World Reimagined

A Look at Samotsvety Forecasting, One of the World's Best Predictors of the Future

Peering into the future is incredibly hard, even for people who have access to the most advanced data and forecasting tools. And that makes the accuracy rate of Samotsvety Forecasting all the more impressive.

The group, which is filled with some of the best forecasters in the world, doesn’t pull data from intelligence sources or spend its days in a think tank. Instead, it’s a group of online friends, who look at questions in a different way—with remarkable results.

In 2020, the Samotsvety group won one of the top forecasting competitions in the world on Infer, a prediction site that tasks individuals and teams to predict future scenarios, ranging from whether Iran will be at war within the next year to whether the People's Liberation Army will invade, blockade, or attack the main island of Taiwan in the next six months. It won again in 2021, the last full year it participated (And even though it only participated in part of 2022, it was near the top again).

Named after an obscure Russian rock band whose name meant “self-lighting/coloring stones,” Samotsvety was started as a Slack channel, where co-founders Misha Yagudin and Nuño Sempere would chat about forecasting with each other and friends. Today, there are 15 people that make up the group.

Sempere and Yagudin met at a summer fellowship at Oxford’s Future of Humanity Institute, where they both honed their forecasting skills. They soon began competing together. Yagudin is also the cofounder of Arb Research, a consulting firm whose services include forecasting to surveying the views of public figures. Sempere is a consultant, software engineer and forecaster.

Becoming a member of Samotsvety Forecasting is invitation-only. Once someone is accurate enough in their forecasting and manages to catch the group’s collective eye, they might receive an inquiry. Infer has traditionally been the best path to doing so.

The key to Samotsvety’s success, though, isn’t just picking the right members. It also comes down to how they make their predictions. Samotsvety thinks in numbers, perhaps because many have backgrounds in computer science, math and more.

That makes forecasts easier to compare—and it also helps them not over (or under) react to evidence. Because the group works together, that introduces a variety of evidence bases. Often a member will bring up a point that others haven’t considered that could alter the course of the forecast.

By breaking a big question into several smaller ones, Samotsvety deconstructs a scenario down into more palatable queries, which – when those answers are brought together—can help forecast the answer to a bigger one.

This was on display nowhere better than the forecast that brought the group to people’s attention in March of 2022, when it concluded there was just a 0.01 percent chance that London would be hit by a nuclear weapon by mid-April 2022 (At the time, with Russia’s saber-rattling, that was a growing fear amongst people).

To make their forecast, Samotsvety looked at everything from the history of nuclear close calls to the situation at the time in Ukraine. The group considered how Russian oligarchs would react to the use of a nuclear weapon. All of those factors made them confident London would be spared.

Like so many other Samotsvety forecasts, it proved to be entirely correct.

The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.

Chris Morris

Chris Morris is a veteran journalist with more than 30 years of experience, more than half of which were spent with some of the Internet’s biggest sites, including CNNMoney.com, where he was Director of Content Development, and Yahoo! Finance, where he was managing editor. Today, he writes for dozens of national outlets including Digital Trends, Fortune, and CNBC.com.

Read Chris' Bio