Getting Started With Prediction Markets

By Ben Golden on December 04, 2015
Joining a prediction market can be confusing and anxiety-inducing. It's easy to be overwhelmed by all the questions, to not understand the forecasting interface, or to have trouble forming opinions to base forecasts on. All of this is pretty natural--as a now-experienced forecaster, I can remember these feelings the first time I joined a prediction market. In this post I'll address a few specific emotional barriers that make it difficult to start forecasting.


"I'm not qualified"
"There's no way I could be better than other forecasters"

It's easy to be intimated by other forecasters. They've made more forecasts, and have higher scores. In their profiles, they might have called themselves a "superforecaster", or a "PhD". But that doesn't mean they're more qualified, or better forecasters. Indeed the research shows that credentials have little relationship to forecasting ability. The only way to know whether you're a good forecaster is to make some forecasts, and let the prediction market determine who's most accurate.

"I don't understand this question"
"The current forecasts all seem about right"
Some prediction market questions ask about unfamiliar topics, and sometimes even after researching a topic, you'll basically agree with the existing forecast. That's okay--you don't have to forecast on every question. Sometimes I prefer to wait and watch how a question changes over time. If one answer's probability suddenly spikes, try to figure out why. Did the forecaster who changed the probability leave a comment? Was there a major news event driving the change? If you thought a forecast was reasonable and then someone changes it, you may want to revert, or partially revert the change.

Following prediction markets is a great way to learn about a topic (and about forecasting), and over time you may start to notice trends. If a market's probability has been hovering between 40% and 50%, and then jumps way outside that range (say to 10% or 85%), you might now feel that the current forecasts aren't so right.

"I don't feel very confident in my opinion"
"I don’t want people to judge me if I get something wrong"
Getting things wrong is a natural part of forecasting. The keys are 1) getting things right more often than not, and/or getting things really right and only slightly wrong, and more importantly, 2) learning to become a better forecaster. A good approach when starting out is to only make small forecasts. If the current prediction is 75% when you think it should be 35%, you might start by changing it to 60% or 70%. Leave a comment explaining why you think it should be lower, and see how other users react. You can even express your low level of confidence. Here's an example:
"I'm not exactly an expert, but from what I'm reading, this event seems pretty unlikely. Here are a couple articles I found suggesting why."

You might have reservations about "helping" other forecasters, but posting comments is a great way to learn, especially when other forecasters respond. (It also helps you remember why you made a forecast in the first place). It's also important to realize that many other forecasters are feeling the same anxieties you are, even experienced ones. It's often hard to know whether an opinion is sound until you express it, discuss it with others, and wait for predicted events to unfold. The good news is, regardless of whether a forecast is right or wrong, you're going to learn something. Either you learn that you're a great forecaster, or you learn that a prediction was off, and can improve your forecasting skills for the next question.

Sign up for free at SportsCast and AlphaCast to start forecasting today. If you liked this post, follow us on Twitter at @cultivatelabs, and sign up for our monthly newsletter.

Ben Golden/@BenGoldn is an Engineer at Cultivate Labs.

prediction markets crowdsourced forecasting