Video Summary - Turtles Trading Strategy Explained

Finance-focused summary: Turtles Trading Strategy (Richard Dennis / Bill Eckhardt “Turtles”)

The video explains the Turtles trading system, originally tested in the early-to-mid 1980s in futures. It emphasizes a purely mechanical, technical-analysis-only ruleset with defined entry/exit, position sizing, pyramiding, portfolio/correlation limits, and drawdown-based scaling—aimed at producing positive expectancy over many trades despite long losing streaks.

Key people / sources (method origin)

Richard Dennis (believed traders can be taught; trained “turtles”)
Bill Eckhardt (disagreed; believed successful traits are innate)
Mentions:
- A group of 13 traders (“turtles”)
- Later trainees trained by Dennis/Eckhardt

Instruments / tickers / assets mentioned

Crude oil (example used to illustrate breakout/breakdown logic)
Afterpay (Australian Buy Now Pay Later leader) — APT (ASX)
- Mentioned price levels: ~$40 → $8 (post-COVID sell-off); later ~$22, $27.10/27.11, $28.15, $41.14/41.15
Futures markets (general; no specific ticker given)

Explicit numbers and performance claims

Training / proof phase

1983: recruitment advertised in Barron’s / Wall Street Journal
End of 1983: 2-week training course
January 1984: trial on futures markets with $50,000 from Dennis

Capital allocated

Most turtles: $500,000 to $1,000,000 (Dennis’s capital)

Results claimed

Average return: 80% per annum
Total profits: > $175 million (1980s dollars)

Methodology / rule framework (step-by-step)

1) Core premise: mechanical system + positive expectancy (edge)

The system is technical-analysis-only (no news/fundamentals).
You follow the trading rule exactly; no discretion.
“Edge” is framed via expectancy:

Expectancy = (win rate × avg win) − (loss rate × avg loss)

Illustrative expectancy example (hypothetical 56 trades)

Winners vs losers: fewer winners by 24 to 32
Avg win ≈ 4× avg loss
Example result: +12.50 per trade (illustrative)

Recommendation / caution: If your system’s expectancy isn’t positive, the video suggests you shouldn’t be trading it.

2) Trade entry: two systems (breakouts & breakdowns)

Turtles use two lookback rule sets:

System 1 (S1): 20-day look-back
System 2 (S2): 55-day look-back

Common trigger logic (long and short)

Long trigger: breakout above the highest high of the lookback period (+1 tick)
Short trigger: breakdown below the lowest low of the lookback period (−1 tick)
A “tick” is the smallest price increment (for stocks cited as $0.01).

S1 entry rules (20-day)

Long: enter if price trades ≥ (highest high of prior 20 days + 1 tick)
Short: enter if price trades ≤ (lowest low of prior 20 days − 1 tick)
Extra S1 filter / quirk:
- Only take an S1 trade if the last S1 trigger was “unsuccessful.”
- “Unsuccessful” is defined later as a prior trade that hit either stop type (volatility or trailing stop).

S2 entry rules (55-day)

Long: price trades ≥ (highest high of prior 55 days + 1 tick)
Short: price trades ≤ (lowest low of prior 55 days − 1 tick)
No S2 filter quirk: S2 acts as a “catch-all” if S1 was rejected.

3) Position sizing + risk per trade (volatility-based)

Risk model: % of capital at risk per trade
Stated turtle parameter:
- 2% of total capital risked per trade
Stop distance tied to volatility using ATR:
- Stop-loss distance = 2 × ATR over the relevant lookback period
Position size derived from fixed dollar risk:

Shares = (dollar amount risked, net of commissions) / (2 × ATR)

Example using the video’s stock math

Capital: $10,000
%R = 2% → risk budget $200
Commissions example: $16 → risk used for shares calculation: $184
If 20-day ATR = $0.50, then stop distance = 2 × 0.50 = $1.00
Shares = $184 / $1.00 = 184 shares

Implication: higher volatility → larger ATR → wider stop → fewer shares → risk normalizes automatically.

4) Exits: volatility stop + trailing stop (defined by lookback)

System-specific exit rules:

S1 long exit: when price trades below the lowest low of a 10-day look-back (trailing-stop behavior)
S1 short exit: when price trades above the highest high of a 10-day look-back
For System 2, the video describes similar logic using a longer look-back (20-day) for trailing/exit levels.

How the trailing stop behaves:

If the trade moves in favor, the trailing line moves with it
If breached, you exit

5) Pyramiding (adding to winners)

“Pyramiding” = add to the position as the trend strengthens.
Rule:
- Add another position when price moves by 0.5 × ATR
- The added unit’s stop is set at the same amended stop point so total risk on added exposure is structured and controlled.
Limit:
- Max additions: 3 plus the original entry (up to 4 units total per market)

Purpose: amplify gains in strong trends while stops protect against wrong-way moves.

6) Portfolio risk: correlation-based unit limits + directional caps

The video describes portfolio limits based on how markets move together:

Closely correlated markets: max 6 units risk
Loosely correlated markets: max 10 units risk
Directional max exposure:
- max 12 units long or 12 units short

Sector analogy (video framing):

Stocks in the same sector treated as closely correlated
Sectors treated as more loosely correlated as correlation decreases
Portfolio should be skewed:
- Bull market: prefer long bias
- Bear market: prefer short bias

7) Capital management during drawdowns (scaling down risk)

Drawdown: sustained deterioration of trading capital.
Explicit rule:
- When drawdown reaches each additional 10%, reduce new trade position size by 20%
Goal: reduce damage from losses and stress; scale up when conditions improve.

Afterpay (APT) case study numbers used in the rules

Context (video example)

COVID sell-off: approx $40 → $8
By (example) April 9, price ~$22

A) S1 long setup around April 9

S1 trigger:
- Highest high over prior 20 periods → 27.10 + 0.01 = 27.11 (using “+1 tick”)
S1 filter check:
- The prior S1 on Feb 20 was rejected/not allowed because it hadn’t yet been proven “unsuccessful” (still above the ATR-based stop at that time)
- Presented as a guard against a “false breakout”
Volatility and stop:
- Current ATR20 = 2.86
- Stop distance = 2 × 2.86 = 5.72
- Stop level = 27.11 − 5.72 = 21.39
Position size example:
- With assumed $10,000 capital, 2% risk, $16 commissions
- Shares calculated as 32 shares

B) S2 long setup when S1 was “missed”

On May 8, the video switches to S2 because the earlier S1 opportunity rules-out entry.
S2 trigger:
- Highest high over prior 55 periods:
- 41.14 + 0.01 = 41.15
Volatility and stop:
- ATR55 = 2.29
- Stop distance = 2 × 2.29 = 4.58
- Max stop = 41.15 − 4.58 = 36.57
Shares:
- Shares calculated as 40 shares (larger than 32 due to lower volatility at the later time)

Exit example outcomes mentioned (high-level)

S1 trade:
- Hit trailing stop earlier
- Exit around $65 in early August
- Characterized as ~140% profit
S2 trade:
- Held longer
- Exit into early 90s
- Characterized as well over 100% profit

Recommendations / cautions stated or implied

Psychology is discussed, but the video argues expectancy matters more:
- If your system has positive expectancy, psychology may be easier.
Follow rules mechanically; no discretion.
Verify/compute expectancy for any strategy.
Expect long drawdowns and losing streaks; the framework includes drawdown scaling.
S1 includes a “last unsuccessful trade” filter to reduce false breakouts (presented as an effective but described as an “undocumented” quirk).

Disclosures / disclaimers

Subtitles include marketing/training content and promotional links, but no explicit “not financial advice” disclaimer appears in the provided text.

Presenters / sources (end)

Presenter not explicitly named in the subtitles.
Primary credited people:
- Richard Dennis
- Bill Eckhardt
Recruitment mention:
- Barron’s and Wall Street Journal (as publication sources for the ad).

Turtles Trading Strategy Explained - COMPREHENSIVE

Key takeaways