-water mark, Chuck Baldwin came in third in 20 of Utah's 29 counties, in 12 counties in Idaho, in Elko County, NV, in two relatively small counties in Colorado (Dolores and Rio Blanco), and in one (small) county in New Mexico (Catron). 

The Mountain West (which Chuck Baldwin called 'the tip of the spear in the freedom fight') had always been a focus of Constitution Party support, but arguably it had never been as expansive across the region as in 2008. (Baldwin was not on the ballot in either Arizona or Montana, the latter the state he ended up moving to.) 

Baldwin came in third in at least three counties that cast over 100,000 votes, and in which he received at least 1,000 votes: Utah County, UT (where he got 1.83% and 2,877 votes), Davis County, UT (where he got 1.06% and 1,175 votes), and Greenville County, SC (where he got 0.82% and 1,566 votes). He also--almost certainly--came in third in his then-home county of Escambia, FL (another county casting over 100,000 votes), although with just 0.33% and 508 votes. ('Other', not including Nader, received 0.34%/533 votes.) He came in third in Al Gore's home county of Smith County, TN, and in both Pottawatomie and Wabaunsee Counties, Kansas. 

In 2012, the party nominated Virginia Congressman Virgil Goode. This was arguably the first time--the second being with Blankenship--that the party sought (or welcomed) a big-name outsider, to its detriment. Goode ended up getting 0.09%, less than either Peroutka or Baldwin, and only a couple hundredths of a percent more than Phillips had gotten in 2000 (when, again, Pat Buchanan was a national candidate). Goode came in a reasonable third in a number of counties in his native Virginia, with his best result being 2.58% in Franklin, VA (he also came in third in Henry, Pittsylvania, Halifax, Charlotte, Mecklenburg, Lunenburg, Nottoway, Cumberland, Buckingham, Highland, Bath, Alleghany, Botetourt, Appomattox, Patrick, Pulaski, Craig, Bland, Wythe, Grayson, Carroll, Buchanan, Lee, Brunswick, and Sussex Counties). However, despite being a Southerner, he failed to come in third in Union County, Arkansas, a county in which Pat Buchanan, Peroutka, and Baldwin had all come in third in 2000, 2004, and 2008, respectively. He fell to fourth place, well below Gary Johnson's level, in Greenville County, SC. He did still come in third in Smith County, TN, although he ran well below Baldwin's percentage.

While not doing particularly better than his predecessors in the South (outside Virginia), Goode plummeted in the Mountain West. In Utah, he came in third in one county, Millard. In Davis County, he came in fifth, behind Rocky Anderson. In Idaho, he likewise came in third (probably) in only one county, Bear Lake (where he and 'others' both got 0.39%--or 11 votes). In Nevada, he did come in third in Lincoln and Lander Counties, even getting over 1% in both, but these were much less significant than Elko--Baldwin got 156 votes from Elko County (even though this was just under 1%); Goode got 23 votes from Lander and 25 from Lincoln. In Colorado, he came in third in deep-blue Costilla County, and even got nearly 1% of the vote--but again, this was an insignificant enough county that this amounted to just 18 votes.

(Goode did come in third in a number of counties in Michigan, which would ordinarily be impressive, except that Gary Johnson was not on the ballot in Michigan--so the race for third place was essentially between Goode and Stein. Even so, he did not come in third in Macomb, Hillsdale, or Ottawa Counties.)

In 2016, the CP nominated Darrell Castle. Now, Nick Conway pointed out specifically that the Constitution Party had come in third in no counties, but this is incorrect: Castle came in third in two counties (and, as far as I can tell, in only two counties): Wilkinson and Franklin Counties, MS. The number of votes he received in these two counties was 19 and 23, respectively, which means that the size of the counties and/or his performance in them (or a combination--in these cases, it was a combination) was underwhelming (unlike the Prohibition nominee's impressive third-place finish in Arkansas County, Arkansas). Still, Castle did keep alive the party's streak of coming in third in at least one county somewhere. The best county I've found for Castle is Big Horn, WY, where he got 2.11% (but ran well behind Gary Johnson).

Castle actually didn't do that badly, at least relative to previous CP nominees. His share of the national vote was only slightly lower than Baldwin's; and he received over 1,000 votes in Utah County (despite not coming in third), which Goode had failed to do. And, curiously, he got over 1% of the vote in Hawaii. But, in a year when the third-party vote grew dramatically (at least in proportional terms), the CP stayed relatively static.

And in 2020, Blankenship, again, did keep alive the CP's streak of coming in third in at least one county in every election, despite his miserable national showing. He even managed to come in third in one county in the Mountain West: tiny Mineral County, Colorado, where he got 0.66% and...five votes. (He also tied with Jorgensen for third place in Issaquena County, MS, with...five votes apiece.) 

I haven't found Blankenship's likely best county. It may have been Little River, where he got 1.60% (and 83 votes); he also exceeded 1% in Nevada County (with 1.37%/46 votes). By contrast, Baldwin had received 5.57% in Millard County, UT, and Peroutka received 2.87% in Juab County, UT (I haven't confirmed that those were their best counties, but they were their best counties in either Idaho or Utah).

He was on the ballot in Utah, traditionally one of the CP's best states, and there was no other candidate one would expect Utah to particularly like (say, Romney, or McMullin). Indeed, the Republican nominee was one who had put Utah off in his initial run. But Blankenship failed to benefit from this, finishing below not only Jorgensen but Kanye West (who has a significant presence in the Mountain West). In vote-rich Utah County, he likewise finished below both Jorgensen and West (although he did receive over 1,000 votes).

As in 2020, the party chose a Southerner, but Blankenship did not do particularly well in his home region. It probably didn't help that he wasn't on the ballot in his home state of West Virginia, or in any neighbouring state--although he did have write-in access in West Virginia, and is listed as having received...25 votes in the state. He was on the ballot in Tennessee, but failed to finish third in Smith County. He also failed to finish third in Union County, Arkansas (a significantly larger county than either Nevada or Little River). 

At the same time, he failed to match his fellow Southerner's abilities, meagre as they were, in the Mountain West. He came in third in no significant county (and barely in any county at all) in the region. In Lander, he got five votes (or 0.18%). In Lincoln, NV, he got eight votes (or 0.33%). (In Elko, he got 0.31%, as compared to Goode's 0.92%.)

---

Third-place finishes are relatively hard to find unless they are particularly noteworthy; historically, they have also been little to celebrate. John Schmitz in 1972 and Ross Perot in 1996 came in second in counties (although, somewhat surprisingly, John Anderson did not), which is qualitatively more significant than coming in third, inasmuch as it means one is beating at least one of the two major-party nominees. Since 1996, again, only Evan McMullin has managed this anywhere (despite that many counties are small and landslide counties--in Billings and Slope Counties, ND, Hillary Clinton received 59 and 43 votes, respectively, and Gary Johnson, 33 and 19, in 2016; although Johnson didn't come that close to catching Clinton percentage-wise in either county, that's not a lot of individual persons who would have had to change their minds).

One interesting third-place county finisher, however, was John Hagelin, of the Natural Law Party; he came in a very strong third in his home county of Jefferson, IA 1992, 1996, and 2000 (which meant, of course, that in the first two instances, he was beating Ross Perot). In 1992, he received 22.77%; in 1996, 21.28%; and in 2000, 14.69%; and although Jefferson is not an extremely large county, Hagelin got over 1,000 votes each time. (Jefferson, IA wasn't even the only county Hagelin came in third in in 2000, however; he also came in third in Perry County, Arkansas, with 1.87%/75 votes.) Particularly in 1992 and 1996, coming in third in a county was a substantial achievement for anyone who wasn't the Republican or Democratic nominee or Ross Perot, let alone to get over 1,000 votes in a county one was coming in third in.