Glossary

Every key term from the course, in plain English. Each one links to the lesson where it's explained properly — and you can deep-link any definition with its #anchor.

Adjusted price

A price series rewritten so stock splits and dividends don't appear as phantom jumps. Always backtest on adjusted prices. Module 3

Backtest

Running a strategy's rules over historical data to tally how it would have performed. Module 4

CAGR

Compound annual growth rate — the single yearly rate that takes you from start balance to end balance, compounding. Module 2

Correlation

A −1 to +1 measure of whether two assets move together. Not the same as causation or cointegration. Module 1

Deflated Sharpe Ratio

A Sharpe ratio discounted for how many configurations you tested — the honest number after a big search. Module 5

Discipline gap

The shortfall between backtested and realised returns, caused by human behaviour overriding the rules. Module 6

Discretionary trading

Deciding trades by in-the-moment human judgement rather than a fixed rule. Module 0

Drawdown (max)

The largest peak-to-trough drop in an equity curve — the worst loss a strategy put you through. Module 2

Edge

A repeatable reason a strategy wins over many trades — a small, fading tilt in the odds. Module 0

Event-driven backtest

Stepping through time one bar at a time, only ever seeing the past — slow but realistic. Module 4

Expectancy

Average profit per trade, combining win rate and payoff. Positive expectancy is the whole point. Module 2

Fixed-fractional sizing

Risking a constant small percentage of the account (often 0.5–2%) on every trade. Module 6

Kelly criterion

The bet fraction that maximises long-run growth; full Kelly is too aggressive, so use a fraction of it. Module 6

Look-ahead bias

Using information at a moment you couldn't actually have had it — a classic way to fake a great backtest. Module 3

Mean

The plain average return — your typical period gain or loss. Module 1

Moving average

The average price over the last N periods, recomputed each step, used to smooth out noise. Module 4

Multiple testing

Trying many variants and keeping the best, which inflates the winning result by pure chance. Module 5

Net return

Return after commissions, spread, and slippage — the only return you can actually keep. Module 5

Normal distribution

The bell curve; ~68% of values fall within ±1σ. Real markets have fatter tails than it predicts. Module 1

OHLCV

Open, High, Low, Close, Volume — the five numbers in a single price bar. Module 3

Out-of-sample

Data held back from tuning and used once, at the end, to judge a strategy honestly. Module 5

Overfitting

Tuning a strategy so tightly it memorises noise instead of signal — great on the past, useless ahead. Module 5

Paper trading

Running a strategy live with fake money to test execution and your own discipline before risking real capital. Module 7

Point-in-time data

A dataset reflecting exactly what was known on each historical date, delisted names included. Module 3

Position sizing

How much capital to commit per trade — the dial that turns a signal into a survivable strategy. Module 4

R-multiple

A trade's result measured in units of the risk taken — risk $100, make $300, that's +3R. Module 2

Return

The percent change in price — the common scale we analyse instead of raw prices. Module 1

Risk of ruin

The probability a losing streak wipes out your account before the edge has time to pay off. Module 6

Sharpe ratio

Return divided by volatility — return earned per unit of risk. The headline quality number in quant. Module 2

Signal

The condition a strategy watches for that triggers a potential trade. Module 4

Survivorship bias

Testing only on companies that survived to today, so the failures are invisible and the past looks safer than it was. Module 3

Systematic (quant) trading

Deciding trades by a fixed, testable rule a computer can follow. Module 0

Validation

Checking whether a backtest's result is real, not luck, hindsight, or a biased dataset. Module 0

Vectorized backtest

Computing all signals across history at once — fast and simple, but easy to leak future data into. Module 4

Volatility (σ)

The standard deviation of returns — how widely they scatter around the mean. Module 1

Walk-forward

Repeatedly fitting on one window and testing on the next, rolling through time for an honest verdict. Module 5