10 Legendary Traders — Their Systems Decoded for Algorithmic Trading | ForexAlgoTrading

Not every great trader leaves behind a rulebook. Most discretionary legends took their edge to the grave. But a select few — through published research, contest disclosures or systematic design — have left behind systems precise enough to code. This post covers ten of them, filtered specifically for traders building automated Expert Advisors in MetaTrader 5.

Each system is rated for codifiability: how much of the documented edge survives translation into mechanical rules, with honest notes on what does and doesn't port to a single-symbol EA.

Trend Following

Richard Dennis & The Turtles

Donchian Channel Breakout (S1 / S2)

Codifiability

5 / 5

Dennis trained a group of complete novices — the Turtles — into traders who reportedly earned $175M+ in roughly five years using a fully mechanical system. It remains the most completely documented trading system in public record, published verbatim in the so-called Turtle rules document.

Entry	›System 1: buy the break of the 20-day high / sell the break of the 20-day low — skip the signal if the last S1 trade was a winner ›System 2: buy the break of the 55-day high / sell the break of the 55-day low — always taken
Exit	›S1 exits at the opposite 10-day extreme. S2 exits at the opposite 20-day extreme. ›Hard stop at 2N from entry, applied to all open units simultaneously
Risk	›Unit size = 1% equity ÷ N, where N is the 20-period ATR (volatility-normalised sizing) ›Add one unit every half-N in favour, maximum 4 units. Stop trails to 2N from the most recent unit.
Markets	Originally futures — translates directly to XAUUSD, FX majors and indices on D1 / H4

MT5 Build Notes

The ideal first build: 100% mechanical, zero discretion. Donchian channel indicator plus EA with N-sizing, a four-unit pyramiding ledger, and the S1 whipsaw filter running a paper-trade tracker. Hedging account recommended for clean per-unit tracking. This EA is in development — sign up at ForexAlgoTrading.com for release updates.

Volatility Breakout

Larry Williams

Volatility Breakout + Gap Failure Reversal

Codifiability

5 / 5

Williams won the 1987 World Cup Trading Championship turning $10,000 into $1.1M in 12 months — a verified, audited result of +11,376%. His daughter Michelle replicated comparable performance using the same systems a decade later, providing rare cross-operator evidence that the method transfers.

Entry	›Volatility breakout: place a buy stop at today's open plus k × yesterday's range (k ≈ 0.5–0.8); sell stop mirrored below ›Gap failure reversal: when price gaps beyond the prior session's extreme and then fails to follow through, enter in the opposite direction at that extreme level ›Supplementary filters: trade-day-of-week bias and a directional trend filter
Exit	›First profitable open (the "bailout" exit) or a fixed-dollar / ATR stop. No positions held overnight speculatively.
Risk	›Aggressive fixed-fractional sizing. Williams has stated that position sizing — not entry selection — was the primary driver of the contest return, and the source of its intraday drawdowns.
Markets	Index and commodity futures — strong fit for XAUUSD, US30 and NAS100 on D1 session-open logic

MT5 Build Notes

Maps cleanly to pending stop orders placed at session open and cancelled at session close. First-profitable-open exit is a one-liner on each new D1 bar. Day-of-week filter is an input bitmask — the same pattern used in the FAT_Portfolio EA.

Mean Reversion

Larry Connors

RSI(2) Pullback + Double 7s

Codifiability

5 / 5

Connors is a quant researcher whose RSI(2) studies show historically high win rates — 75–90% per trade in equity-index backtests — making this the most heavily replicated mean-reversion edge in retail quantitative literature.

Entry	›RSI(2): close above the 200 SMA and RSI(2) below 5 → buy at close. Optional scale-in on continued weakness. ›Double 7s: close above the 200 SMA and close at a 7-day low → buy
Exit	›Close above the 5-period SMA, or RSI(2) above 70, or 7-day high (Double 7s). No stop loss in original research — exits are condition-based only.
Risk	›Edge degrades with tight stops. Risk is bounded by the 200 SMA regime filter and disciplined position sizing.
Markets	Equity indices primarily (US500, US30, NAS100) on D1. Weaker raw edge on FX pairs.

MT5 Build Notes

Approximately 300 lines of MQL5. The no-stop design is framed as a configurable max-position-loss circuit breaker (default off, matching original research) rather than an absence of risk management. High win rate is a strong marketing angle.

Mean Reversion / Hybrid

Linda Bradford Raschke

Turtle Soup + Holy Grail (ADX Pullback)

Codifiability

5 / 5

A Market Wizard (New Market Wizards, 1992) and decades-long professional trader. Her co-authored book Street Smarts is rare: patterns documented as testable, mechanical rules rather than vague discretionary guidelines.

Entry	›Turtle Soup: when a new 20-day low appears where the prior 20-day low was at least four bars earlier and the breakout stalls — buy a stop just above the old low, directly fading the Turtle System's own breakout signal ›Holy Grail: ADX(14) above 30 and rising, price pulls back to the 20 EMA → enter on the break of the pullback bar's high in the trend direction
Exit	›Turtle Soup: stop below the breakout extreme, trail quickly — this is a 2–5 bar reversal trade ›Holy Grail: target the prior swing high; stop below the pullback low
Risk	›Small fixed risk per trade with emphasis on quick partial profits at the first resistance level
Markets	FX, indices and gold on H1 to D1. Turtle Soup is a false-break system — directly adjacent to the SR FalseBreak indicator already built for this platform.

MT5 Build Notes

Pairs naturally with the Turtle EA as a matched product — "the system and its counter-system" is a strong marketing narrative. Holy Grail reuses existing pullback and EMA scaffolding from the ERT derivative family.

Trend / Multi-Timeframe

Alexander Elder

Triple Screen

Codifiability

4 / 5

Elder's Trading for a Living introduced Triple Screen in 1986 — the canonical multi-timeframe filter architecture. Most multi-timeframe EAs implicitly replicate it whether their authors know the source or not.

Entry	›Screen 1 (tide, e.g. Weekly): trade only in the direction of the weekly MACD-histogram slope or 26 EMA slope ›Screen 2 (wave, Daily): oscillator pullback against the tide — Force Index(2) below zero, or Stochastic below 30 in an uptrend ›Screen 3 (ripple): trailing buy stop placed above the prior bar's high, triggered intrabar
Exit	›SafeZone trailing stop based on ATR of adverse excursions, or a weekly tide reversal
Risk	›2% per trade maximum, plus a 6% monthly drawdown halt — Elder's "six percent rule"
Markets	Any liquid market; FX majors and gold on a Weekly / Daily / H4 stack

MT5 Build Notes

The 6% monthly halt is a prop-firm-mode feature with existing code patterns in the ORB Trader. Main design decision: Screen 2 oscillator as an input enum (Force Index / Stochastic / RSI).

Box Breakout

Nicolas Darvas

Darvas Box

Codifiability

4 / 5

Darvas turned approximately $36,000 into over $2M in 18 months in the late 1950s — trading by telegram while on tour as a professional dancer. Documented in his 1960 book How I Made $2,000,000 in the Stock Market.

Entry	›Box top: a high not exceeded for three consecutive bars. Box bottom: the subsequent low not undercut for three bars. The box is confirmed once both levels hold. ›Buy stop placed just above the box top — only on instruments near their 52-week highs (momentum regime filter)
Exit	›Stop just below the box bottom. Ratchet the stop up box-by-box as new boxes form above entry. Never lowered; no fixed profit target.
Risk	›All risk lives in the stop discipline. Darvas ran highly concentrated positions and accepted large daily swings.
Markets	Originally stocks; the box logic applies to XAUUSD and indices on D1 and H4 in trending regimes

MT5 Build Notes

The box state machine — three-bar confirm top, then three-bar confirm bottom — is the only subtle component. The indicator draws box rectangles (OBJ_RECTANGLE); the EA places a buy stop and manages the ratcheting stop level.

Trend Following

Ed Seykota

EMA Crossover with Portfolio Heat Management

Codifiability

4 / 5

A Market Wizard who reportedly grew a client account approximately 250,000% over 16 years. One of the earliest fully computerised trend-following systems. His stated edge is not the signal but the sizing discipline across a diversified portfolio.

Entry	›Fast EMA crosses above slow EMA (published teaching example: approximately 15/150 on daily data) ›Whipsaw guard: N consecutive closes beyond the cross, or an ATR buffer past the cross level
Exit	›Opposite crossover, or an ATR-based trailing stop. "Ride your winners, cut your losses" — implemented through the stop rather than the exit signal itself.
Risk	›Fixed-percentage heat per position with a portfolio-level heat cap across all open symbols simultaneously
Markets	Diversified across futures — best deployed as a multi-symbol portfolio EA

MT5 Build Notes

Entry logic is trivial. The product value is the portfolio heat manager: aggregate open risk across symbols with a global cap. That differentiates this from every other MA-cross EA on the Market.

Trend / Chaos Theory

Bill Williams

Profitunity — Alligator + Fractals + Awesome Oscillator

Codifiability

5 / 5

Williams authored Trading Chaos and developed the Alligator, Fractals, Awesome Oscillator and Gator Oscillator — a suite so influential it shipped as native indicators inside every version of MetaTrader. This is the only system here where the entire indicator toolkit is already built into the platform.

Entry	›Alligator (13/8/5 SMMA, shifted 8/5/3): when the three lines are ordered and spreading, the trend regime is active ›First entry: fractal breakout beyond the Alligator's teeth in the trend direction ›Add-on entries: Awesome Oscillator momentum signals (saucer, zero-cross) and subsequent fractals — pyramiding is explicit in the method
Exit	›Close when price closes through the Alligator's jaw or lips, or a reverse fractal triggers against the position
Risk	›Stops behind the opposing fractal level; pyramid adds sized decreasingly from the initial unit
Markets	FX and gold on H4 and D1

MT5 Build Notes

The fastest build of the ten — iAlligator, iFractals and iAO are native handles, no custom indicator required. The EA is purely signal orchestration and a pyramiding ledger. Strong commercial angle: every MT5 user already recognises these indicators.

Momentum Swing

Kristjan "Qullamaggie" Kullamägi

Consolidation Breakout / Episodic Pivot / MA Surf

Codifiability

3 / 5

Kullamägi's reported growth to approximately $100M between 2011 and 2021 was achieved with a 25–30% win rate — carried entirely by winners averaging 5–20 times the initial risk. He has documented his methodology publicly through his blog and live-stream sessions.

Entry	›Prior momentum leg (typically 30%+ in weeks) → tightening consolidation above the rising 10 and 20 MAs → buy the breakout of the consolidation range on volume expansion ›Market regime filter: index 10 MA above 20 MA — stand aside when the 10 MA crosses below
Exit	›Sell one-third to one-half at 2–3× risk after 3–5 days; trail the remainder on the 10 or 20 MA with an exit on a close below
Risk	›Stop never wider than 1× the 20-day average daily range; 0.25–1% account risk per trade
Markets	Momentum stocks natively; the consolidation-after-impulse template adapts to XAUUSD and indices on H4 to D1

MT5 Build Notes

Rated 3/5 because stock selection (identifying relative-strength leaders) is half the edge and does not port to a single-symbol EA. The tradeable core — impulse-leg detector, range-tightening detection, ADR stop, tiered partial exits with MA trail — is entirely mechanical and buildable.

Macro / Regime

Paul Tudor Jones

200-Day MA Regime Filter + 5:1 Asymmetry Rule

Codifiability

2 / 5

Jones is a billionaire macro trader who famously tripled his money shorting the 1987 crash. His publicly stated rules are sparse but precise: nothing good happens below the 200-day moving average, and he requires a minimum 5:1 reward-to-risk on every position.

Entry	›Regime gate: long exposure only above the 200-day MA. Research on this filter alone shows materially improved risk-adjusted returns versus buy-and-hold across most instruments. ›Actual entries are discretionary macro. For an EA, combine the regime filter with a mechanical trigger such as a swing pullback.
Exit	›5:1 minimum reward-to-risk — wide target, tight stop. Being wrong four times out of five and still finishing flat.
Risk	›Never average into a losing position. Reduce size after losses and return to full size only after the account recovers.
Markets	Indices and gold on D1

MT5 Build Notes

Only the regime filter and asymmetric R:R discipline are codifiable from public sources. Best shipped as a regime overlay module that bolts onto other EAs rather than a standalone system.

Build Priority Matrix

Ranked by codifiability and MT5 fit, with estimated build effort.

Priority	Trader	Style	Codify	Effort	Target Markets
P1	Richard Dennis — Turtle	Trend	5/5	Medium	XAUUSD · FX · D1/H4
P2	Larry Connors — RSI(2)	Mean-rev	5/5	Low	Indices · D1
P3	Linda Raschke — Turtle Soup / Holy Grail	Hybrid	5/5	Medium	FX · Gold · H1–D1
P4	Bill Williams — Profitunity	Trend	5/5	Low	FX · Gold · H4/D1
P5	Larry Williams — Volatility Breakout	Breakout	5/5	Medium	Gold · Indices · D1
P6	Nicolas Darvas — Box	Breakout	4/5	Medium	Gold · Indices · D1
P7	Alexander Elder — Triple Screen	Multi-TF	4/5	Medium	FX · Gold · W1/D1/H4
P8	Ed Seykota — Portfolio Trend	Trend	4/5	High	Multi-symbol · D1
P9	Qullamaggie — Momentum	Momentum	3/5	High	Gold · Indices · H4/D1
P10	Paul Tudor Jones — Regime Module	Macro	2/5	Low	Bolt-on overlay

Common thread across all ten: the entry signal is the smallest part of the edge. Every system here succeeds or fails on position sizing, loss discipline and exit execution. Building correct risk mechanics matters more than finding a perfect entry trigger — and it is almost always what commercial derivatives get wrong.

Methodology note: Codifiability scores reflect how much of each trader's documented edge survives translation into mechanical MQL5 rules — not a judgement on trader quality. Every system must pass the realism mandate before live deployment: bar-close entries, no Strategy Tester fill assumptions that cannot be honoured live. The EAs from this research series are currently in development. Vol. 2 covers verified high-leverage retail traders and documented small-account flips.

All articles Get in Touch

10 Legendary Traders: Their Systems Decoded for Algorithmic Trading

Richard Dennis & The Turtles

Larry Williams

Larry Connors

Linda Bradford Raschke

Alexander Elder

Nicolas Darvas

Ed Seykota

Bill Williams

Kristjan "Qullamaggie" Kullamägi

Paul Tudor Jones

Build Priority Matrix