Explore What Online Gaming Can and Cannot Predict About Property Prices

A reader who works as a data analyst sent us a long email last month. He had spent a year playing an online economic simulation where thousands of players buy, build, and trade property on a shared map. Prices in the game rise and fall with player activity, in-game events, and updates from the developers. He had tried to build a model to forecast which districts would appreciate. Some parts of the model worked remarkably well. Other parts were no better than a coin flip. He wanted to know whether the same split applied to real housing markets. It does, and the line between the two is worth understanding before you buy.

He is not the only player chasing forecasts. Many people in his game community spend hours comparing projections from fan-made spreadsheets, streamers who analyze the map, and prediction sites like ankertoto. What he found most useful was not any single forecast but the habit of asking which parts of a market move for knowable reasons and which parts are noise.

What His Game Model Got Right

The parts of his model that worked were all tied to structural features of the game. Districts next to newly built transit lines appreciated reliably over the following weeks, because the game's rules increased foot traffic there. Districts where the developers announced upcoming content rose as players positioned themselves. Districts with limited buildable land held value better than open ones, because supply could not expand.

In other words, the model did well when prices were driven by rules and announcements that anyone could read in advance.

What His Model Got Wrong

The parts that failed were tied to player behavior over short periods. A district might surge for three days because a popular streamer moved there, then fall back. A rumor about a hidden bonus could inflate prices for a week before collapsing. Week-to-week swings in any individual district had almost no relationship to his forecasts. Over 40 weeks, his short-term calls were right about 52 percent of the time, which is essentially random.

His conclusion: long-term direction driven by structure was partly predictable. Short-term movement driven by crowd mood was not.

The Same Split in Real Housing Markets

Real markets behave similarly. Some forces are slow, structural, and visible ahead of time:

• Supply constraints. Areas with strict zoning, limited land, or geographic boundaries tend to hold value over long periods.
• Infrastructure. New transit, schools, and major employers tend to raise nearby values over years, and they are usually announced well in advance.
• Demographics. Population growth and household formation in a region shape demand over a decade.
• Interest rate direction. Not the exact level, but the general environment affects affordability in predictable ways.

Other forces are fast, noisy, and nearly impossible to time: month-to-month price changes, the exact moment a market turns, how a particular house will sell on a particular weekend. We discuss how location-driven value works in games and in reality in what online games understand about location and what makes value.

"Experts Can Tell You Where Prices Are Going Next Year" Deserves Skepticism

Every year brings a wave of price forecasts for the year ahead. Some come from serious researchers, some from industry groups with a stake in optimism. Even careful forecasts are often wrong by several percentage points, and they frequently miss turning points entirely. That is not because forecasters are bad at their jobs. It is because short-term prices depend on things like rate surprises, economic shocks, and changes in sentiment that no model captures well.

A more useful approach is to treat one-year forecasts as weak signals and focus on the structural factors you can actually verify. A home in an area with constrained supply, solid employment, and good infrastructure is likely to do reasonably well over ten years, whatever next year's forecast says.

A Practical Way to Use What Can Be Predicted

When evaluating a property, separate your reasoning into two lists:

1. Structural factors you can check. Zoning, planned infrastructure, school quality, employment base, flood risk, and the condition of the house itself. These deserve most of your attention.
2. Short-term factors you cannot control. Next quarter's prices, whether rates dip next month, whether a bidding war appears. These deserve a margin of safety, not a bet.

If your purchase only makes sense when the second list goes your way, it is a speculation. If it makes sense on the first list alone, with the second list as upside, it is a sound purchase. Our breakdown of what cap rate actually tells an investor applies the same thinking to income properties.

What the Analyst Changed After a Year

By the end of his experiment, the reader had simplified his in-game approach. He stopped trying to forecast weekly swings and focused on buying in districts with structural advantages, holding for longer stretches, and keeping enough reserve currency to ride out dips. His in-game net worth grew more slowly in good weeks and fell far less in bad ones. Over the full year, he finished ahead of most of the players who had tried to trade short-term momentum.

When he and his partner started looking at real homes, they used the same filter. They chose a neighborhood with a planned light rail extension, limited new construction, and a stable school district, and they ignored forecasts about whether prices would rise or fall in the next twelve months. They were comfortable with the payment, planned to stay at least eight years, and accepted that what happens in any single year is mostly out of their hands.

Randomness Is Not the Enemy

It is tempting to see unpredictability as a problem to solve. In both games and markets, it is simply part of the system. The goal is not to eliminate randomness but to structure decisions so that it matters less. That means buying for structural reasons, keeping reserves, holding for long enough that short-term noise averages out, and being honest about which parts of a forecast rest on evidence and which rest on hope. A buyer who can tell the difference has already done most of the work that matters.