Home/Systematic Trading/Edge Identification

Edge Identification

model-buildingLevel 1 — Novice

What It Is

The process of distinguishing genuine, persistent sources of positive expected value from random noise, data-mined patterns, and unfounded beliefs — and then determining whether the edge is a structural risk premium or a temporary market inefficiency.

Correct Execution

Practitioner can articulate the specific observable phenomenon behind each trade (e.g., "vol is mean-reverting after spikes" or "stocks go up over time") and trace that phenomenon to a real economic cause independent of any backtest result. Edge is quantified numerically (expected value, confidence interval). Position sizing is derived from edge magnitude, not conviction or narrative.

Progression Levels

Diagnostic Tree

Coaching Cues

  • "Start with the phenomenon. The indicator is just the measurement tool." — Euan Sinclair; when constructing any strategy
  • "Risk premium or inefficiency? That question determines everything about how you trade it." — Euan Sinclair
  • "If you've got the edge, press it as hard as you can for as long as it lasts. It won't last forever." — Euan Sinclair; counterpoint to the common "keep positions small" advice
  • "Can you express the edge in numbers? If not, it may still exist — but you can't trade it." — Euan Sinclair

Common Errors

  1. Starting with the indicator: Building strategies by testing hundreds of moving average combinations and finding the best one → data mining, not strategy discovery → always start with the economic phenomenon, then find the measurement tool.
  2. Confusing volume for independent observations: 1,000 data points in a vol strategy backtest may be only 10 independent regime observations → massively overstates statistical confidence → count independent regimes, not calendar days.
  3. Treating domain expertise as trading edge: Being a biotech scientist doesn't generate trading edge in biotech stocks if everyone else in the market already knows what you know → "what you know is already baked in" → trading edge must be about information or processing the market doesn't already have.
  4. Working on weaknesses rather than strengths: In many professions, improving weaknesses is valuable; in trading, the payoff is heavily skewed to concentrated strengths → find what you do better than the market and maximize it, rather than diversifying into mediocrity.

Edges

Conventional Wisdom Is Wrong

Discipline Without Edge Is A Loss Machine

The trading world treats discipline and process as the primary success determinants. In reality, discipline is an execution multiplier — it amplifies whatever edge (or non-edge) you have. A disciplined trader with no edge loses more consistently than an undisciplined one because they execute their non-edge strategy more efficiently.

What most people do
When losing consistently, diagnose the problem as a discipline or psychology failure. Work on emotional control, journaling, and rule-following.
What the best do
First ask: can I quantify a positive expected value for this strategy? If not, stop trading it regardless of discipline. Discipline is the final step after edge is confirmed, not the first step.
Why it's an edge: Reframes the entire diagnostic process. Losses are not psychology problems — they are edge problems until proven otherwise. This prevents years of "improving discipline" on a fundamentally non-edge strategy.
How to exploit: Before any discipline work, articulate the specific, numerical source of positive expected value. If you cannot state expected value as a number with a confidence interval, you don't have a diagnosable strategy.
Euan Sinclair, "Edge is in the Numbers," Futures Radio Show, 2020-09-25
💎 Elite-Only Behavior

Press The Edge Hard While It Lasts

Common risk management advice is to keep positions small, diversify broadly, and never bet too aggressively. Expert edge holders do the opposite: when they have confirmed edge, they size it as large as Kelly allows and run it aggressively — because edges decay, and the window is finite.

What most people do
Bet conservatively even with confirmed edges, treating small sizing as universally prudent risk management.
What the best do
Differentiate between "protecting against uncertainty about having edge" (bet small) and "confirmed edge with documented decay timeline" (bet large while it lasts, then exit). Maximize edge exploitation before crowding or regime change erodes it.
Why it's an edge: Most practitioners leave significant P&L on the table by under-sizing genuine edges out of generalized risk aversion.
How to exploit: For each confirmed edge, estimate the expected decay timeline (based on crowding dynamics and economic rationale persistence). Front-load the position sizing while the edge is fresh; systematically reduce as crowding indicators rise.
Euan Sinclair, "Edge is in the Numbers," Futures Radio Show, 2020-09-25; Flirting with Models S3E12, 2021-04-10
Conventional Wisdom Is Wrong

Domain Expertise Is Not Trading Edge

The assumption that knowing more about an industry translates to trading edge is incorrect. Trading edge requires information the market doesn't already have, or a better way to process information the market has. If your domain knowledge is widely shared among market participants in that sector, it is already priced in. The expert knows things, but the market knows them too.

What most people do
Assume that expertise in biotech, technology, or commodities generates trading edge in those sectors. Build strategies based on industry-specific insight.
What the best do
Evaluate not just what they know, but whether the market knows it too. Test whether their domain expertise actually generates systematic positive returns vs. the benchmark, with proper controls for factor exposures. If it doesn't, the expertise is not a trading edge.
Why it's an edge: Prevents the common trap of sophisticated knowledge with no actual alpha — and forces rigorous testing of whether "knowing the industry" translates to predictive returns.
How to exploit: For any domain-expertise-based strategy, test it out-of-sample against a simple factor benchmark. If it does not outperform on a risk-adjusted basis, the domain knowledge is not generating trading edge regardless of its quality.
Euan Sinclair, Positional Option Trading, Flirting with Models S3E12, 2021-04-10
🔑 Hidden Causal Lever

Count Independent Regime Observations, Not Calendar Days

A 4-year vol strategy backtest with daily data may contain only 5-10 independent vol regime observations, massively overstating statistical confidence. Practitioners count data points when they should count regime transitions. 1,000 daily data points during 3 regime states gives you n=3, not n=1,000.

What most people do
Report statistical significance based on daily data points. "We have 4 years of daily data = 1,000 observations."
What the best do
Count the number of independent regime transitions in the sample. If your strategy depends on regime behavior, your effective sample size is the number of regimes, not the number of days.
Why it's an edge: Most backtests dramatically overstate confidence in regime-dependent strategies. The practitioner who correctly counts degrees of freedom avoids deploying strategies that look significant but aren't.
How to exploit: For any regime-dependent strategy, identify regime transitions in the backtest period. If there are fewer than 15-20 independent regime observations, the statistical evidence is insufficient regardless of daily Sharpe.
From Common Errors #2 and Progression Level 3

Sources

  • Euan Sinclair, "Edge is in the Numbers," Futures Radio Show (2020-09-25) — foundational framework: risk premium vs. inefficiency, Kelly criterion, statistics as the basis for every trade
  • Euan Sinclair, Positional Option Trading, Flirting with Models S3E12 (2021-04-10) — phenomenon-first strategy development, market efficiency as organizing principle
  • Euan Sinclair, "0DTE The Trap," Sharpe Two Podcast (2026-02-26) — when to change your mind vs. when to maintain conviction through underperformance
  • Euan Sinclair, Options Trading Course Overview, Money Talks (2023-03-17) — practical framework for teaching edge identification to traders