We could spend all day analyzing what went “wrong” in the 2016 election, specifically how the major polling outfits missed by that much and did so almost uniformly. In fact, we already have done that here, as has everyone else. (Remington actually pegged the race pretty closely but nobody ever seems to figure them in.) Four days should be a sufficient amount of time for navel gazing even on a subject of this magnitude, so it’s time to move on to the solution. Let’s just do away with the pollsters for elections. What’s going to be a lot more fun is when we can replace them with artificial intelligence which bases its results on tweets and Facebook updates.
And don’t worry about tasking somebody with developing one because it’s already been done. In fact, there’s one which has been running since before the 2004 contest and it’s gotten every presidential winner correct, including Trump. (CNBC)
An artificial intelligence system that correctly predicted the last three U.S. presidential elections puts Republican nominee Donald Trump ahead of Democrat rival Hillary Clinton in the race for the White House.
MogIA was developed by Sanjiv Rai, founder of Indian start-up Genic.ai. It takes in 20 million data points from public platforms including Google, Facebook, Twitter and YouTube in the U.S. and then analyzes the information to create predictions.
The AI system was created in 2004, so it has been getting smarter all the time. It had already correctly predicted the results of the Democratic and Republican Primaries.
MogIA had the race called almost two weeks in advance and it did it without bothering a single person during dinnertime with an annoying phone call. But I know what you’re thinking… there’s all sorts of problems with handing over this sort of task to artificial intelligence. I agree. We’d be taking away jobs from Americans and giving them to yet another robot. Besides, we don’t need to make the IA systems any smarter, particularly after the Internet of Things knocked out half the web a couple of weeks back.
Okay, so robots are out. We still don’t have to use the pollsters. Why not just clone this monkey? He picked Trump for the win months ago. (Washington Post)
His name is Geda, and local legend has pegged him as a mystical monkey with prophetic insight — one with an impressive record of correctly predicting the winners of European soccer matches.
This time, Geda was asked to predict the winner of the U.S. presidential election and — unlike many a highly-paid pollster — he did exactly that, AFP reported, and with absolute conviction, going as far as planting a smooch on the winner’s face.
(Click for horrifying, full size picture.)
No? Okay, I’ll grant you that the monkey is kind of creepy and that cardboard cutout of Hillary in an orange pantsuit was unsettling as well. So no monkeys. Then could I interest you in this shark? (International Business Times)
Nova Southeastern University in Fort Lauderdale, Florida, for example, has set up a competition between two sharks to predict the next president. They’ve tagged a Clinton shark and a Trump shark and are tracking the miles each travels. The shark who goes the farthest by noon on Friday will be the winner, according to a news release from the university.
As of press time, the Clinton shark had swum 510 miles to the Trump shark’s 652.
When you match up the final lead in the electoral college and pair that with the registration advantage averaged over nine states, the net difference in the distance traveled by the Trump shark almost exactly matched the spread. This was clearly the most accurate prediction method tested and certainly blew the doors off of Quinnipiac. The one drawback here is you can only test two to three races with the same shark before the data retrieval person inevitably winds up being eaten. But hey… it’s a new system. We knew there would be some bugs to be worked out.