The psychology of size

It is 10:05 ET (22:05 ICT, winter). The method is validated, the plan is written, the journal is honest. Then the trader doubles size and suddenly cannot execute: entries hesitate, exits panic, the plan might as well be in another language. Nothing about the market changed. The size changed the trader. Often the size changed more than they think, because volatility rose the same week. This last lesson of Module 7 is about the final variable: you, at size, measured in the unit the market moves in.

Key terms

Dollars per σ (số đô la trên mỗi σ): what one normal day’s move is worth on your position: price × implied volatility ÷ √252 × dollars per point × contracts Definition. For NQ the volatility input is VXN, the Nasdaq-100 volatility index (M8.05).

σ, the one-day implied move (biên độ kỳ vọng một ngày): price × IV/√252, a ruler for how big a normal day is today Definition. It is a scale, not a forecast of direction or of where price turns. Only the ±0.5σ and ±1σ bands are used in this course.

Sizing ladder (thang tăng khối lượng): the rule for when you may add one contract step, and when you step back down House rule.

Size is dollars per σ, not contracts

A contract count is not a size. Two MNQ on a quiet day and two MNQ on a volatile day are different positions, because every swing you watch is measured in points × $2 × 2, and the points grow with volatility.

Worked through with a synthetic NQ at 20,000 (MNQ is $2 a point, NQ $20; tick 0.25) Definition:

Your stop still caps a planned loss (M6.09 owns that formula). What changes is everything around it: on a VXN 30 day your structure stops sit further away, normal pullbacks are bigger in dollars, and unrealised P&L swings harder while you wait. If you keep the contracts and widen the stop, your dollar risk rises; if you keep the contracts and keep the old stop, the stop sits inside the day’s noise. The consistent rule is the one M6.09 already gives: hold dollar risk fixed and let the contract count fall as σ rises House rule.

Do this once before each session: write today’s half-σ in dollars for the size you plan to trade. If that number makes your stomach tighten, the size is too big for today, whatever the formula allows.

What the data says

Read it as two facts. The ruler is honest enough to price your size with MeasuredNasdaq index. And it is not a ceiling: on more than a third of days the regular session touches ±1σ somewhere. Size so that a full σ against a mistake (a missed stop, a fill through a fast market) is survivable, not just the half-σ you expect to watch.

Why size changes the trader

Stakes change performance. In a set of experiments, people offered very large rewards (about five months of their normal spending) did worse on most tasks that needed concentration and memory than people offered small rewards MeasuredOther market. That was measured on cognitive and motor tasks, not on traders; that a bigger size can degrade your execution in the same way is plausible, not measured PlausibleOther market. The losses-loom-larger side of this, loss aversion, is in M7.05.

The symptoms are your measurement. You cannot see the stakes effect directly, but your journal can: hesitating on valid setups (M7.06’s fear), moving stops (M7.05’s loss aversion), checking P&L mid-trade instead of structure. Tag them. If the tags rise after a size step, the size is wrong for you right now, regardless of what the formula allows House rule.

Size up one earned step at a time

The ladder House rule: validate at micro size (M7.10) → trade 2–3 months with logged positive expectancy → increase one step (for example 1 → 2 MNQ) → re-measure for a month → repeat. Each step re-tests whether you hold up under the new dollars per σ. Drop back down at the first sign the symptoms return. There is no shame in that; the shame is in blowing up at a size you had not earned. These numbers are policy, not a measured optimum: they exist so that each step is judged on enough trades (M7.07, M7.10) and so that the account survives the step that fails.

Price each step in dollars per σ, not contracts. A step from 1 to 2 MNQ in a week when VXN rose from 18 to 26 is not a doubling: half-σ goes from about $227 on one MNQ to about $655 on two, nearly three times the swing.

Environment and sleep

The environment (môi trường): make the plan the path of least resistance House rule. A dedicated desk, the phone in another room during the session, one fixed layout, the plan taped to the monitor, the sim account one click away for “let me try something” urges: curiosity gets a sandbox, never the live account. The goal is not more discipline; it is needing less of it.

Sleep is a trading input. NQ’s regular session is 09:30–16:00 ET, which is 21:30–04:00 ICT in winter and 20:30–03:00 ICT in summer: your night. In a laboratory study, people kept to 6 hours of sleep a night for two weeks ended up as impaired on attention tests as after one to two nights without sleep, while rating themselves only slightly sleepy MeasuredOther market. The part that matters for you is the second half: you do not feel the deficit accurately. Transfer to trading decisions is plausible, not measured PlausibleOther market. Log hours slept as a journal field and compare deviation tags on short-sleep days with the rest; that comparison on your own data is the only measurement you will get.

The identity shift (chuyển đổi tư duy): beginners try to make money (outcome); professionals execute the plan (process) House rule. A working test: when a red day with full adherence genuinely feels like a good day (M7.03), the identity has moved. Until then, keep score on adherence; identity follows the scoreboard.

Worked example: a size step that went wrong, and the fix

Synthetic trader and numbers, for illustration. The trader validates at 1 MNQ for 3 months (about 120 logged trades, +0.31R per trade after costs, adherence 90%). They step to 2 MNQ, $40 risk per trade on a $5,500 account. That week VXN has climbed from 18 to 26: their half-σ goes from about $227 to about $655.

Week one: they hesitate on two valid setups and exit one winner early. Symptoms. Instead of “pushing through”, they drop back to 1 MNQ for two weeks and journal the hesitation triggers: both were setups right after a loss (M7.06’s fear, at the new size). Then they re-step to 2 MNQ with two written rules: the first trade of each session at 1 MNQ, size up only after a clean first execution; and on days when today’s half-σ on 2 MNQ is above $600, trade 1 MNQ. The workaround holds; a month later 2 MNQ is the new normal. The trader did not conquer fear. They sized around it until the new size felt like the old one.

Prop-firm note. In an evaluation or funded account, the firm’s position cap and daily loss limit are hard ceilings, and some firms scale the cap with the account balance. Your ladder lives under those ceilings, never above them. Rules change, so check the firm’s help center for current numbers; the Prop Firm course covers them.

Common mistakes

Common mistake: Sizing up because the account grew (“I’m at $8k now, I can trade 2 NQ”). Account balance is not a sizing credential; logged expectancy at the current size is. An account that grows while execution stays unmeasured is how traders arrive at size with money they have not learned to keep.

Common mistake: Counting size in contracts. “Still 2 MNQ” on a VXN 30 day is about 1.5 times the position it was at VXN 20. Price the size in dollars per σ every session.

Common mistake: Reading “about three days in four close inside 1σ” as “a stop at 1σ is safe three times in four”. That number is close-to-close on the index; the session high or low reaches ±1σ on more than a third of days, and the ruler says nothing about direction.

Drill: price your size in σ for ten sessions

1. Before each of your next ten sessions, write the prior-close VXN, the NQ price and your planned contracts.
2. Compute the 1σ day in points (price × VXN/100 ÷ 15.87) and your half-σ in dollars.
3. After each session, add your symptom tags (hesitation, moved stop, P&L-watching, early exit) to the journal row.
4. Sort the ten sessions by half-σ dollars and compare the symptom count in the top five with the bottom five.

Graduation checklist

Module 7 is complete when all of these are true House rule:

• One-page plan written, versioned, taped to the monitor (M7.04, M7.09)
• 50+ sim trades logged with expectancy computed (M6.08, M7.03)
• Adherence ≥ 85% over the last 20 trades (M7.03)
• One full drawdown weathered without rule-breaking (M7.08)
• Today’s half-σ in dollars written before each session (this lesson)
• Can narrate any session in three questions (M1.07)

Recap

• Size is dollars per σ, not contracts: the same contracts are a bigger position when VXN rises. Hold dollar risk fixed and let contracts fall.
• The VXN/√252 ruler is measured on the Nasdaq-100 index as slightly wide, with about a quarter of days closing beyond 1σ; NQ is extrapolated.
• Hesitation, stop-moving and P&L-watching are your journal’s measure of oversize; the size is wrong for you right now, whatever the formula says.
• Size up one step at a time, priced in dollars per σ, and step back at the first symptom. The ladder is a house rule, not a measured optimum.
• Environment beats willpower, and sleep loss is hard to feel: log it. Both are inputs to execution.

Further reading

• Large Stakes and Big Mistakes (Ariely, Gneezy, Loewenstein & Mazar, 2009): experiments in which very high rewards lowered performance on most tasks.
• The Cumulative Cost of Additional Wakefulness (Van Dongen et al., 2003): two weeks of restricted sleep, cumulative deficits, and why people do not feel them.
• Prospect Theory (Kahneman & Tversky, 1979): the value function behind loss aversion and the shrinking feel of each extra dollar.
• Our lab: on the Nasdaq-100 index, price × VXN/√252 slightly overstates a normal day, about a quarter of days close beyond ±1σ and more than a third touch it during the session; on NQ futures this is extrapolated, not measured.

Connects to: M7.11 (the review that measures the psychology) → M8.01 (macro: the weather around your now-systematic trading). M6.09 owns the sizing formula; M8.05 introduces VXN.

Educational content only. Not financial advice. Futures trading involves substantial risk of loss.