Predicting the Future

We have all heard about prophets and sages, and fortune tellers, but you can’t really know what the future holds. Can you?

What if I told you that there may be a scientific way to predict elections, stocks, and political changes?

You may have heard about ‘crowdsourcing’, where lots of small inputs from lots of people on the web, are combined into usable information. For example, Google Maps knows if there are cops on the freeway nearby. It gets this information from drivers using the program and pressing a button when they see a police car. The Maps program can see traffic jams in a similar way by judging numbers of cars in an area and how fast (or slow) they are going. A bunch of cars going 10mph in a 65 zone, is likely a traffic jam.

Now think about ibkr.com/forecastnow or RobinHood’s SuperBowl or Vegas sports betting. These are a form of crowdsourcing or polling. People put money on one outcome or the other. You can bet on whether Republicans gain control of the House of Representatives, for example. You can gamble on the elections. And oddly enough, all of this together is better at predicting things than polls.

Maybe because this is a sort of survey, and includes many more people than you get in an average survey. And this is people “putting their money where their mouth is”. People lie on surveys. Or they say what they would LIKE to happen, instead of what they THINK will happen. But I expect people are less likely to throw their money away. When it comes to gambling with their own money, they vote how they really expect it to happen. I think.

Limitations

Is this accurate? Betting has done a decent job or predicting many recent elections, getting 77% right according to Newsweek. [1] From experience, I would put Google Map’s ability to predict radar traps at more like 90%.

Of course this won’t help you predict the next asteroid strike, or anything that the ‘wisdom of the crowd’ can not see. But crowdsourcing can tell you where the police radar is likely to be running, what the economy might do, who might win the election, and who might likely win a football game. And maybe a gamble event where only astronomers are allowed to participate, might be better at predicting how likely an asteroid strike is, in a given year? I suspect that some crowds are better at predicting things than others. And some groups are better at making predictions in their area.

That said, don’t blame me, if you are gambling and lose. That is on you.

References

[1] https://www.newsweek.com/election-betting-odds-accurate-prediction-us-presidential-election-1964752