A Question Nobody Has Been Asking

The results of the 2016 election surprised everybody. But there was one group of people who should not have been surprised: the leadership and data scientists of Facebook, Amazon and Google.

The whole premise of Facebook, Amazon and Google is that everyone uses their products, and so they have access to everyone’s data. They know whom you’re talking to, what you’re saying, how that person reacts to your face and what they then proceed to do. They also have the best data scientists and machine learning experts money can buy. Thus, the idea goes, they can accurately predict their users’ preferences, personalities and choices, which makes them great at putting the right ads in front of the right people at the right time, which allows them to charge advertisers a premium for performance.

So, since they had many months of everybody’s posts, conversations, browsing history and purchases leading up to the election, how is it that they were not able to accurately predict the election results weeks out? A vote is a binary decision which people make every four years after discussing with their family and friends in depth beforehand. If you have those conversations on record, and all the processing power you want, and still can’t predict their decisions, how are you going to predict whom to sell flatscreens to and what price to set?

And yet, Silicon Valley (except for Thiel) was apparently flabbergasted by the electoral outcome. They are still losing their shit over stuff Trump says and does.

How could this possibly have blindsided them?

I’m honestly asking-I don’t know.

Here are some possibilities. This is not a comprehensive list, so if you can think of anything else, let me know.

1. The motivation was not there. These companies’ senior and middle management is so professional and apolitical that it never thought of using its massive clout to forecast and influence political outcomes. This is unlikely. Eric Schmidt, Google’s executive chairman, played a key role in Clinton’s campaign. Amazon’s Bezos bought Washington Post, which has been used as a megaphone with which to attack the right (zero pretense of objectivity remains.) Zuckerberg decried “fake news” on Facebook influencing the election (a dogwhistle about people sharing things like the Wikileaked Podesta and DNC emails-nothing “fake” about them-and one which Trump completely reversed on his opponents.) Years before the election, Zuckerberg launched a group dedicated to driving US immigration policy in a way calculated to ensure a continual supply of H1B wage slave programmers. Both the staff and management of Silicon Valley companies donated very actively to Hillary, and not at all to Trump. In short, the rank-and-file, leadership and owners of Facebook, Amazon and Google are heavily politically invested in the Left.
2. The motivation was there, but the tech just kind of sucks. It’s been hugely overhyped. It’s one thing to teach a computer to play Go and another to actually process and analyze real human behavior. This is directly contrary to everything I know about the field of machine learning. On the other hand, Facebook has been getting caught faking audience engagement with ads since 2009 (and again in 2012) , charging advertisers for faked or nonexistent clicks. Google is refunding massive advertisement charges that went to fake engagement. Procter and Gamble cut $100 million from its quarterly digital advertising budget with no impact on revenues, suggesting that it was almost all wasted. This is not how I would expect companies sitting on massive high-resolution data to perform if they know how to process it into actionable intelligence at scale. But if this is the case, these people have been lying to everyone for a decade and a half. Maybe it’s just greed-if you can make $Y with good, targeted delivery, and $XY with garbage delivery, it might be a no-brainer. But I don’t think so.
3. The motivation is there, the tech is there, but the brains aren’t. Meaning that all the really smart, driven guys who could think outside cliches took their payouts and left, same as in politics. What’s left are the yes-men and idiots. Any remaining intelligent and independent-minded people are keeping their heads down. The recent Damore fiasco at Google shows the mindset, as does the fact that Schmidt’s recommendations to Hillary were blown off. Everyone there was so blinded by group-think and indimidated into conformity that they either couldn’t see what the data was telling them or didn’t want to say it.
4. Some other game is going on which is orthogonal to votes, and it went wrong, throwing an unexpected and unpredictable result. For instance, voting machines have been known to be hackable for over 10 years: see this, this, this and this. Other, more traditional methods of vote fraud have been reported; for instance, more votes were cast in Detroit in the last election than there were voters, and there are millions more registered voters than there are adult citizens. There have been reports going back years of voters who seemed like they were out of place showing up en masse. And thanks to Project Veritas, we have the guys in charge of having those voters show up from out of state explaining, in detail, exactly how they organize it. So, there’s a possibility that the election was rigged to go to Hillary, and then something went unexpectedly, systemically wrong at the last minute, in which case Facebook, Google and Amazon can’t really have been expected to predict it.
5. They did accurately predict the election’s results, but didn’t publicize their prediction. I would expect to see some serious money made by eerily prescient deals made in the 1–2 months before the election by people connected to Facebook, Google and Amazon.

These are just speculations. What’s not speculative is that if any non-governmental entity was in a position to predict Trump’s victory, had the data, processing power, personnel and motivation to do so, it was these three companies. And they didn’t. And nobody has offered an explanation of why, or even pointed out how remarkable their failure to do so was.