Base-rate neglect

Base-rate neglect is the tendency to overlook the actual statistical baseline — the prior probability of a category — when evaluating the likelihood of a specific case. Instead of starting with “90% of companies that pursue this strategy fail,” you focus on the individual company’s appealing features and estimate its success as likely, ignoring the base rate entirely.

The core mechanism

When you are told “the manager has beaten the market for five years,” your brain does not ask “what fraction of managers beat the market for five years by chance alone?” Instead, it focuses on the impressive fact (five-year outperformance) and estimates a high probability the manager has skill. But the base rate matters enormously.

If there are 10,000 fund managers and the market is random, we expect roughly 2^5 = 32 managers to beat the market five years in a row by chance. So the base rate of “five-year outperformers who have skill” is low. An investor who neglects this base rate and estimates the manager as skilled is committing base-rate neglect.

Why it happens

Psychologically, specific information feels more real and more relevant than statistical averages. The manager’s track record is concrete and vivid; the base rate is abstract. Your mind naturally weights the concrete information more heavily, even when the statistical baseline is more predictive.

Additionally, focusing only on the specific case feels more rational than relying on an average. “This manager is special because of [specific reasons]” feels like a better basis for judgment than “most managers do not beat the market.” But it is not.

Base-rate neglect in investment selection

Startup investing. The base rate of startups that fail is roughly 90%. Yet investors regularly fund startups with compelling pitches, brilliant founders, and large markets — while neglecting that even companies with all three typically fail. The base rate should anchor expectations low; specific information should modulate that estimate upward, but probably not all the way to “likely to succeed.”

Stock picking. An investor researches a company, finds strong fundamentals, excellent management, and a growing market. All of that is true and impressive. But the base rate of non-indexed companies that outperform the market is below 50% (after fees). Neglecting that base rate and estimating a high return probability is base-rate neglect.

Emerging markets. After a period of outperformance, investors flood into emerging markets, believing the region will continue to outperform. But the base rate of markets that outperform over the next decade is close to random — the long-run return differential is tiny. Neglecting that base rate while focusing on recent performance (specific, vivid information) is classic base-rate neglect.

Base-rate neglect and fund manager evaluation

One of the clearest applications is evaluating fund managers. A manager outperformed for 10 years. Investors often interpret this as evidence of skill. But consider the base rate:

• If 10,000 managers compete and performance is random, we expect some to get lucky for 10 years.
• The probability of any one manager beating the market 10 years in a row by chance is (0.5)^10 = 0.001, or 0.1%.
• But with 10,000 managers, the expected count of 10-year lucky streaks is 10.

So 10-year outperformance is not strong evidence of skill; it is what we expect from random variation in a large population. Base-rate neglect causes investors to ignore this and hire managers right after their lucky streak is over.

Base rates and the overconfidence bias

Base-rate neglect and overconfidence often work together. You ignore the base rate (most investors underperform), focus on your own positive specific attributes, and conclude you will outperform. This combination is lethal for portfolio returns.

Distinguishing base-rate neglect from representativeness

Base-rate neglect is about ignoring the statistical baseline. Representativeness is about judging probability by similarity to a stereotype. They often occur together and are sometimes hard to separate, but they are distinct.

In base-rate neglect, you know the base rate but fail to use it (“I know 90% of startups fail, but this one is special”). In representativeness, you judge by similarity without accessing the base rate at all (“this looks like a winner”).

Defenses against base-rate neglect

• Always ask: what is the base rate? Before investing in a category, find out the actual historical success rate. What fraction of growth stocks deliver high returns? What fraction of emerging markets outperform developed markets over the next decade?
• Start with the base rate as your estimate. Do not start with the specific case and adjust upward. Instead, start with the base rate and ask “what specific information would move me away from it, and by how much?”
• Use Bayesian frameworks explicitly. In professional investment settings, formalize how new information should update your prior (base rate) belief. This forces accountability and prevents base-rate neglect.
• Diversify within a category. Rather than trying to pick the “winner” among startups, hold a portfolio of startups. This forces you to confront the base rate of failure directly.
• Track outcomes. Keep a record of your predictions and actual outcomes. Over time, you will see that cases you rated as high-probability failures sometimes succeed, and vice versa — a sign you are overweighting specific information relative to base rates.

See also