Prop Research

The rules are a range, not one company

Iron Ledger tests its models against the range of rule sets prop firms publish — profit targets, drawdown limits, daily loss limits, consistency requirements and position sizing. No firm is named and no company’s account is modelled. The result is a second layer of research alongside each model’s underlying performance record.

What a rule set contains

Every prop research test is defined by one rule set: the values a firm publishes, swept across the range the market sells.

The range swept is deadline 14, 30, 60, 90 days or none; drawdown USD 250 to USD 8,000; target 1.0x, 1.5x or 2.0x the drawdown; daily loss cap none or half the drawdown; consistency none or 40%. The complete rule set is shown with each model’s research, together with its clock, so the results can be assessed against the actual thresholds.

Which rule sets each model fits

Models that have completed Iron Ledger’s prop research carry a prop research designation. The model’s standard performance record remains unchanged. The prop research adds another question: which rule sets does the model fit, and which does it miss?

A strategy can produce strong historical results and still behave differently when drawdown, daily loss and profit targets become hard limits, and a deadline decides whether any of it can be kept. Iron Ledger therefore measures the model against the range the market publishes and records the results alongside its standard research.

Scope: every cell on this page is measured on the chart the model was built on — five minutes for the session scalps and the Dow entry models, fifteen minutes for Floor Traders Fulcrum (heating oil). Timeframe is a parameter with no result yet: the one-minute chart, the most common chart for scalping, and the ten- and fifteen-minute charts have not been tested. They are the next axis of this research, together with further markets. A breach rate is part of the measurement, not a hidden failure: the clock beside every pass rate is what shows whether a pass could actually have been kept.

Method: Evaluation started at the first trade of each month and ran the real fill sequence forward, with a 40% consistency rule applied, over the month-starts the record contains - 11 for every model except FSMI, which skips July by design and therefore has 10 (Oct 2025 - Aug 2026). Position size fixed at the start of each evaluation; no resizing. Measured before commission. Each cell is then read against the range of rule sets prop firms publish (see prop_rule_range) and states its clock: how many month-starts finished inside 90, 60 and 30 days, not only how many reached the target at all.

The research record

Each completed test shows the actual numbers.

YM · Entry Flash

Tested against one rule set from the range prop firms publish. The model’s standard performance record is unchanged; this records how it behaved inside those thresholds at the stated size.

Result: tested against the range of rule sets prop firms publish, at 2 contracts per signal (as traded). Rule set: trailing drawdown USD 5,000, profit target USD 7,500 (1.5x the drawdown), no daily loss cap, 40% consistency. No breaches of the tested drawdown or daily-loss parameters over the test period.

MES · MES Session Scalp

Tested against one rule set from the range prop firms publish. The model’s standard performance record is unchanged; this records how it behaved inside those thresholds at the stated size.

Result: tested against the range of rule sets prop firms publish, at 2 contracts per signal (as traded). Rule set: trailing drawdown USD 1,000, profit target USD 1,500 (1.5x the drawdown), no daily loss cap, 40% consistency. No breaches of the tested drawdown or daily-loss parameters over the test period.

FDXM · Mini-DAX (FDXM) Session Scalp

Tested against one rule set from the range prop firms publish. The model’s standard performance record is unchanged; this records how it behaved inside those thresholds at the stated size.

Result: tested against the range of rule sets prop firms publish, at 2 contracts per signal (as traded). Rule set: trailing drawdown USD 2,000, profit target USD 3,000 (1.5x the drawdown), no daily loss cap, 40% consistency. No breaches of the tested drawdown or daily-loss parameters over the test period.

FESX · FESX Session Scalp

Tested against one rule set from the range prop firms publish. The model’s standard performance record is unchanged; this records how it behaved inside those thresholds at the stated size.

Result: tested against the range of rule sets prop firms publish, at 2 contracts per signal (as traded). Rule set: trailing drawdown USD 1,000, profit target USD 1,500 (1.5x the drawdown), no daily loss cap, 40% consistency. No breaches of the tested drawdown or daily-loss parameters over the test period.

FSMI · FSMI Session Scalp

Tested against one rule set from the range prop firms publish. The model’s standard performance record is unchanged; this records how it behaved inside those thresholds at the stated size.

Result: tested against the range of rule sets prop firms publish, at 2 contracts per signal (as traded). Rule set: trailing drawdown USD 2,000, profit target USD 3,000 (1.5x the drawdown), no daily loss cap, 40% consistency. No breaches of the tested drawdown or daily-loss parameters over the test period.

HO · Floor Traders Fulcrum (HO)

Tested against one rule set from the range prop firms publish. The model’s standard performance record is unchanged; this records how it behaved inside those thresholds at the stated size.

Result: tested against the range of rule sets prop firms publish, at 2 contracts per signal (as traded). Rule set: trailing drawdown USD 7,500, profit target USD 11,250 (1.5x the drawdown), no daily loss cap, 40% consistency. No breaches of the tested drawdown or daily-loss parameters over the test period.

Prop research library

The models tested against the published rule range, with the rule set each result belongs to. A pass rate is only meaningful with its rule set, its clock and its contract count, so all three are printed beside it.

MYM, FDXS are live models that are not published as prop research models.

Research methodology

Prop research follows the same research discipline as the rest of Iron Ledger.

The underlying model remains the starting point. Where research produces a materially different configuration, that configuration is documented separately rather than blended into the original performance record.

Important

• 11 month-starts is a small sample. Treat pass rates as indicative, not predictive.
• Start count. An evaluation can only start in a month the model actually traded. FSMI skips July by design, so it is measured over 10 month-starts, not 11. Every percentage is taken over the starts that exist for that model, and the cell states its own count.
• Time. These figures measure whether a run reaches the target, not whether it does so inside a firm's deadline. Every cell therefore states its clock as well: how many month-starts finished inside 90, 60 and 30 days, not only how many reached the target at all. Most runs reach the target after 90 days, and at the traded size, on an account whose drawdown matches the risk the model itself needs, no model on this book finishes a month-start inside 30 days.
• A monthly or daily loss brake does not protect a prop account: tested at 1.5-5% of the account, breach rates were unchanged on every product while the record net fell.
• Breaches come from intra-trade excursions and the trailing floor, which a brake watching realised loss cannot see. The lever is size and tier.
• Every firm's rules differ and change without notice. These figures measure risk the strategy produced in the tested window at the stated size; they are not a promise of a pass.
• Currency. Results are shown in each market's native currency - USD for MES, EUR for FDXS, FDXM and FESX, CHF for FSMI. The test parameters are USD-denominated. Where a EUR result is stated against them it is converted once, at the end of the evaluation, at the fixed research rate used throughout it - EUR 1 = USD 1.08 - the same rate the evaluation applies to the EUR point values (EUR 1 to USD 1.08, EUR 5 to USD 5.40, EUR 10 to USD 10.80). No daily or per-trade conversion is applied.
• Currency (FSMI). FSMI results are stated in CHF, the currency of the contract and of the export. Where they are compared against the USD-denominated test parameters they are converted once, at the end of the evaluation, at the rate used for the FSMI headline - CHF 1 = USD 1.12. No per-trade or daily conversion is applied.
• Units (HO). HO's risk figures are published in ticks, because a tick is what its chart prints: one tick is USD 4.20 a contract. Each one therefore carries its account-money equivalent after the tick count, and both are re-derived from the export on every build. A tick count without that equivalent is not comparable with a USD drawdown and must not be read as one.

Hypothetical performance results have many inherent limitations. Results are prepared with the benefit of hindsight, do not involve financial risk, and may not reflect commissions, slippage or actual fills. Past performance is not indicative of future results.