Bet Sizing Bias: Why Most Investors Get Position Size Wrong
Feb 3, 2026
Two types of investors consistently destroy their returns. The first concentrates heavily in high-conviction plays, certain their research provides an edge. They blow up spectacularly when a single position moves against them. The second spreads capital across dozens of holdings seeking safety in numbers. They underperform index funds year after year while doing more work and paying higher fees.
Both errors look different on the surface. One appears reckless, the other cautious. Both stem from the same psychological root: bet sizing bias driven by an inability to accurately assess one’s own edge. The concentrated investor overestimates their information advantage. The over-diversified investor underestimates it or never tries to measure it at all. Neither approach is based on honest self-assessment. Both are reactions to uncertainty dressed up as strategy.
The paradox runs deeper than most realize. Conviction and accuracy are not the same thing. Studies consistently show that confidence levels bear almost no relationship to actual predictive ability. Traders who feel most certain about a position are not more likely to be right. They are simply more likely to size the position larger. That gap between feeling and reality is where fortunes disappear.
The Overconfidence Engine
Bet sizing bias operates through a specific mechanism: overconfidence in edge estimation. Most investors believe their best ideas deserve outsized positions. The logic feels sound. If you have done extensive research and hold high conviction, why would you not bet big? The problem is that conviction measures how you feel, not how much you actually know.
Information asymmetry is real but rare. Most retail investors trade on the same information available to everyone else. They read the same filings, watch the same earnings calls, and consume the same analysis. The belief that this process generates unique insight is almost always wrong. What feels like edge is usually pattern-matching on public data that millions of others have already priced in.
The overconfidence compounds with experience. The more you know about a sector or company, the more confident you become. But your accuracy does not increase at the same rate. Expertise creates conviction faster than it creates correctness. This is why seasoned investors often size positions more aggressively and sometimes blow up more spectacularly than novices. They trust their judgment precisely because they have been rewarded for it before.
Bet sizing bias turns this confidence into capital allocation. The feeling of certainty translates directly into position size. When the position works, it reinforces the confidence. When it fails, the loss is catastrophic because the size was too large. The feedback loop runs in one direction: toward overconcentration in ideas that feel right.
Long-Term Capital Management: Genius Meets Reality
If overconfidence could be solved by intelligence, Long-Term Capital Management would have succeeded. The hedge fund launched in 1994 with two Nobel laureates on its team and some of the sharpest minds on Wall Street. Their models were sophisticated. Their track record was exceptional. Annual returns of 40% in the early years. Their position sizing was based on mathematical precision that retail investors could not match.
By 1998, LTCM had leveraged its positions to roughly 25-to-1. The models said this was safe because the positions were hedged and correlations were stable. Then Russia defaulted on its debt. Correlations that had held for years broke simultaneously. Positions that were supposed to offset each other moved in the same direction. Within weeks, LTCM lost $4.6 billion and required a Federal Reserve-coordinated bailout to prevent broader market contagion.
The lesson is not that models are useless. The lesson is that bet sizing bias affects everyone, including Nobel laureates. LTCM sized positions based on confidence in predictions that assumed the future would resemble the past. That assumption was embedded so deeply in their framework that they could not see it as an assumption at all. They treated it as fact. The market treated it as hubris.
If the smartest people in finance cannot accurately assess their own edge, retail investors should be deeply skeptical of their ability to do so. Bet sizing bias does not discriminate by IQ. It discriminates by humility.
Why Buffett Is Not Your Template
Warren Buffett runs a concentrated portfolio. His largest positions often represent 30-40% of Berkshire’s equity holdings. This seems to contradict everything about the dangers of concentration. If concentration destroys returns, how does Buffett survive?
The answer is that Buffett’s concentration differs structurally from retail concentration. He has an actual information advantage. His edge is not stock-picking in the traditional sense. It is access to deals that regular investors cannot get. When Buffett bought $5 billion of Goldman Sachs during the 2008 crisis, he received 10% preferred dividends plus warrants. Regular investors buying Goldman stock on the same day got none of those protections.
Buffett also has scale advantages that invert the usual dynamics. He can influence management, negotiate terms, and wait indefinitely without redemption pressure. His holding period is measured in decades, not quarters. Mistakes have time to correct. Volatility becomes noise rather than risk.
Retail investors who concentrate like Buffett without Buffett’s structural advantages are not following his strategy. They are imitating the visible output while lacking the invisible inputs. The bet sizing bias here is particularly dangerous because it comes wrapped in the authority of a successful example. Feeling like you have an edge because Buffett has an edge is not the same as actually having one.
Sizing by Volatility and Liquidity
If you cannot accurately assess your own edge, and you probably cannot, position sizing needs to be based on something more objective. Two factors work better than conviction: volatility and liquidity.
Volatility-based sizing adjusts position size inversely to price movement. A stock that swings 5% daily gets a smaller allocation than one that moves 1% daily, regardless of how confident you feel about either. This approach acknowledges that bigger swings create bigger risks of being stopped out or panicking at the wrong moment. It removes conviction from the equation entirely.
Liquidity-based sizing limits positions based on how easily you can exit. If a stock trades thin volume, a large position becomes a trap. You cannot sell without moving the price against yourself. Limiting position size to some fraction of average daily volume ensures you can exit within a reasonable timeframe without becoming your own worst counterparty.
Neither approach requires you to know your edge. Both impose external discipline that protects you from your own overconfidence. The Kelly Criterion looks elegant on paper, but it requires accurate inputs for win probability and payoff ratio. Garbage in, garbage out. Volatility and liquidity are measurable. Your edge is not.
The Structural Problem No Formula Solves
Most position sizing advice assumes you can rationally assess your own skill level and then size accordingly. This assumption is false. Decades of research on overconfidence bias show that people systematically overestimate their abilities. Expertise makes this worse, not better. The more you know about a subject, the more confident you become, but your accuracy does not increase at the same rate.
This creates an unsolvable problem at the heart of active investing. Optimal position sizing requires knowing your edge. Knowing your edge requires accurate self-assessment. Accurate self-assessment is psychologically impossible for most people. The investors who think they have solved this problem are often the ones most at risk.
Bet sizing bias cannot be eliminated through awareness. Knowing about the bias does not make it disappear. The solution is not a better formula or more sophisticated model. It is building systems that do not require accurate self-assessment in the first place.
Hard position limits. Volatility-based sizing. Forced diversification rules that trigger regardless of conviction. These approaches feel unsatisfying because they ignore your research and opinions. That is precisely why they work. Your opinions about your own edge are the least reliable input in your entire investment process. Build a system that does not need them.
