Colombian presidential election, June 17 2018

[Hash]

Hash - correct me if I'm wrong - but it seems like Fajardo lost because of his weakness in the Carribean region. He was unable break 10% in any of the coastal departments NE of Antioquia. What was that the result of?

He was expected to be very weak in the Caribbean, so this isn't really a surprise although in such a close race for second it definitely hurt him significantly.

The Caribbean, to paint with a broad brush and reinforce stereotypes, likes machines or populists. In this election, particularly the latter. Petro is a native son (born in Ciénaga de Oro, Córdoba) and the Caribbean likes their native sons, since they haven't had a president since Rafael Núñez in the nineteenth century, even if they've moved away.

[seb_pard]

Petro doing really good with the afro-colombian vote.

Honestly I think these results open a new campaign, Petro did as expected but Fajardo did much better (and I think Duque performed badly), so this is an opportunity to Petro to reapproach his campaign. But I believe Duque is the favorite by far (90%).

[Oryxslayer]

Petro doing really good with the afro-colombian vote.

Honestly I think these results open a new campaign, Petro did as expected but Fajardo did much better (and I think Duque performed badly), so this is an opportunity to Petro to reapproach his campaign. But I believe Duque is the favorite by far (90%).

Definitely agree. Pre-election, this was a Duque walk in the park. Post-election, there are now enough left-wing voters for Petro to win. However, if he is going to win over those types he is going to need to change his whole style, something that I doubt is possible.

Statistically, Pre-First Round was probably 99% Duque, 1% others. Post-First round, 90%-10% is definitely a good rating. If Fajardo made it in, then that number would be much closer to 50-50.

[Hash]

I've taken the freedom of using red rather than purple for Petro, although I may change my mind later.

[Velasco]

I'm a bit shocked by these results. The performance of Sergio Fajardo and the failure of Vargas Lleras and his machinery network are extraordinary. Also this is the best result ever for a left-wing candidate in Colombia. Iván Duque performed better than last polls predicted. Petro, Fajardo and De la Calle got more votes than Duque and Vargas Lleras. I expected the opposite. It's a pity that Fajardo is not going to fight the runoff, because he could have had better chances to defeat Duque. I guess that Polo and Alianza Verde will endorse Petro, but the endorsement of Fajardo is leds likely. Duque is the favourite, but his margin of victory might be narrower than expected. Fascinating outcome, in any case.

[RodPresident]

I'm a bit shocked by these results. The performance of Sergio Fajardo and the failure of Vargas Lleras and his machinery network are extraordinary. Also this is the best result ever for a left-wing candidate in Colombia. Iván Duque performed better than last polls predicted. Petro, Fajardo and De la Calle got more votes than Duque and Vargas Lleras. I expected the opposite. It's a pity that Fajardo is not going to fight the runoff, because he could have had better chances to defeat Duque. I guess that Polo and Alianza Verde will endorse Petro, but the endorsement of Fajardo is leds likely. Duque is the favourite, but his margin of victory might be narrower than expected. Fascinating outcome, in any case.

But Petro is somewhat weak and as president, he'll need to compromisse with some of the establishment if he wins. But as Brazilian, i fear a Duque win as he's very likely to push for a war with Venezuela and that he can bring Brazil into the war as one way of keeping Congress and Temer's coalition in power.

[LimoLiberal]

Damn. I was hoping for Fajardo. Duque it is :-(

Anecdotally, my friends in Santa Marta were all in for Fajardo, and I thought they might be a better representation of the Carribean than they are. I guess not...

Our taxi driver in Santa Marta (family visited Colombia earlier this year) who was a recent university graduate was also a staunch Fajardo supporter. He made it clear that most of his cohorts liked Petro, however. He disliked the leftist guerrilla groups and Petro's past membership in M-19. Shows you that radical centrists exists everywhere, lol.

[Skye]

I like how Duque easily won most municipalities that border Venezuela.

[Velasco]

I'm a bit shocked by these results. The performance of Sergio Fajardo and the failure of Vargas Lleras and his machinery network are extraordinary. Also this is the best result ever for a left-wing candidate in Colombia. Iván Duque performed better than last polls predicted. Petro, Fajardo and De la Calle got more votes than Duque and Vargas Lleras. I expected the opposite. It's a pity that Fajardo is not going to fight the runoff, because he could have had better chances to defeat Duque. I guess that Polo and Alianza Verde will endorse Petro, but the endorsement of Fajardo is leds likely. Duque is the favourite, but his margin of victory might be narrower than expected. Fascinating outcome, in any case.

But Petro is somewhat weak and as president, he'll need to compromisse with some of the establishment if he wins. But as Brazilian, i fear a Duque win as he's very likely to push for a war with Venezuela and that he can bring Brazil into the war as one way of keeping Congress and Temer's coalition in power.

Of course Petro would have to reach a compromise with some elements of the rstablishment, in case a miracle occurs and he manages to defeat Duque. It'd be quite complicated, given the personality of the candidate. Maybe one day a leftist will live in Casa de Nariño, but I don't think that Petro will win this time. I wish I'm wrong. Even Uribe's boy will have to forge alliances with other groups in the fragmented Congress. I don't think that Duque is going to launch a war against Venezuela.There's  a difference between realpolitik and campaign rhetoric. Iván Duque is a 'techocratic' neoliberal boy, not a lunatic right-winger like Alejandro Ordóñez.

The addition of vote percentages got by Petro, Fajardo and De la Calle is similar to the percentage obtained by Santos in the 2014 runoff. The combined vote percentages of Duque and Vargas Lleras similar to the result of Zuluaga in the 2014 runoff.

Polls suggest that Fajardo would have capitalized better the anti-Uribe vote.

[Skye]

I'm a bit shocked by these results. The performance of Sergio Fajardo and the failure of Vargas Lleras and his machinery network are extraordinary. Also this is the best result ever for a left-wing candidate in Colombia. Iván Duque performed better than last polls predicted. Petro, Fajardo and De la Calle got more votes than Duque and Vargas Lleras. I expected the opposite. It's a pity that Fajardo is not going to fight the runoff, because he could have had better chances to defeat Duque. I guess that Polo and Alianza Verde will endorse Petro, but the endorsement of Fajardo is leds likely. Duque is the favourite, but his margin of victory might be narrower than expected. Fascinating outcome, in any case.

But Petro is somewhat weak and as president, he'll need to compromisse with some of the establishment if he wins. But as Brazilian, i fear a Duque win as he's very likely to push for a war with Venezuela and that he can bring Brazil into the war as one way of keeping Congress and Temer's coalition in power.

Of course Petro would have to reach a compromise with some elements of the rstablishment, in case a miracle occurs and he manages to defeat Duque. It'd be quite complicated, given the personality of the candidate. Maybe one day a leftist will live in Casa de Nariño, but I don't think that Petro will win this time. I wish I'm wrong. Even Uribe's boy will have to forge alliances with other groups in the fragmented Congress. I don't think that Duque is going to launch a war against Venezuela.There's  a difference between realpolitik and campaign rhetoric. Iván Duque is a 'techocratic' neoliberal boy, not a lunatic right-winger like Alejandro Ordóñez.

The addition of vote percentages got by Petro, Fajardo and De la Calle is similar to the percentage obtained by Santos in the 2014 runoff. The combined vote percentages of Duque and Vargas Lleras similar to the result of Zuluaga in the 2014 runoff.

Polls suggest that Fajardo would have capitalized better the anti-Uribe vote.

What? I dunno, but the prospect of a presidential candidate who idolizes Hugo Chávez and took way too long to denounce Maduro's regime just because it became a hot topic for the election should scare anyone.

[Velasco]

What? I dunno, but the prospect of a presidential candidate who idolizes Hugo Chávez and took way too long to denounce Maduro's regime just because it became a hot topic for the election should scare anyone.

Don't forget Fidel Castro, Karl Marx and Vladimir Lenin

[seb_pard]

The husband of an uncle is Colombian (living in Chile) and voted for Fajardo, he is anti-Uribist and voted for Fajardo. He is sad and depressed and says that at the moment he will not vote for Petro (although he says that he will think deeply) in the run-off, so his evolution would be a good reference on the campaign.

[Hash]

A map of results by municipality, with 100% reporting. I used official colours (except orange for Vargas), but I realize the contrast might be strange and not to everyone's liking. Zooming in will show a much clearer contrast between colours.


Full size image: https://i.imgur.com/7LyqDaX.png

This is a fascinating map with some intriguing patterns which I'll spend a few days going over. However it's very late, so I will have more in-depth analysis and commentary on this map and the results more broadly tomorrow and in the coming days (as well as more analytical maps) -- in the hopes that someone is interested in my lengthier analysis and discussion, rather than the quick and easy tweets and whatever.

[Tender Branson]

^ No Austriaposting, please.

It's not.

I just said that 53% turnout is mediocre at best.

[Lexii, harbinger of chaos and sexual anarchy]

^ No Austriaposting, please.

It's not.

I just said that 53% turnout is mediocre at best.

It's a very high turnout for a presidential election in Colombia, the highest since 1974

80% is a pathetic turnout for Malta OTOH, where you have to go to 1966 to get a turnout in the very high 80

[WMS]

However it's very late, so I will have more in-depth analysis and commentary on this map and the results more broadly tomorrow and in the coming days (as well as more analytical maps) -- in the hopes that someone is interested in my lengthier analysis and discussion, rather than the quick and easy tweets and whatever.

Doooooooooo iiiiiiiiiiiiiiiit.

Well, damn it, I was within 2 percentage points of having someone to root for in the runoff. Because between a Uribist and a Chavist, I would very reluctantly vote for the Uribist, or just not vote. And as others in this thread have shown, there's plenty of actual Colombians who agree with me.

[DavidB.]

Only had time to read this now. Amazing contributions, Hash, thanks so much.

Got 57% for Duque and De La Calle, 39% for Fajardo and Vargas Lleras and 29% for Petro in the Votematch test.

[Velasco]

You may like him or dislike him, but regardless of what Uribe and his supporters say Petro isn't a Chavista. He has tried to explain during the campaign that his model has little to do with Venezuela. Also, he has condemned the "democratic involution" (or something in the like) triggered by Maduro in the neighbouring country. You can argue that he was a bit too late in making such statements. Anyway no: Gustavo Petro is a leftist but he's not "castrochavista" (translation: diehard supporter of Hugo Chávez and Fidel Castro).  

Regarding Venezuela: is there a correlation between the strong performance in the city of Cúcuta and Norte de Santander department with the refugee flood?

Also, average turnout in Colombia has been around 45% in the last 15 years or so. 53% is a modest turnout, but it's rather good for Colombia (and even better for Chile).

[Starry Eyed Jagaloon]

Does anybody know why Fajardo didn't do better in Antioquia, seeing as he's from there, and Medellin would seemingly like him?

[Hash]

Does anybody know why Fajardo didn't do better in Antioquia, seeing as he's from there, and Medellin would seemingly like him?

LSV had a very good article about this about a month ago: http://lasillavacia.com/silla-paisa/cinco-razones-por-las-que-fajardo-no-es-profeta-en-su-tierra-65584

If you can't read Spanish, here's a summary of the five reasons they cited (which I agree with)

1. The political class opposes him. He never governed with them as mayor or governor, he didn't negotiate 'quotas' as is usually the norm and often publicly attacked them as corrupt or incompetent.

2. He's paying for past administrative mistakes and failures as mayor and governor. The most oft-cited one is that he 'bankrupted' the department, which isn't true, but he did increase the debt as governor.

3. The infrastructure symbols of his administrations aren't aging well. The biggest high-profile one is the Parque Biblioteca España, inaugurated by King Juan Carlos of Spain and celebrated as a symbol of Medellín's revival and urban renewal in a low-income area (Comuna 1), is now a white elephant and falling apart. Of the much vaunted 80 'parques educativos' he promised as governor, not all were built and others were left unfinished and others are now abandoned for lack of local funds.

4. He lost power in Antioquia to his political rival, Luis Pérez, in 2015. Fajardo's gubernatorial candidate was creamed in 2015, finishing third, and Luis Pérez became governor and has spent a lot of time throwing dirt on Fajardo's record and overturning many of his landmark policies. Luis Pérez, a former Liberal elected in 2015 with CR, Liberals and Conservatives, is/was close to Vargas Lleras but likely played for Duque this year, because his nephew is a CD senator-elect.

5. His coalition with the Greens and especially the Polo didn't sit well in conservative Antioquia and with his former allies. As mayor and governor, Fajardo's strength was to be a progressive reformist while not upsetting paisa uribismo, and throughout his political career, he was supported by a lot of the regional business elite in the Grupo Empresarial Antioqueño, the Colombian keiretsu. The GEA didn't appreciate his alliance with Robledo and López, and he lost much of his business support, particularly when it appeared that he would relegated to a distant third. At the other end, the local Polo was Fajardo's most consistent opposition as mayor and governor, so like elsewhere, they felt ill at ease in this coalition (but the Polo is weak in Antioquia).

I would also add that Antioquia is Uribe country, and it has voted for the uribista candidate in a landslide in every single national election since 2002.

[Hash]

OK, some comments on the actual results before looking at some of my map's revelations.

1. This is a major victory for uribismo and the new, modern, '2.0' face of uribismo, Iván Duque. He didn't win by the first round, but nobody really expected him to do so anyway, so his result is being interpreted as it should be: a clear victory which confirms the strength of uribismo and the right in Colombian politics. With 7.56 million votes, Duque shatters records: 1.43 million more votes than the 11-M primary (which many thought was a ceiling), 652.6k more votes than Zuluaga in the 2014 runoff and - perhaps most strikingly - 171.8k more votes than Uribe himself in 2006 (when he was reelected with 62%, amidst much lower turnout). This result confirms Duque's individual strength as a candidate, able to placate the far-right/furibista core with a traditional, conservative message, while also appealing to a centre-right/centrist electorate which isn't necessarily uribista (if we use congressional results as a baseline, the uribista core vote is 2.5 million) with a more modern, technocratic neoliberal/neoconservative message which emphasizes his youth. The result also confirms that Uribe, even if he may act like a deranged lunatic on Twitter, is one of the smartest politicians out there: he correctly assessed the electorate's mood, and went for a young candidate with no baggage and no direct connection to Uribe's two terms and the tons of scandals associated with them. As I've been saying for months, Duque is the only candidate who has both opinion and machines, and even though the former was more important for him yesterday, the latter shouldn't be underestimated. It was/is a winning strategy, although what I'm already more interested by is what this winning electoral strategy means for uribismo 2.0 in government.

2. Gustavo Petro's result (25.1%) is not as high as some polls had suggested and he is 14% and 2.7 million votes behind Duque, a difference which will be very challenging to make up in three weeks (the gap is similar to that between the two primaries in March, where the right attracted 2.6 million more voters than the left). But there's no beating around the bush: his result (4.85 million votes) is a record-breaking historic success for the Colombian left, which is higher both than Carlos Gaviria's previous record in 2006 (2.6 million) and the turnout in Petro's own 11-M primary (3.5 million). Given the way in which Petro left Bogotá city hall in 2015, the criticism (and disciplinary proceedings) he faced and his lack of party structures accompanying him, nobody took his candidacy seriously until, basically, January and everyone felt that his strong numbers in early polls was merely name recognition. He confounded everyone, creating a genuine groundswell of popular support and public enthusiasm for his candidacy and his agenda, appealing to the indignados with an anti-establishment/anti-mafias message which spoke to many given Santos' second term, and proving that his obsession with the 'new citizenships' (nuevas ciudadanías) was actually a tangible electoral base (and he appealed to them with a modern, new/alternative left message). In an election driven by emotions and passions, Petro's success is an individual success which another left-wing politician wouldn't have been able to replicate: he is a confident, charismatic, rabble-rousing populist who can connect with voters which don't generally vote for the left and with his style he appealed to the emotions of historical Colombian liberalism (and he didn't hesitate to repeatedly draw comparison to Liberal icons like Rafael Uribe Uribe, Alfonso López Pumarejo, Jorge Eliecer Gaitán and Luis Carlos Galán).

3. My view on the runoff and its possible outcome hasn't changed. Duque is the runaway favourite to win. He needs to do very little to win, while Petro needs to do a lot and he isn't the right person to do it. Duque is the only candidate with both opinion and machines (let's not pretend that uribismo doesn't do machine politics), and he is the only candidate with substantial support in basically every region of the country (Petro was crushed in paisa country). My view was reinforced upon listening to both Petro and Duque's speeches - both were good and traced a path to the runoff, but the former couldn't help but get into Trump-like boasting about his numbers and 'I have no ceiling' (he does) and regurgitating the stump speech which may not be a great idea when everybody says he needs to tone it down a bit; Duque, meanwhile, gave shout-outs to all candidates by mentioning the importance of specific issues and themes their campaigns emphasized (particularly Fajardo w. education) and clearly wanted to give a message of 'national unity' similar to that of Santos in 2010. Petro's strategy, meanwhile, seems to be similar to Santos' 2014 runoff black-and-white dichotomy (war vs. peace, politics of death vs. politics of life), which was superficially effective for Santos in 2014 but will be tougher for Petro in 2018.

4. A real success for Fajardo, even if he fell (tragically) short in the end. He defied predictions and did better than anyone, myself included, could have predicted. In an election where opinion, emotions and passions were unexpectedly powerful, his 'force of hope' anti-polarization campaign did finally end up striking a chord with, from what I am gathering, an educated and young middle-class and student electorate in the well-off major cities. The green wave which never was in 2010 finally occurred in 2018, although it wasn't strong enough to overwhelm leftist petrismo. Unlike with Petro, who has a personalist cult following, Fajardo's success is potentially replicable by others in the future (I am thinking of Angélica Lozano and Juanita Goebertus, who are both amazing women). With Petro, Fajardo's success is the proof that this election may mark a new era in Colombian presidential politics - one where machines alone can't put presidents, but where opinion with few machines may potentially put a president.

5. I don't think casual one-off observers of this election understand how massive Vargas Lleras' humiliation was. This is a man who, like Santos, has been dreaming of the presidency since he was a teenager, went through all the usual steps, waited patiently for his turn to come and followed the old playbook. In early 2017, Vargas was still the man to beat - here's a guy who had been senators for years, already run for president in 2010, served in cabinet or as vice president for 7 years and had contacts with all the congressmen and their machines. He was backed by over 50 structures and machines, no matter how nasty, from different parties who had won over 5 million votes in March, and which would have been unbeatable in any past presidential election, 2014 included. And yet, despite all that, and everyone, myself included, assuming that he was underestimated and that the machines would put him stronger than 7%, even if not in the runoff, he performed astoundingly badly. I'm still in shock at how absolutely horrible he did. He won less votes than he did in 2010 (1.40 vs 1.47 million), and as I said in my benchmarks post, that result was "an absolute humiliation for him". I kind of feel sorry for him, almost, but then I remember that his party endorsed a murderer in 2011 and that he was supported by the machine of a fat war criminal who ordered a paramilitary massacre, so sucks to be you. He won less votes than CR alone did in March (2 million). What did him in was that:

(i) this election, unlike any other before, was won by opinion-emotions-passion, and Vargas Lleras doesn't inspire any of that (quite to the contrary, people dislike him because he's kind of a dick and doesn't hide it)
(ii) because of that, machine support hurt his image, even if he did his best to avoid direct association with CR's closet of skeletons (#Mejor Vargas Lleras!)
(iii) this election wasn't played according to the old playbook and, perhaps most importantly
(iv) because of Odebrecht and the authorities paying more attention to vote buying since the Aída Merlano case in March, both the campaign and the machines were careful and the campaign didn't distribute the money needed to the machines and their operations yesterday were underfunded and unable to operate as they need to (providing transportation and refreshment to voters, buying votes etc.). It is also quite possible that people took advantage of public transportation and free refreshments courtesy of #Mejor Vargas Lleras to vote for Duque and Petro, which is a fear which one of LSV's sources mentioned a few months ago. It seems like I was quite right to be skeptical of the true strength of Vargas Lleras' machinery, and even then I still overestimated him.

Also, I guess it's also a fitting end to Juan Carlos Pinzón's pathetic candidacy.

6. An absolute disaster for the Liberal Party and sad end to De la Calle's political career. The Liberals won their worst presidential election, with just under 400k votes compared to 638.3k votes for Pardo's disastrous candidacy in 2010. Many people will be angry at de la Calle and the Liberals for potentially spoiling this for Fajardo: after 11-M, there were talks of a centrist Fajardo/de la Calle alliance and the famous 'cafecito' between both men, but it was a legal and logistical impossibility which never came through and, moreover, was shut down by the Liberal caucus and party leader Cesar Gaviria (although Fajardo's obstinate refusal to enter into an alliance with the Liberal Party, only willing to ally individually with de la Calle, didn't help). In the end, Gaviria artificially kept de la Calle's candidacy in order to negotiate his endorsement for the runoff - and, mark my words, that slimy old turd will negotiate with Duque - even as the Liberal machines completely abandoned him. The result puts the Liberal Party in its most dire straits yet - after 2010, they got out of it by finding new life in Santos' administration and coalition, and making the most of that, so my hunch is that they will do the same in 2018 by finding new life in Duque's administration and coalition (or Petro, on the off-chance he wins), regardless of how often the party has waxed lyrical about peace since 2010 (and, hell, if MIRA, a party with more principles than the reds, supports Duque after supporting the peace process, then anybody can!). The result also finally kills off the idea that Gaviria is some kind of multidimensional chess player and master political strategist (an idea created by his ostensible success in managing Santos' 2014 runoff campaign), because, yeah, you failed (just like how you failed with Pablo Escobar in 1992, you turd bucket) -- and will strengthen Juan Fernando Cristo's Liberal dissidence (Cristo has already accused Gaviria of spoiling the de la Calle-Fajardo alliance).

7. The polls were right! After the nasty memories of 2010 and 2016, I was very very skeptical of them and took them with buckets of salt, but they were quite close to the ball. I guess this just confirms what I said: "this election, unlike any other before, was won by opinion-emotions-passion". Polls measure that, so they were right. On the other hand, the Nate Silver-like forecast model (which I was always a bit skeptical of because it made very generous assumptions on machines' ability to transfer votes) was a complete disaster and I think the guy behind it should go hide in Vaupés for the next decade (he can join Vargas Lleras and Pinzón there).

In another post, I will explain some of the main takeaways from my map. I welcome any questions.

[Skye]

Who wins Bogotá in the runoff? It seems to lean slightly to the left of the country.

[Oryxslayer]

Who wins Bogotá in the runoff? It seems to lean slightly to the left of the country.

Bogota gave over 60% to Petro and Fajardo combined. This year it isn't slightly, though it previously has been marginally left. Easily Petro barring a massive dropoff in Fajardo turnout - something that isn't hard to see.

[Hash]

The map.

1. The general patterns are similar to those seen in the 11-M primaries (see my maps on p. 2 comparing left/right primary turnout), but they are no less fascinating and incredible. The patterns observed confirm what I said in my last post: this election may mark a new era in Colombian presidential politics - one where machines alone can't put presidents, but where opinion with few machines may potentially put a president. Broadly speaking, at a macro-level, the spatial patterns are not particularly unusual compared to 2014/2016: uribismo dominant in Antioquia, the central Andean region and the Llanos Orientales; Petro dominant in the Pacific, ahead but closely matched in the Caribbean.

2. There are substantial regional differences in the strength of the various candidates. As I said in my last post, Duque is really the only 'national' candidate, although he does fall below 25% in Cauca, Vaupés, Putumayo. Petro is strong on the Pacific and Caribbean coastlines, in Bogotá and remote Amazonia but very weak - too weak - in Antioquia and the Eje Cafetero, and not particularly strong in much of the Llanos Orientales or the Santanderes; Fajardo was strong in Antioquia, the Eje Cafetero, Bogotá and relatively strong in Santander, Boyacá but very weak - too weak - in the Caribbean, and not particularly strong in the Pacific. The Valle and Bogotá were the only major closely matched 3-way races.

3. I will talk in greater detail about Bogotá later, but Fajardo's performance was impressive, and a silver lining in his defeat to many of his supporters. It is a vindication of his more low-key strategy of direct contact with smaller groups of voters, in contrast to Petro and the others' mass meetings on public squares. In contrast, Petro's performance, while strong and respectable, was unimpressive to me: although he won over 1 million votes, in relative terms his percentage (29.9%) was smaller than the 32% he won with in 2011, which shows that Petro is still a polarizing figure in Bogotá and that his mixed/unpopular record as mayor weighs him down locally.

4. I'm surprised at how well Duque did in the Caribbean, and how poorly Vargas Lleras did, despite the latter having the backing of the vast majority of the region's most famous and powerful machines like Char and Gnecco. Duque came first or a strong second in every Caribbean department. Petro's support in the region was predominantly urban, winning in Barranquilla, Cartagena, Santa Marta, Sincelejo, Riohacha as well as other major towns like Soledad, Turbaco, Tolu and Dibulla. Duque won in Montería (the most traditionally uribista capital in the region, although its local Conservative machines were ostensibly for Vargas), Valledupar and San Andrés (and also in Ciénaga, Magangue and Aguachica among others). I'm surprised at how well Duque did in the Montes de María region of Bolívar, an area ravaged by the conflict. His strong results in southern Bolívar and Cesar makes more sense to me (and winning Valencia and Tierralta in Córdoba is fitting...). Petro won 57% in Ciénaga de Oro, where he was born, and also won in Sahagún (Córdoba), the old bulwark of Musa Besaile and Ñoño Elías (both in jail) and seemingly the epicentre of all corruption scandals.

5. In Sucre, where Vargas Lleras was supported by basically all of the department's traditional political machines (which had won 44% for CR in March) as well as the governor and mayor of Sincelejo, he won only 11.7% of the vote in a distant third while Petro won 44%. In Barranquilla, the fiefdom of the Char family, Vargas Lleras won only 16%, again a poor third to Petro's 38%. In Soledad, known for its corrupt politics and dirty elections, Vargas did even worse, at 12.8% while Petro won 49.6%. Vargas Lleras saved face by winning a few municipalities in Atlántico and Magdalena (where he had the governor, from the Cotes clan, and the CR machine - 34% in March), although he won just 17% overall in Magdalena. In Cesar, stronghold of the criminal Gnecco family, Vargas won just under 10% (!). It is a similar story in Córdoba and La Guajira. But the machines failed for him everywhere: in the Valle, he won only 7.9% despite being backed by governor Dilian Francisca Toro, one of Colombia's main regional barons (as well as by Roy Barreras, local CR etc), in Cauca, where he had former governor/CR senator-elect Temistocles Ortega he got only 9%.

6. Antioquia remains uribista country, now more than ever, with over 1.3 million votes for Duque in the department against less than 240,000 for Petro who finished a very poor third with 9.3%, failing to grow sufficiently from the 11-M primary. Following usual patterns, Duque did best in the rural paisa cultural heartland and the urban Valle de Aburrá (Medellín) with 53% in Medellín and 57% in Envigado (and close to 50% in Bello, Itaguï, La Estrella). In many municipalities in the rural Oriente, Suroeste, Norte and Occidente provinces he got over 70%. Petro only did respectably well in Urabá (winning in Turbo and Apartadó and coming a strong second in Carepa and Chigordó), a banana-growing region which was a leftist stronghold in the late 1980s (but later paraco-narco territory). I'm intrigued by Fajardo's victories in the mining towns of Segovia and Remedios, which also had a leftist history in the 1980s before paramilitarism killed them off.

7. Duque's best department was actually Norte de Santander (61%) rather than the usual Casanare (60.3%), which just further underlines the massive shift to uribismo in Norte (and Cúcuta) since 2014: Santos won it in the runoff in 2014, but it voted No with over 60% in 2016. Norte de Santander is in direct contact with the Venezuelan crisis and directly affected by the Venezuelan migrant crisis, so castrochavismo and 'becoming Venezuela' is even more real. Petro finished fourth with less than 10%, only winning four leftist strongholds in the Catatumbo. Duque also did well in Santander (44%), despite the CD lacking much of a local base there, and finished first by a close margin in the Bucaramanga metro area, where Fajardo performed very well with the strong alternative voto de opinión there. Petro only won the oil port of Barrancabermeja, where he had the support of the main oil workers' union (USO).

8. Fajardo's support was very predominantly urban, unlike with Duque but also Petro. Outside of Bogotá and some 'freak' cases, his only wins were in cities like Cali, Palmira, Buga, Tunja, Duitama, Sogamoso, the Sabana de Bogotá (major towns like Zipaquira and Funza, middle-class Bogotá overspill in Chía, Sopo etc) and, most strikingly to me, all three capitals of the Eje Cafetero - Armenia, Pereira (and suburban Dosquebradas, and a very strong second in Sta. Rosa de Cabal) and Manizales (and suburban Villamaría, and even a strong second in Chinchiná). In Manizales, which has a big student vibe because it has five major universities, he won nearly 50% (!). A common trait to these cities, with the exception of Palmira (which is crappy) and maybe Buga, is that they're all relatively affluent and desirable major cities, well connected with educated populations and universities.

9. Petro did well with Afro-Colombians and indigenous voters, especially in the Pacific/the south and Amazonia because he was supported by most of the major indigenous and Afro organizations and his platform on economic, social, ethnic and environmental issues speak directly to the concerns of these marginalized communities and their territories. 'Indigenous politics'/activism is strongest in Cauca (and to a lesser extent in Nariño), and he did extremely well in those areas. Petro won 55% in Buenaventura (Valle) and 47% in Quibdó (Chocó), two impoverished major cities (heavily Afro) ravaged by violence and mass corruption where there were big civic strikes/demonstrations last year, and Petro reaped the benefits of those protests (which are credited with having raised local collective conscience) in cities traditionally controlled by corrupt machine politics. However, Duque narrowly defeated Petro in Tumaco (Nariño), the most violent major municipality in the entire country, which is somewhat surprising because uribismo has always been very weak there. However, indigenous communities in the Caribbean don't have a strong tradition of leftist political activism or political organization -- the Wayúu in La Guajira, for example, basically don't vote (turnout in Uribia was 17.5%).

10. It is interesting to compare today's map to the 'old map' of bipartisan politics. It is striking how many things have changed, although there remain some similarities - Petro has inherited the Liberal tradition of the Pacific coast and parts of the Caribbean - but they're probably more superficial than anything else.



Questions welcome. I'm currently working to figure out results by 'comuna'/localidad in the major cities, where possible. More maps soon.

[RodPresident]

Hash, I have a doubt about a hypotesis that I know that isnt't happening, but can be made. Can Petro drop out from run-off to allow Fajardo entering it or Colombia is like Argentina (Menem dropout that allowed Nestor Kirchner winning without runoff)? And if this is allowed, how would electors react?