Apex trailing drawdown: $3,000 of room at any peak — never more ($100k)

Ask traders who failed an Apex evaluation what ended it and most will say "one bad day." The account statements usually say something more precise: the trailing threshold caught up with a retracement — often on an account that was comfortably in profit minutes earlier. This page describes the mechanism, the arithmetic that follows from it, how it differs from the other two floor types, and how to test your own statistics against the rule before the fee leaves your card.

Scope: this page describes Apex Trader Funding's Intraday Trailing Drawdown evaluation, as transcribed in our rule corpus and dated below. Apex also publishes a separate end-of-day trailing product, which behaves differently and is modeled separately in the engine. For the general mechanic — independent of any firm — see Trailing drawdown, explained, or test the trailing floor against your own numbers in the free trailing-drawdown calculator.

What the rule actually does

The evaluation uses a trailing drawdown computed on intraday equity peaks, including unrealized profit. Two properties make it the most frequently-adjusting floor in our rule corpus — it is the only variant we transcribe whose threshold can move on any tick rather than once a day:

• It ratchets on peaks, not closes. If an open position spikes your equity to a new high and then retraces before you exit, the elimination line has already moved up — permanently. You never banked that profit; the threshold doesn't care.
• It only moves up. Every new peak drags the floor behind you at a fixed dollar distance. There is no reset and no downward move.

A third property — whether the threshold ever stops trailing inside an evaluation — is the one thing Apex's own documentation does not say consistently, and our corpus records it as unresolved rather than resolving it on the firm's behalf. The section below sets out both readings and which one the engine models.

A worked example

Say your trailing threshold is $3,000 on a $100,000 evaluation:

1. You start at $100,000. The elimination line sits at $97,000.
2. A strong open position pushes your equity to an intraday peak of $102,400 — unrealized. The line is now $99,400.
3. The trade retraces and you exit at $100,900. You're up $900 on the evaluation. The line stays at $99,400.
4. The next morning a normal two-loss sequence takes equity to $99,350. Evaluation over — on an account that never closed a losing day badly and was net positive throughout.

Under a static floor at the same $3,000 distance, this sequence never comes within $2,350 of failure. Floor mechanics — not strategy quality — decided the outcome.

Now express the same day in units of risk. At $1,000 risk per trade, you started the evaluation exactly three average losses from elimination. After the day described above — a day you finished up $900 — your room to the threshold was $1,500: one and a half losses. You made money and became easier to kill. Static floors cannot produce that sentence, and end-of-day trailing floors can't either: only an intraday trailing floor can turn a green day into lost ground.

Notice what the rule is actually reading. Two traders can post identical closed-trade statistics — same win rate, same average R, same daily P&L — and die on different days, because the threshold is computed from the path equity travels between entry and exit, not from what gets banked. A style that lets winners breathe — wide targets, holding through data releases, scaling in — shows the rule its highest excursion on every trade. A style that banks quickly shows it almost nothing. Under intraday trailing, trade management isn't adjacent to your risk rule; it's an input to it.

The rule as arithmetic

Write the rule out and the whole evaluation collapses into one expression. Your room to elimination is:

Peak minus equity is simply your current giveback from the high — so read the formula as: a fixed $3,000 allowance, minus whatever you've handed back. Banked profit appears nowhere in it. At your peak you have exactly $3,000 of room — never more, no matter how much you're up: profit raises the peak, the peak drags the floor, and the ceiling on your safety stays fixed. Under this rule you cannot buy safety; you can only avoid spending it. And the evaluation has exactly two exits — your equity reaches the profit target, or it touches the floor. There is no waiting it out: the floor never descends, so flat time spends calendar without buying back an inch of room.

On this account size the two exits sit at $106,000 (the 6% profit target) and at whatever the trailing threshold currently reads. The threshold distance and the target both vary by account size, and the size ladder is not a straight-line scale — the engine loads the figures for the size you pick rather than interpolating them, and we do not print a ladder here.

The one thing the firm's documentation does not settle

Traders routinely repeat that the Apex threshold "freezes at your starting balance." Our corpus does not record that as an evaluation rule, and the engine does not model it. What the firm's pages actually contain is three separate statements that do not reduce to one:

• Pages covering Tradovate evaluations describe a threshold that keeps trailing the peak for the whole evaluation, with no lock.
• Pages covering Rithmic and WealthCharts evaluations describe a freeze once the threshold reaches the profit-target balance — which is also the balance at which the evaluation ends. No path inside an evaluation can tell the two readings apart, because reaching that point ends the account either way.
• The freeze at $100 above the starting balance — the version most often quoted secondhand — is documented as a property of the funded Performance Account, not of the evaluation.

So the corpus leaves the field unresolved and the engine takes the conservative branch: the threshold trails for the entire evaluation. That is the reading that produces the higher ruin estimate, which is the direction an honest model should err in when a firm's own documentation is ambiguous. The engine labels this field as unresolved in the Apex preset's modeling notes rather than hiding the choice.

The practical translation: the evaluation is a race to the profit target run on a fixed, non-renewable allowance. Every dollar of giveback from a peak is spent permanently, and nothing you earn afterwards refills it. The same structure appears wherever a floor trails a high-water mark — our risk-of-ruin analysis works through what a non-renewable allowance does to ruin probability in general.

The rule also re-prices a habit most traders consider free: letting a winner run past the plan and giving some back. In the worked example, the $1,500 given back from the $102,400 peak wasn't just unrealized profit evaporating — it was survival room, permanently spent. Under this rule an un-banked winner is not neutral; it draws on the same buffer a losing trade does.

Three species of floor — and why the difference is not cosmetic

Drawdown floors come in three species, and traders routinely price them as if they were one. The difference is structural and follows from the rule text alone:

Exhibit — the three drawdown species

SCROLL →

Rows name specific evaluation products, not firms: several firms publish more than one rule set, and Apex publishes both an intraday and an end-of-day trailing product. Which product uses which mechanic is transcribed from each firm's own published documentation and dated on the rule changelog. One further difference the table cannot hold: in each of the end-of-day rule sets above, the firm's documentation states the floor stops trailing once it reaches the starting balance (or just above it) — the point Apex's intraday documentation leaves unresolved. The ordering is by how often the floor can move, not by how hard a rule set is to pass; that depends entirely on the trader, and the engine simulates yours.

Why size matters more under this floor than under a static one

This follows from the mechanic, not from a table. Under a static floor, position size affects only how fast your equity falls toward a line that never moves. Under an intraday trailing floor, size does that and feeds the mechanic that raises the line: bigger positions produce bigger unrealized swings, and every unrealized swing that makes a new high ratchets the floor up behind you, permanently. Size therefore appears twice in the arithmetic of an intraday-trailing account and once in a static one.

That is why the sizing question has no universal answer and why we do not print one. The magnitude depends on your win rate, your average win, your trade frequency and your variance — the engine sweeps risk per trade across the range and shows the whole curve for your own statistics, simulated in your browser.

What changes the odds under this mechanic

These are properties of the rule, not recommendations. Each one is a statement about how the mechanic responds to an input; what any of them is worth for a given trader is a number the engine simulates from that trader's own statistics.

• Risk per trade. Because the allowance is fixed at the threshold distance and never refills, risk per trade sets how many independent givebacks the allowance can absorb before it is gone. Sizes that change partway through an evaluation are a configuration the engine can model, not a plan it endorses.
• Unrealized givebacks. Under an intraday trailing floor a giveback from an intra-trade peak is survival room destroyed permanently, dollar for dollar, whether or not the trade closes green. Exit behaviour therefore affects the floor here in a way it does not under a static or end-of-day floor.
• Room expressed in losses rather than dollars. Room ÷ average loss states the same fact as a dollar figure in the unit the mechanic actually consumes. The two are arithmetically identical; the ratio simply makes the remaining count legible.
• Knowing the number before the fee. A simulated pass probability under this rule is derivable from win rate, average R and risk size. The engine simulates yours, locally, before the fee leaves your card. It is a model output under the model's assumptions, not a forecast of your account; what tolerance to hold it to is yours to set — we do not publish one.

Questions traders actually ask

Companion reads: the same mechanic with no firm attached — Trailing drawdown, explained — and the other mechanism that ends evaluations after the hard part is done, Topstep's 50% consistency rule, explained with the math.

PropSurvival is independent analytical software and is not affiliated with, endorsed by, or sponsored by Apex Trader Funding. "Apex Trader Funding" is used here only to identify the rule set being described. Rule descriptions are transcribed from the firm's published documentation on the date shown above; firms change rules without notice, and the firm's own documentation is always the final authority. Nothing on this page is investment advice, trading advice, or a recommendation to purchase any evaluation program. Every figure the engine produces is a simulation over user-supplied inputs and transcribed rules, under the assumptions stated in the methodology — not a prediction of any actual account's outcome.