What it means
A quantitative strategy starts with a hypothesis about market behaviour, such as that stocks which have fallen sharply over five days tend to bounce slightly. Researchers test that idea against historical data, measure how often it worked and how much it earned, and only then write the rules that will execute it.
The output is a set of explicit conditions for entering and exiting a position, with no discretion left at the moment of the trade. Individual trades in these strategies often have a tiny expected gain, sometimes only a few cents per share.
The profit comes from repetition and from strict position sizing, in the same way an insurance book makes money across thousands of policies rather than on any single one. This is why quantitative firms care obsessively about the number of independent trades they can make.
The central danger is overfitting, which means building a model so closely tailored to past data that it has learned the noise rather than the pattern. A strategy that looks spectacular on ten years of history and falls apart in live trading has usually been overfitted.
Careful teams guard against this by holding back data the model never saw during development and by insisting on an economic reason why the pattern should exist. Costs decide most outcomes.
Commission, exchange fees, the bid-ask spread and slippage, which is the gap between the price you expected and the price you got, all bite into a thin edge. A strategy earning twenty cents a trade before costs and paying fifteen cents in costs is a very different business from one paying five.
Risk control sits alongside the model rather than inside it. Firms cap exposure to any one stock, sector or factor, set a maximum daily loss that switches the strategy off, and monitor live results against what the model predicted.
When live performance drifts away from the backtest, the honest response is to reduce size and investigate rather than to hope.
In practice
Real-world examples.
Example
A small proprietary trading firm runs a pairs strategy on two listed supermarket chains, buying the cheaper one and shorting the dearer one whenever their price gap widens beyond two standard deviations. It closes both legs when the gap narrows, and over a year the trades average a modest gain that compounds into a solid return on the capital committed.
Example
A commodities desk builds a model that trades natural gas futures around weekly inventory reports, sizing each position by recent volatility. When live results underperform the backtest for two months running, the desk halves its position size while it investigates whether market structure has shifted.
Example
An asset manager uses a quantitative model to rank 2,000 stocks on value, quality and momentum, then holds the top decile and rebalances monthly. The approach is quantitative even though it trades slowly, because the decisions come from the model rather than from analyst conviction.
Think of it
“Quant trading uses math and data to find trades-systematic, model-driven investing.
Formula
Calculation
Expected value per trade = (win rate x average win) - (loss rate x average loss). Net profit = (expected value per trade x number of trades) - total transaction costs.
Take a strategy that wins on 55% of trades, earning an average of $120 on a winner and losing an average of $100 on a loser. Expected value per trade = (0.55 x $120) - (0.45 x $100) = $66 - $45 = $21. Over 10,000 trades in a year, gross profit = $21 x 10,000 = $210,000. If each round trip costs $6 in commission and slippage, total costs = $6 x 10,000 = $60,000, leaving net profit of $210,000 - $60,000 = $150,000. Note how a rise in costs to $12 a trade would halve the strategy's remaining profit.Case study
Seen in the real world.
Meridian Quant Partners is a fictional, illustrative eight-person trading firm running a short-horizon equity strategy with $40 million of capital. Its backtest showed a 22% annual return, and the founders raised outside money on the strength of it.
In the first live year the strategy returned 6%. The post mortem, conducted honestly, found two causes: the backtest had assumed fills at the midpoint of the bid-ask spread when the desk was in fact paying most of the spread, and the model had been tuned on the same three years of data used to validate it. Neither problem was fraud; both were ordinary research errors.
Meridian rebuilt the process, holding back two years of data entirely and rerunning every backtest with conservative fill assumptions. The rebuilt strategy showed a 9% backtested return and delivered 7% live, which the founders found far more useful than the flattering number they started with. The illustrative point is that a smaller, honest expected return beats an inflated one.
Watch out
Common mistakes.
- Judging a strategy on its backtest return alone. A backtest without realistic assumptions about spreads, fees and slippage flatters almost every idea, and the gap between the two is where most strategies die.
- Believing quantitative trading removes risk because it removes emotion. The rules remove hesitation at the point of trade, but the model itself embeds assumptions that can be wrong for months at a time.
- Confusing quantitative trading with high-frequency trading. Speed-based trading is one branch of it, and plenty of quantitative strategies hold positions for weeks.
Questions
People also ask.
Do you need a large team to trade quantitatively?
No, individuals do run rule-based strategies, but competing on speed or on crowded short-horizon signals requires infrastructure that only funded firms can afford.
How many trades does a strategy need before you trust it?
There is no fixed number, but a few hundred independent trades is a common minimum before results start to say more about the edge than about luck.
Why does the same strategy stop working?
Edges get crowded as other participants find them, and market structure, regulation and participant behaviour all change, so most strategies decay rather than fail suddenly.
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%