In the four years that it took to negotiate this peace deal, Colombia has been moving inexorably towards October 2, the day that the people could have their say about the deal that would end the five-decade long war. The polls predicted an easy win for the "yes" side. The government's negotiators and the guerrillas (FARC) campaigned for a strong "yes" vote. This was the best deal that could be had, they said.
There are around 34 million Colombians eligible to vote out of a population of around 48 million. My own prediction was that about half of that would vote in the referendum and around 70% of them would vote "yes".
But the 'No' won with 6,422,136 votes, defeating the 'Yes' who came in with 6,361,762. A difference of 60,374 voters. A difference of less than half a percentage point of the 12.8 million who voted. An even smaller fraction of the 21.2 million who didn't vote.
With peace at stake, why was abstention so high? High abstention is a feature of recent elections - it was high in the 2014 election as well - with 32.9 million eligible voters in 2014, just 14.7 million voted, and only 7.8 million of those voted for the winner, President Santos.
But Santos and the peace bloc weren't able to get that number to vote "yes". What happened?
The polls predicting an easy "yes" victory may have played a role. Why would "yes" voters feel the need to vote if the outcome was a foregone conclusion? The "no" side, by contrast, was mobilized by the ever-polarizing ex-President and war candidate Alvaro Uribe Velez.
The areas most affected by the war voted "yes", while most of the cities voted "no" (Bogota and Cali, however, voted "yes"). Hurricane Matthew may have played a role, since the Caribbean Coast has been severely affected by the war and was expected to be a "yes" stronghold.