Backtest Research
Simulated parameter analysis for FSD-X ORB PRO (Knightfall MK2) from June 18, 2019 through September 30, 2026 ($250 and $400); September 4, 2026 ($500, $650, $850 and $1000). Knightfall now has two variants built on the same breakout rules: Variant 1 (stock) and Variant 2 (runner). Pick one with the Variant switch below. Risk is scaled dynamically by setup grade. Hypothetical results — past performance is not indicative of future results.
These are ORB PRO (Knightfall) results. They come from a strategy that grades every breakout, filters the ones that don't qualify, and sizes by grade. Raven does none of that — it's an indicator that marks every breakout and draws the levels, with no backtest and no track record of its own. Nothing on this page describes Raven.
⚠ All data on this page reflects hypothetical simulated backtest results, not actual trading. Past performance is not necessarily indicative of future results. Figures are gross of costs — they exclude commissions, fees, and slippage, which vary by broker and reduce net performance. Results shown are from a specific risk profile and are not typical.
Knightfall runs in two variants. Both use the same breakout rules, grading and sizing. Variant 1 is the original: it takes profit at TP1. Variant 2 holds the full position for TP2 and trails the stop once TP1 is reached. A trailed exit can free up a later entry the same day, so trade counts differ slightly. Every figure below belongs to the variant selected above.
Same chart, same dates for both: . Hypothetical, simulated results, not actual trading. A higher backtest figure is not a promise of a better future result.
$400 is where the selector starts. It isn't a recommendation.
Every setting takes the same trades. What changes is how big the swings are, and whether your account has room for them. Three questions decide it.
A 50K has less room to absorb a losing run than a 150K. The bigger the account, the more setting it can carry.
Moving stops, re-entering, running a drawdown protocol. If you'll take every signal exactly as printed and do nothing else, size down.
Not what you can survive on paper. What you'll actually hold through without switching it off mid-drawdown.
Straight losses at full size. Real sequences mix wins, losses and smaller grades.
Newer, or one account you can't easily replace? $250 is the most forgiving place to start. It's slower. It's also the setting where the drawdown has come closest to never being a problem.
The deepest drawdown in the backtest is the worst so far, not the worst possible. Size for what you can afford to lose, not for what the strategy did.
Pick a setting above, then open to see how it held up at your account size.
Hypothetical, simulated results, not actual trading. Past performance is not indicative of future results. Not financial advice.
// RECONCILES WITH TRADINGVIEW
Every figure on this page comes from the TradingView Strategy Tester export for the selected risk setting, so you can open the same strategy on your own chart and match it line for line. The grey figure under each number is what the Strategy Tester prints.
One figure is measured differently, and it is the drawdown. TradingView reports the deepest point a trade reached while it was still open. This page reports the drop from one closed trade to another. Both are real; they answer different questions. TradingView's is what your account screen would have shown at the worst moment — the number to size a prop account against. Ours is what the strategy actually banked, which is the one that matters for comparing risk settings. The two are shown side by side rather than picking one.
Profit factor is the same number rounded differently — TradingView prints three decimals, this page prints two. Net profit, win rate and the winning/losing split are identical. Strategy Tester settings: Deep Backtest, MNQ 5-minute, the risk setting selected above. Figures are gross of commissions, fees and slippage.
// EQUITY CURVE — CUMULATIVE P&L
// YEAR BY YEAR — BACKTEST DATA
Deepest dip is closed-trade, measured inside each year. 2019 begins June 18 and 2026 runs through September 4.
// LONG VS SHORT — SIMULATED
Direction split at the selected risk setting and window.
// PERFORMANCE BY GRADE — SIMULATED
Grade sets position size under the risk ceiling. A+ through C are all traded; D and F are not.
// SYSTEM ANATOMY — SIMULATED
The full stat sheet at the selected setting and window. Every figure is labelled in plain terms so it reads correctly on its own.
The same simulated trades, cut by month, quarter, weekday and week of month. Setups arrive at a similar rate all year — what each setup returned did not. Win rates and trade counts are identical at every risk setting; only the dollar figures rescale.
// AVERAGE RESULT PER TRADE, BY MONTH
Simulated dollars per trade, pooled by calendar month.
// BY QUARTER
// BY DAY OF WEEK
// BY WEEK OF MONTH
Week 5 is a partial bucket — only months with a 29th onward contribute. Read it as a footnote, not a finding.
// EVERY MONTH, EVERY YEAR
Simulated net per calendar month. Values are printed in every cell, so nothing is carried by colour alone.
// WHAT THIS DATA DOES NOT SAY
Seven years is six to eight observations per calendar month, not ninety. Nothing here is tested for statistical significance. The quarterly and day-of-week splits pool the most trades and are the most dependable; Week 5 pools the fewest and is the least.
This is not a forecast, and it is not a filter we apply. The strategy takes every qualifying setup in every month. We publish it so a slow stretch reads as a slow stretch instead of a broken system — not so anyone sits out a quarter.
A net figure tells you where the record ended up, not how it got there. This is the shape of the curve — the climbs and the dips, how long each lasted, and how many trades ran back to back in the same direction.
// THE TEN STRONGEST CLIMBS
Trough of one dip up to the peak before the next.
// THE TEN DEEPEST DRAWDOWNS
Peak to trough, then trough back to a new equity high.
// NET BY YEAR
// DEEPEST DIP BY YEAR
// RUNS — HOW OFTEN
Consecutive trades with the same outcome. Identical at every risk setting — sizing changes the dollars, not the sequence.
// BEST MONTHS
// WORST MONTHS
// HOW TO READ THIS
Slow stretches are normal. Here is how normal.
Four questions, answered from seven years of simulated results. If you are in a flat patch right now, this tells you whether the strategy is behaving the way it always has.
The part most people get wrong
Traders assume a working system sits at a new high most of the time. It does not — and neither does this one.
› Show the numbers behind this
How often a stretch finished lower than it started
Take any day in the sample, look ahead 30, 60 or 90 days, and see whether the account is higher or lower. Repeated from every possible start date, one day at a time. These figures shift between risk settings — grade-based sizing changes each trade’s weight, so a stretch can finish green at one setting and red at another.
Stops at 90 days deliberately. Seven years does not contain enough separate year-long stretches for a one-year figure to mean anything, and publishing one would read like a promise the sample cannot support.
Every year had a bad month
The worst 30-day stretch inside each calendar year. Note 2022 — the deepest bad month in the sample landed in the strongest year on the page.
Red days and red weeks
Counted day by day and week by week. The share of red days holds steady across risk settings, but weeks and the time-below-a-high figure shift a little, because grade-based sizing changes how much each trade contributes to a day’s total.
How to read all of this
The setting changes the size. It doesn't change the edge.
Read the risk-profile table column by column. In Variant 1, over the same dates, win rate is identical at every setting and the trade count never changes, because they are the same trades. Profit factor moves inside a tenth.
What does move is net and drawdown, and they move together, in both directions. Turning the setting down makes the swings smaller. It does not make the strategy worse.
Why that matters if you run more than one account
A trader who wants more exposure has two paths. Turn the risk up on a single account, or run the same conservative setting on more than one. The table above is the argument against the first — a higher setting buys bigger swings, not a better edge.
The same breakout, the same grade, the same stop and target. An account does not get its own analysis — there is only ever one read of the morning.
Adding an account adds no work. You are not managing several strategies, you are running one set of rules in more than one place.
Copy-trade tooling routes the same alerts to each account. Every account runs the setting you chose for it, entirely independently.
This section describes how the strategy is structured, not what any account will earn. There is no multi-account backtest — every figure on this page is a single-account simulation at the stated risk setting. Hypothetical and simulated, not actual trading, gross of commissions, fees and slippage. Drawdown figures are backtested maximums, not limits, and not a prediction of what any account will experience. Past performance is not indicative of future results.
Check the numbers yourself.
These are the unedited TradingView exports behind every figure on this page — every trade, win or loss, nothing removed. Updated October 2026.
Same entries, same exits, same dates. Only the risk-per-trade setting differs, which is what changes the dollar columns — the win rate and profit factor are the same in both.
These exports run past the page cutoff. This page counts trades through Sep 30, 2026 so both variants cover the same dates; later trades are left in the files, unedited. Each variant's two files hold the same trades; only the size differs. All runs use TradingView's High detalization.
Members can run either file through the Trade Importer to turn it into a Backtest session and review it year by year and month by month.
Hypothetical simulated backtest results, not actual trading. Figures are gross of costs — the Commission column in these files is zero, and real commissions, fees and slippage vary by broker and reduce net performance. Past performance is not necessarily indicative of future results.
How the same simulated signals behave against a prop firm evaluation. Pick your account size below and the risk setting above. Evals are modelled on an end-of-day trailing drawdown and run back to back — the moment one passes or fails, the next begins. Each calendar year starts fresh on January 1.
// EVALS BY YEAR
Each year simulated independently from January 1. The last column shows when that year’s final eval completed, which may fall in the following year.
// ALL RISK SETTINGS AT THIS ACCOUNT SIZE
The same account, every risk setting side by side. The row matching your selection is highlighted.
// HOW THE EVAL IS MODELLED
This is the data. The next step is your own chart.
Run the same ORB engine that produced this research on live NQ/MNQ price action. 30-day free trial, cancel anytime.
$129/mo after trial · Cancel anytime in one click via Whop
Trading futures involves substantial risk of loss and is not suitable for every investor. Past performance is not necessarily indicative of future results.
Not ready to join? The Discord is free.
Market talk, our free indicator, and traders working on the same thing. Look around before you decide.