Pairs Trading ES vs NQ: How a Spread Trade Works

A pairs trade asks whether two related instruments have moved apart more than usual, and whether that gap tends to close. Direction matters less than the spread. With ES and NQ, you'd be long one and short the other, sized so the position is roughly neutral to the overall market. It sounds tidy. It also has two legs, two sets of costs, and a relationship that can change without warning.

Why ES and NQ Are the Obvious Candidates

The S&P 500 and Nasdaq 100 share many large holdings, both contracts trade on CME Globex with deep liquidity, and they tend to move in the same direction on most days. NQ usually moves more in percentage terms than ES, so a day where NQ rallies much harder or falls much harder than its usual ratio to ES is the kind of gap a pairs trader watches. None of that guarantees the gap closes. The two indexes differ in composition and can diverge for as long as a sector theme persists.

Correlation Versus Cointegration

Correlation says two series tend to move in the same direction. It doesn't say the distance between them will shrink. Cointegration is a stronger idea: each series can wander, but a particular combination of them (the spread) stays within a range and keeps returning to a long-run average. Pairs trading depends on that second property. Statistical tests such as Engle-Granger check for it on a window of data, and the results move as the window moves. A pair that passes in one period can fail in the next, so a passed test isn't a permanent license. Any parameters, such as a 60 or 90 day lookback, are choices you have to test, and I wouldn't copy numbers from an article without checking them on data you control.

The Spread and the Z-Score

A common approach is to compute a spread, for example the log of NQ minus a hedge ratio times the log of ES, then standardize it: subtract a rolling mean and divide by a rolling standard deviation. That number is the z-score. A reading of +2 says the spread is two standard deviations above its recent average. A simple rule might enter when the z-score passes 2 and exit when it returns near 0.5. A stop might sit at 3.5. Those thresholds are examples, not recommendations. Each has a trade-off: tighter entries mean more trades and more noise, wider entries mean fewer trades and fewer samples to judge them on.

A Hypothetical Dollar-Neutral Calculation

Contract counts alone don't tell you exposure, because ES is $50 per point and NQ is $20 per point. Use the notional value: index level times multiplier. Suppose, hypothetically, ES is at 5,600 and NQ is at 21,000.

Three ES against two NQ is about dollar-neutral at those prices. Prices change, so the ratio has to be recalculated. This is a hypothetical illustration, not a recommended position. With micros (MES at $5 per point and MNQ at $2 per point) you can size in smaller steps.

Costs and Execution Problems

• Slippage on every leg. One tick of slippage is $12.50 per ES contract and $5 per NQ contract. On 3 ES and 2 NQ, one tick on every leg is 3 x 12.50 plus 2 x 5, or $47.50. An entry and an exit with one tick each on all four fills is $190, before commissions.
• Leg risk. The two orders can fill at different times and prices. If one leg fills and the other doesn't, you're holding a directional position you didn't plan.
• Margin. A hedged pair still needs margin on both legs. Margin is collateral, not a cap on loss.
• Stops slip. A stop on the spread has to be implemented as orders on each leg, and either can slip in a fast market.

When the Trade Breaks

• Sector shocks that reprice tech against the broad index for weeks, so the spread doesn't revert on your timeline.
• Crisis conditions where both fall together and the spread behaves differently from its recent history.
• A relationship that changed because the index composition or volatility ratio changed, which makes your hedge ratio stale.
• Too little history. A handful of trades isn't evidence the spread reverts reliably.

Testing and Rule Checks

1. Pull both series on the same sessions and bar timing, and watch for contract roll dates, which aren't universal. Use consistent front-month handling.
2. Compute the spread and test the rule on one period, then on a separate period you didn't tune on.
3. Include commissions and at least a tick of slippage per leg per side.
4. Run it forward on SIM, remembering that simulated fills aren't a validation of exchange queue position or slippage.

If you plan to run this on a prop account, read the firm's rules first. Hedging between ES and NQ, trading two legs at once, and any automation of the legs are all subject to firm policy, and rules differ by firm and account type. Apex, for example, explicitly prohibits automated or algorithmic preset bots that place, modify or cancel trades without personal real-time review (see the Apex prohibited activities page).

Related Reading

Spread trading is mean reversion applied to a gap between two prices. The mean reversion guide covers the single-instrument version, and ES vs NQ covers how the contracts differ. The edge calculator helps you see how costs change expectancy.