A career ladder gated by measurements

It is Friday, 16:05 ET (04:05 ICT Saturday, winter). A learner closes the replay window after three weeks on a sim account. Twenty-two trades, two NQ contracts each, +$3,100 on the screen. The checkout page for a prop-firm evaluation is open in another tab. Three weeks feels like long enough, and the number is green.

Nothing in that scene measures skill. Three weeks is a date, $3,100 is a sum, and 22 trades is a sample a coin could produce. This lesson replaces “how long have I done this?” with “what does my journal show?”, in five gates.

Key terms

Career ladder (lộ trình nghề nghiệp): the order of rungs from practice to risking real money, and later the firm’s money House rule. Ours has five: sim, micros, one mini, evaluation, funded.

Gate (cột mốc mở khóa): a written test you pass before you climb a rung House rule. It names a metric and a threshold, and it opens on your journal, never on the calendar.

Micro contract (hợp đồng micro): a futures contract one tenth the size of the E-mini Definition. MNQ moves $0.50 per 0.25-point tick ($2 a point); NQ moves $5 a tick ($20 a point). MES is $1.25 a tick, ES $12.50.

Sim account (tài khoản mô phỏng): an account that fills your orders against live or replayed prices with no money at risk Definition.

Terms you need: Expectancy is your average result per trade in R, where 1R is the planned risk to your stop (loss exit order; RSK1.01). A 95% confidence interval is average ± 1.96 × σ ÷ √n, where σ is the standard deviation of single-trade outcomes in R (EV1.01; EV3.01 owns the sample-size formula). Adherence is the share of trades in your journal (trade log) that followed every written rule (DESK1.01). Drawdown is the drop from a peak to a low in cumulative R, slippage is worse fills than planned, and a consistency rule is a firm rule capping how much of your total profit can come from one day.

Chart first: two ways to climb

How it works

Each rung changes one thing

A good ladder changes one variable per rung, so that when results get worse you know why. The specific numbers below (100 trades, 200 trades, 60 sessions, 85% adherence) are academy house rules House rule chosen so a sample spans multiple weeks and regimes rather than one lucky streak:

1. Sim, micro size (Gate 1 → live micros). Metric: at least 100 planned micro-sim trades with ≥ 85% adherence. Tests your process: do you write a plan, place a stop, log the trade?
2. Live micros (Gate 2 → one mini). Metric: at least 200 live micro trades across ≥ 60 sessions, ≥ 85% adherence, drawdown within your written limit, and the 95% confidence interval of net expectancy above 0R. Tests real fills and real nerves where measuring is cheapest.
3. One mini (Gate 3 → evaluation). Metric: at least 100 trades at 1 NQ or 1 ES with ≥ 85% adherence, positive net R, and live slippage logged in ticks against your micro baseline. Tests whether 10× dollar size changes your behaviour or fills.
4. Evaluation (Gate 4 → funded). Metric: pass the evaluation under the firm’s dated rules (loss limit, profit target, and any consistency rule; PROP course) while keeping ≥ 85% adherence.
5. Funded (ongoing gate). Metric: keep the same adherence and drawdown gates running on every rolling 50 trades. A funded account is a new sample, not a graduation.

Why micro size in sim, even though sim is free

The house rule for path A is micro-only sim through the Practitioner level House rule. Sim at 2 NQ teaches numbers you cannot afford live: $40 a point on a 15-point stop is $600 of risk per trade. Practise at the size you will first trade live, and your R, your stops and your slippage carry straight over. The sizing formula itself belongs to RSK1.05.

Why measurements and not time

Time measures nothing about you: in three months one learner logs 300 planned trades, another 40 bored ones. The gate asks for what time was supposed to stand in for: enough trades, spread over enough sessions, executed as written, with a net result whose interval sits above zero.

The last part is the hard one. Your journal’s average is an estimate, and a small sample swings a lot. EV2.01 showed that the base rate for retail day traders is very low MeasuredOther market, so the honest starting assumption is “no edge until my data shows one”. A gate that only asks for a positive average lets luck through, as the next figure shows.

What the data says

The model’s message for the ladder: the evaluation is a place to use an edge you already measured, not a place to discover whether you have one. Gate 2 sits on live micros because that is where measuring is cheapest.

Worked example: gate 2 on 200 micro trades

Illustrative journal, one MNQ per trade. A typical trade: long at 21,050.00, stop 21,042.00 (8 points, 32 ticks, $16 risk = 1R), target 21,066.00 (2R). Costs (commission and slippage) are set at 0.06R per trade here; put in your own numbers from EV2.02.

After 200 trades over 64 sessions: win rate 41%, average win 1.9R, average loss 1.0R.

• Net expectancy = 0.41 × 1.9 − 0.59 × 1.0 − 0.06 = +0.13R per trade Definition.
• For outcomes that are +1.9R (41% of the time) or −1.0R (59% of the time), the single-trade standard deviation (EV1.01) is (1.9 − (−1.0)) × √(0.41 × 0.59) ≈ 1.43R. Using the 95% interval formula average ± 1.96 × σ ÷ √n from EV1.01, the interval is +0.13 ± 1.96 × 1.43 ÷ √200 = −0.07R to +0.33R (EV3.01 derives the sample-size rule in full).

Four of five pass, and the gate stays shut. If the true edge really is +0.13R, the interval clears zero at about (1.96 × 1.43 / 0.13)² ≈ 465 trades: roughly 155 sessions at three trades a day. That is slow. It is also cheap: each extra month on one MNQ costs a small fraction of one evaluation, and the waiting is what turns “I think I have an edge” into a number.

The mirror and the failure

The mirror: stepping down. A ladder needs a rule for going down as well as up House rule. Ours: if adherence over your last 20 trades drops below 85%, or your drawdown reaches the written limit, drop one rung. Climbing back needs the same gate again. Stepping down is the system working, not a failure.

Run the gate by side too. If your shorts clear zero and your longs do not, the gate is open for shorts only.

The failure: a positive average that means nothing. The figure below takes two trader types from our lab’s model: one with no edge (net −0.05R after costs) and one with a real +0.15R net edge. Each bar counts synthetic traders by their average after 50 or 200 trades.

So “positive after 50 trades” is a gate that four in ten no-edge traders pass by luck. The interval rule is stricter on purpose: in the same simulation, under 1% of no-edge traders cleared it after 200 trades, against 31% of real-edge traders. A real edge often needs more than 200 trades to prove itself.

Do it on the platform

Prop-firm note. Evaluation and funded-account terms are firm rules with dates; the PROP course shows them in dated cards. Firms also publish rules on order handling and simulated fills; check your firm’s help center (PROP1.04 covers the dated rules).

Common mistake: treating a positive average as an edge. “+$3,100 in 22 trades” is a result, not a measurement: with no interval and no baseline, it fits a no-edge trader just as well.

Drill: which rung are you on?

A journal shows 120 live micro trades over 41 sessions: win rate 45%, average win 1.5R, average loss 1.0R, costs 0.05R per trade, single-trade spread about 1.24R, adherence 91%. Which gates are open?

What we don’t know yet

• The gap between sim fills and live fills on your orders is not yet measured by our lab; MKT2.03 explains why sim tends to fill more kindly, and gate 3 measures your own gap.
• Whether gated ladders improve traders’ outcomes is not yet measured. The gates are a house rule, chosen because each one is checkable.

Recap

• Five rungs: sim at micro size, live micros, one mini, evaluation, funded. Each rung changes one thing.
• Each gate opens on a journal metric, never on time spent.
• Gate 2 asks for an interval above zero, because a positive average after 50 trades is common without any edge.
• A step-down rule is part of the ladder.

Further reading

• Barber, Lee, Liu and Odean (2014), “The cross-section of speculator skill: Evidence from day trading”, Journal of Financial Markets 18. The base rate behind EV2.01.
• CME Group, Micro E-mini Nasdaq-100 and Micro E-mini S&P 500 contract specifications (public): tick sizes and values.
• Our lab: a model of the firm’s published evaluation rules, in which a zero-edge trader loses per attempt once costs are charged (an upper bound, not a measurement of any trader).

Prerequisites: EV2.01 (base rates) and EV2.02 (your P&L statement). Next: EV2.05 applies the same “measure before you pay” rule to the courses, desks and chat rooms you might learn from.

Educational content only. Not financial advice. Futures trading involves substantial risk of loss. Not affiliated with Topstep; firm rules change; check the Topstep help center.