What it means
Statistical arbitrage is a trading style that looks for small, repeating price gaps between related shares. The idea is that two similar companies, such as two large banks, tend to move together, so if one jumps ahead of the other, the gap is likely to narrow again.
The s-score turns that idea into a single number that can be compared across many stocks. The best-known version comes from academic work on trading stocks against sector funds.
A model first strips out the part of a stock's return that is explained by its sector, and the leftover movement is called the residual. The s-score measures how far today's residual is from its long-run average, divided by its usual variability.
A score near zero means the stock is trading in line with its peers. A score of plus 2 means the stock is two standard deviations above its normal position, which could signal that it is overpriced relative to peers.
A score of minus 2 suggests it is cheap relative to peers. In the published approach, traders open a short position (a bet on a price fall) when the score is high and a long position (a bet on a price rise) when it is very negative.
Entry thresholds of around 1.25 standard deviations were used in one well-known study, and positions were closed when the score moved back towards zero. Real strategies set their own thresholds and test them carefully.
For a non-specialist, the useful point is the logic. The score does not predict that a stock will go up or down overall, only that it has moved away from its usual pattern.
It works only if the old relationship still holds, which is the main risk. The name is not fully standardised.
Some data vendors and researchers use "S-score" for other ratings, such as sentiment or sustainability measures, so always check what the author means.
In practice
Real-world examples.
Example
A quantitative hedge fund calculates s-scores each morning for 400 shares against their sector funds. The system flags a retailer with a score of minus 1.8 as cheap compared with its peers and opens a small long position.
Example
A risk manager at a trading firm reviews the open positions and finds that most of them rely on scores above 2. She asks the team to test what happens if the sector relationships break down, as they might in a crisis.
Example
A finance student builds a simple spreadsheet that computes an s-score for two oil companies over three years. She finds that the gap between them returns to its average within about two weeks on most occasions.
Formula
Calculation
S-score = (Current residual - Average residual) / Standard deviation of the residual.
Suppose a bank share's residual return, after removing the sector effect, is 2.4% today, its long-run average residual is 0.6% and its standard deviation is 1.2%. The s-score is (2.4% - 0.6%) / 1.2% = 1.8% / 1.2% = 1.5. Because 1.5 is above an entry threshold of 1.25, a trader following the strategy would sell the share short and buy the sector fund as a hedge, expecting the gap to close.Case study
Seen in the real world.
Marlow Quant Partners is an entirely fictional trading firm that runs an s-score strategy on $50,000,000 of capital. In this illustrative story, the strategy makes steady small profits for three years by trading 300 shares against sector funds.
In a sudden market shock, many stocks move together and the old relationships break. Scores that looked extreme become even more extreme, and positions that were supposed to close quickly lose money for days.
The risk team cuts position sizes and adds a rule to pause the strategy when sector relationships become unstable. The illustrative lesson is that an s-score depends on the past pattern holding, and that risk limits are needed for the times it does not.
Watch out
Common mistakes.
- Treating a high s-score as a certain sign that a share will fall, when it is only a signal that a pattern may reverse.
- Using the score without checking that the relationship between the stocks is stable.
- Assuming every "S-score" in a report is the same measure, when different providers use the name for different things.
Questions
People also ask.
What does an s-score of 2 mean?
It means the residual is two standard deviations above its average, which is unusual and may signal a stock priced high relative to its peers.
Is this the same as a z-score?
It is very similar, since both express a distance from the average in standard deviations, but the s-score is applied to a residual from a market model.
Do traders always act on large scores?
No, they usually combine the score with position limits, trading costs and risk checks.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%