Evaluation difficulty is a quotient. The profit target is the distance you must travel. The max drawdown is the room you have to be wrong on the way up. Divide the first by the second and you get one dimensionless number that ranks any two offers on that single dimension, at any account size, from any firm. Firms rarely print it. Their own rules page gives you both inputs.
Prop firm evaluation difficulty is set by the ratio of profit target to max drawdown, not by either number alone. An illustrative $3,000 target with a $2,000 drawdown gives a ratio of 1.50, while a $6,000 target with a $3,000 drawdown gives 2.00, so the second account is the harder product despite a 50% larger dollar buffer. Compute target divided by drawdown, then daily loss limit divided by drawdown, then confirm whether the drawdown is static or trailing, and only then look at the fee.
What does the target to drawdown ratio actually measure?
Call it R, where R = profit target / max drawdown. R answers one question: how many full buffer widths of net progress does this evaluation demand? At R = 1.00 you must net exactly one buffer's worth of profit to pass. At R = 2.00, two. The buffer is the only unit the rules enforce. Dollars are the unit the marketing page enforces.
R is size invariant, which is the whole point. A $150,000 nominal product with a $9,000 target and a $6,000 max drawdown computes to 9,000 / 6,000 = 1.50, identical to a small account with a $3,000 target and a $2,000 drawdown (3,000 / 2,000 = 1.50). The headline dollars differ by 3x. Difficulty on this measure does not move. Account size leads most marketing, and the ratio deletes it.
The CFTC's fraud advisory on commodity trading systems sold online makes the same point about simulated track records: hypothetical results "may not adequately take into account the ability of a trader to absorb trading losses or to meet margin calls". Capacity to absorb loss is exactly what the drawdown buffer measures.
Every account figure in this article is illustrative, not any firm's current terms. Firms change targets, drawdowns, drawdown type, daily caps, fees and splits without notice. Compute your own numbers from the written rules document rather than a comparison table, and recompute at purchase time, because promotional terms and standard terms frequently differ.
Three accounts, ranked exactly
Account A: target $3,000, max drawdown $2,000, R = 3,000 / 2,000 = 1.50. Account B: target $6,000, drawdown $3,000, R = 6,000 / 3,000 = 2.00. Account C: target $2,500, drawdown $2,500, R = 2,500 / 2,500 = 1.00. Easiest to hardest: C at 1.00, A at 1.50, B at 2.00. Account B carries the largest dollar buffer and the hardest ratio.
Invert it and the asymmetry gets tangible. Buffer as a percentage of required progress: A = 2,000 / 3,000 = 66.67%, B = 3,000 / 6,000 = 50.00%, C = 2,500 / 2,500 = 100.00%. Account C hands you a dollar of room for every dollar of progress it demands. Account B hands you fifty cents.
The dollar figures tell you how big the account is. Only the ratio tells you how hard it is.
Draw it and it stops being arithmetic. Put the starting balance on a horizontal line. The buffer is a band hanging below that line, its depth equal to the max drawdown. The target is a second line above the start. Now measure the gap from start to target in buffer depths instead of dollars and stack those depths upward: Account A's target sits one and a half buffer depths above the start, Account B's sits two full depths, Account C's sits exactly one. Account B's band is physically the deepest in dollars, and Account B still has the tallest stack to climb. The stack is the difficulty.
What does a higher ratio cost you in trades?
Normalize risk to the buffer and the ratio shows up as trade count. Risk 10% of the max drawdown per trade at 2:1 reward to risk. Account A: risk unit 0.10 x 2,000 = $200, winner $400. Account B: 0.10 x 3,000 = $300, winner $600. Account C: 0.10 x 2,500 = $250, winner $500. Consecutive losses tolerated before failure: A = 2,000 / 200 = 10, B = 3,000 / 300 = 10, C = 2,500 / 250 = 10, identical by construction. Winning trades needed with no losers: A = 3,000 / 400 = 7.5 so 8, B = 6,000 / 600 = 10, C = 2,500 / 500 = 5. Unrounded, those counts are exactly 5R.
Add a realistic win rate and the gap widens. At p = 0.40 with the same 2:1 payoff, expected net progress per trade in risk units is 2p - (1 - p) = 3p - 1 = 0.20. Required progress is target / risk unit = 10R risk units, so expected trades to finish is T = 10R / 0.20 = 50R. Account A needs about 75 trades, Account B about 100, Account C about 50. Account B demands exactly twice Account C's trade count at the same win rate and risk discipline. Break even for this payoff sits at 3p - 1 = 0, that is p = 33.33%, before commissions and slippage.
Those counts assume fixed fractional risk, a constant payoff multiple and independent outcomes. Real trading violates all three, so read them as structure, not forecast. The consequence stands: a higher ratio buys you more trades, and more trades means more exposure to the daily cap, more commission and more chances to trip a consistency rule. If you are still deciding what fraction of the buffer to put at risk per position, the arithmetic in position sizing on funded accounts is the other half of this calculation.
Translate the buffer into instrument terms. CME's E-mini S&P 500 (ES) is $50 per index point with a 0.25 point tick worth $12.50; the Micro E-mini (MES) is $5 per point with a $1.25 tick. Risking $200, 10% of Account A's buffer, is 200 / 50 = 4.00 index points on one ES contract, or 16 ticks of stop distance. On Account B's buffer, $300 is 6.00 points or 24 ticks. A $1,000 daily loss limit is 20 index points of adverse move on a single ES contract; a $500 limit is 10 points. On MES the same $200 buys 40 points on one contract, or 4 points across ten. Contract specifications and performance bonds are set by the exchange and do change, so verify current figures with CME Group or your broker.
How much of your buffer can you actually reach in one day?
The daily loss limit decides how much of the total drawdown is reachable in a single session, so effective per session room is min(daily loss limit, remaining total drawdown). A generous total buffer paired with a tight daily cap is less usable than the headline suggests: on most days the cap binds and the total never enters the calculation.
A $3,000 total drawdown with a $1,000 daily loss limit needs at least 3,000 / 1,000 = 3 maximum loss days to exhaust, which reads as protection. Read the same two numbers the other way: one maximum loss day consumes 1,000 / 3,000 = 33.33% of the entire buffer, two consume 2,000 / 3,000 = 66.67%, and you have $1,000 of room left for the whole rest of the evaluation. Identical arithmetic, opposite emotional register.
| Account (illustrative) | Profit target | Max drawdown | Daily loss limit | Target / DD | Daily / DD | Max loss days to fail |
|---|---|---|---|---|---|---|
| A | $3,000 | $2,000 | $1,000 | 1.50 | 50.0% | 2 |
| B | $6,000 | $3,000 | $1,000 | 2.00 | 33.3% | 3 |
| C | $2,500 | $2,500 | $500 | 1.00 | 20.0% | 5 |
| D | $4,000 | $4,000 | $500 | 1.00 | 12.5% | 8 |
C and D tie at R = 1.00 and are not the same product. C lets a single session reach 20.0% of its own buffer; D lets it reach 12.5%. Among equal ratios, daily limit divided by drawdown is the tiebreaker, and higher is more usable for anyone whose losses arrive in clusters.
Test the cap against your worst day, not your average one. Take a strategy whose worst 5% of sessions lose $900. Against a $500 daily cap it breaches on any such day regardless of edge. Across 20 sessions, assuming independence between sessions (they are not fully independent, but the direction holds), the probability of at least one such day is 1 - 0.95^20 = 1 - 0.35849 = 0.64151, roughly 64.2%. Against a $1,000 cap that identical day survives and consumes 900 / 3,000 = 30.0% of a $3,000 buffer. The total drawdown appeared in neither calculation.
A strategy whose bad day loses $900 is structurally incompatible with a $500 daily cap no matter how good its edge is. The total drawdown cannot rescue a day the daily rule already ended.
Here is the non obvious result. Assume your average winning day nets half your daily loss limit. Days to pass = target / (0.5 x daily limit). Days to fail = drawdown / daily limit. Their ratio is 2 x target / drawdown = 2R, and the daily limit cancels out entirely. Check it: Account A, 3,000 / 500 = 6 days to pass against 2,000 / 1,000 = 2 days to fail, so 6 / 2 = 3.0 = 2 x 1.50. Account B, 6,000 / 500 = 12 against 3,000 / 1,000 = 3, so 12 / 3 = 4.0 = 2 x 2.00. Account C, 2,500 / 250 = 10 against 2,500 / 500 = 5, so 10 / 5 = 2.0 = 2 x 1.00. Substitute your own winning day assumption and the structure survives. The ratio sets the difficulty. The daily limit sets the clock.
Why the same drawdown number is two different products
Under a static drawdown the cushion is cushion(t) = drawdown + profit(t). It grows with every dollar of progress, so the account gets more forgiving the further you get. Under a trailing drawdown the cushion is cushion(t) = drawdown - (peak(t) - equity(t)), capped at the drawdown amount forever, so it never gets more forgiving no matter how much you have banked.
The sharper statement: a static rule performs one test across the whole evaluation, whether your worst equity ever fell below start minus drawdown. A trailing rule tests every peak to trough retracement you make on the way up. One failure check becomes many. That, not the headline dollar figure, is why identical drawdown numbers behave nothing alike. Under a trailing rule your room to be wrong never grows, so the last dollar of progress is protected by no more cushion than the first.
One equity path, two floors. Illustrative $50,000 start, $6,000 target (end equity $56,000), $3,000 max drawdown. The static floor is a flat line at 50,000 - 3,000 = $47,000 for the entire evaluation. The trailing floor is a staircase that ratchets up to the high water mark minus $3,000 at every new high and never steps back down. Path 1 runs 50,000 to 54,000 to 51,500 to 56,000: the static test sees a low of 51,500 against 47,000, a cushion of $4,500, pass; the trailing test sees the 54,000 peak lift the floor to 51,000 and the low clear it by $500, pass. Path 2 runs 50,000 to 54,000 to 50,800 to 56,000: the static cushion at the low is $3,800, a comfortable pass, while the ratcheted floor still sits at 51,000 and the equity path clips it, failing by exactly $200. Same trader, same $3,000 headline, a $700 difference in one retracement, and one product passes while the other does not.
Quantify the giveback. A trader who peaks at +$4,000 on a $3,000 drawdown can hand back 3,000 + 4,000 = $7,000 from that peak under a static rule before touching the fixed floor, and exactly $3,000 under a trailing rule, permanently. That is 7,000 / 3,000 = 2.33 times the room at the same point in the run. Where the trailing threshold reads unrealized profit in real time it gets sharper: a trade that runs to +$1,500 and closes at +$400 lifts the floor by the full $1,500 while banking $400. Headroom surrendered is 1,500 - 400 = $1,100, which is 1,100 / 3,000 = 36.67% of the buffer spent to gain 400 / 6,000 = 6.67% of the target.
Trailing implementations are not standardized. Some track intraday high water marks including open profit, some recalculate only from the highest closing balance, and some stop trailing once the floor reaches the starting balance while others trail the whole way. Confirm in writing which variant applies, because that choice moves the arithmetic more than the dollar figure does. The full mechanics are set out in trailing vs static vs end of day drawdown. Staged programs need R computed per stage, since a second stage often keeps the same buffer against a smaller target, which is why one step, two step and three step evaluations cannot be compared on their combined headline targets.
Where does the fee belong in the comparison?
Last, and normalized two ways. Illustrative fees: Account A at $150 gives fee/drawdown = 150 / 2,000 = 7.50% and fee/target = 150 / 3,000 = 5.00%. Account B at $200 gives 200 / 3,000 = 6.67% and 200 / 6,000 = 3.33%. Account C at $100 gives 100 / 2,500 = 4.00% on both measures, because its target and drawdown are equal. On fee per dollar of target, Account B looks cheapest at 3.33% while carrying the hardest ratio at 2.00. On fee per dollar of buffer, Account C is cheapest at 4.00% and holds the easiest ratio at 1.00. Cheap per unit of required profit is not cheap per unit of survivability.
Firms price ratio, size, split and fee against each other, so the sticker alone tells you little. The CFTC's guidance on contractual obligations tells prospective commodity pool investors to pay particular attention to the break even analysis and other required fee disclosures before allocating funds. A prop firm evaluation is not a commodity pool, but the habit transfers: compute your ratios from the rules document you accept at checkout, not from the grid that sold you the click.
The four number checklist
- Target / max drawdown. This is R, and lower is more forgiving. An account at R = 2.00 demands two full buffers of net progress and roughly twice the trade count of an R = 1.00 product at the same win rate and risk fraction.
- Daily loss limit / max drawdown. The share of your buffer one maximum loss session can burn. Its reciprocal is the minimum number of maximum loss days to fail. Then compare the cap itself against your realistic worst session, not your average one.
- Drawdown type. Static, end of day trailing or intraday trailing, and whether the floor stops trailing at the starting balance. Get the answer from the rules document, not from support chat, and note whether unrealized profit moves the floor.
- Only now, the fee, as fee/drawdown and fee/target. If two products still tie on all four, the tiebreakers are consistency rules, minimum days and payout terms.
What the ratio does not capture
R ignores consistency rules, minimum trading days, time limits, news blackout windows, scaling plans, payout schedules and activation fees. An R = 1.00 account with a harsh single day profit cap can be harder overall than an R = 2.00 account with no such rule. Treat R as a first filter that eliminates products fast, not as a verdict that picks one.
None of the arithmetic above includes round turn commissions, exchange and market data fees, or slippage. Your effective target is the stated target plus total transaction cost across the evaluation, so realized R is always worse than published R, and the gap scales with trade frequency. A scalper should add an estimated cost figure to the numerator before comparing offers.
The honest tradeoff: a low ratio is easier to pass and almost never arrives free. Firms price these variables against one another, so a forgiving ratio usually comes attached to a smaller account, a lower profit split, a higher fee, a tighter daily cap, or a trailing drawdown instead of a static one. The goal is not hunting the smallest R. It is matching R to your strategy's real drawdown profile: an approach that gives back a lot of open profit wants a high daily/drawdown ratio and a static floor more than it wants a small target, while a patient low frequency approach can carry a high R without noticing.
A copier does not fix any of this. Running one strategy across several evaluations does not multiply your chances, it correlates them. If a single evaluation has a 20% pass probability and five accounts were genuinely independent, at least one pass arrives with probability 1 - 0.8^5 = 1 - 0.32768 = 0.67232, roughly 67.2%. Copy the identical strategy to all five and outcomes become perfectly correlated, so the probability stays at 20%. The downside compounds where the upside does not: a session losing $200 loses 5 x 200 = $1,000 across five copied accounts, and if each carries a $1,000 daily loss limit, one ordinary bad day breaches all five at once. Copying solves execution across accounts you already decided to run. It cannot solve a ratio your strategy cannot clear, some firms restrict or prohibit it in their rules document, and correlated size can degrade your own fills.
Evaluation stages at futures prop firms typically run on simulated accounts, and simulated track records carry limits regulators treat as material: 17 CFR 4.41 requires a prescribed cautionary statement whenever simulated or hypothetical performance of a commodity pool operator, commodity trading advisor or a principal of either is presented, prominently disclosed and in immediate proximity to the figures. Whether a given retail prop firm sits inside that part is a separate question, and the regulatory treatment of these firms keeps moving, so verify status rather than assume it. Keep the CFTC's basic instruction in view: know how much you can afford to lose before committing. Your buffer is that number written as a rule. The ratio tells you how far it has to stretch.
Frequently asked questions
What is a good profit target to drawdown ratio for a prop firm evaluation?
Lower is easier, because the ratio tells you how many full buffers of net profit the evaluation requires: 1.0 means one buffer, 2.0 means two. At a fixed risk fraction and win rate, an account at 2.0 needs roughly twice the trade count of one at 1.0. Which figure suits you depends on your strategy's drawdown profile, not on which number is smallest.
Does a larger dollar drawdown mean an easier evaluation?
No, a larger dollar buffer only helps if the profit target does not grow faster than the buffer does. An illustrative account with a $6,000 target and a $3,000 drawdown (ratio 2.00) is harder than one with a $3,000 target and a $2,000 drawdown (ratio 1.50), even though its buffer is 50% larger in dollars. Compare the quotient, never the headline dollar figure.
Does account size change how hard an evaluation is?
Not on this measure, because the ratio is size invariant. A $150,000 account with a $9,000 target and a $6,000 drawdown computes to 1.50, exactly the same as a much smaller account with a $3,000 target and a $2,000 drawdown. Larger accounts change how many contracts the buffer supports, not the proportional distance you have to cover.
How does the daily loss limit affect evaluation difficulty?
The daily loss limit caps how much of your total drawdown is reachable in a single session, so it sets the pace and the granularity rather than the total difficulty. A $3,000 drawdown with a $1,000 daily cap needs at least three maximum loss days to fail, yet one such day still burns 33.33% of the whole buffer. Compare the cap against your realistic worst session, since a strategy whose bad day loses $900 cannot survive a $500 cap regardless of edge.
Why does a trailing drawdown make the same drawdown number harder?
A static rule tests your equity once against a fixed floor, while a trailing rule tests every peak to trough retracement you make on the way up, turning one failure check into many. Under a static rule the cushion grows as you profit; under a trailing rule it stays capped at the drawdown amount forever. Confirm which variant your firm uses, including whether unrealized profit moves the floor, before comparing offers.
How do I compare prop firm evaluation fees fairly?
Normalize the fee two ways: fee divided by max drawdown and fee divided by profit target. An illustrative $200 fee on a $6,000 target looks cheap at 3.33% of target while that account carries the hardest ratio at 2.00, and a $100 fee on a $2,500 buffer is 4.00% with the easiest ratio at 1.00. Cheap per dollar of required profit is not the same as cheap per dollar of survivability.
Does the ratio account for consistency rules and commissions?
No, and both can flip a comparison. Consistency rules, minimum trading days, time limits, news blackouts, scaling plans and payout terms sit outside the arithmetic entirely, so a forgiving ratio paired with a harsh single day profit cap can be the harder product. Commissions, exchange fees and slippage raise your effective target, which makes realized difficulty always worse than the published ratio, especially for high frequency strategies.
How do I compute the ratio for a two step or three step evaluation?
Compute it per stage, not on the combined headline target, because each stage carries its own target and its own buffer. A second stage that keeps the same drawdown against a smaller target has a lower ratio than stage one and is a different product to pass. The stage with the highest ratio is the one that will decide the outcome, so size and plan around that stage.
Should I just pick the evaluation with the lowest ratio?
No, because firms price the ratio against everything else on the offer. A forgiving ratio usually arrives with a smaller account, a lower profit split, a higher fee, a tighter daily cap, or a trailing drawdown instead of a static one. Choose the ratio that matches how your strategy actually draws down rather than the lowest number available.
Does copy trading several evaluations improve my odds of passing?
No, copying one strategy across several accounts correlates the attempts instead of multiplying them. If five accounts were genuinely independent at a 20% pass rate, at least one pass arrives 67.2% of the time, but identical copied trades keep the probability at 20%. The losses do scale, so a session losing $200 loses $1,000 across five accounts and can breach five $1,000 daily caps at the same moment.