Curriculum·R409 Grid, DCA, and Rule-Based Automation·about 42 min
DCA and averaging bots
By the end of this lesson you can
- →Distinguish scheduled averaging from a martingale safety-order ladder
- →Compute the capital a multiplier ladder requires, and the deviation it covers
- →Compare that coverage against an instrument's largest historical drawdowns
- →State what you hold, and at what average price, when coverage is exhausted
Senior · enrolled learners
This lesson opens with Barings and the doubling sequence, 1992 to 1995.
- What happened
- Nick Leeson recorded losses in Barings' error account 88888 and responded to each one by increasing the position, on the expectation that the Nikkei would recover and the accumulated loss would be erased in a single move. After the Kobe earthquake of 17 January 1995 the index fell instead, and the response was to add again. By mid-February 1995 he held roughly half the open interest in the Nikkei future and about 85 percent of the open interest in the Japanese Government Bond future. The hole was about GBP 827M and Barings, then the oldest merchant bank in Britain, was sold to ING for a token GBP 1. The strategy had worked repeatedly before it did not, which is the property of the structure rather than an accident of that episode.
- The decision point
- A doubling ladder wins on almost every attempt, which is exactly why it is dangerous. Each recovery confirms the method, the required capital grows geometrically while the account grows arithmetically, and the terminal case arrives with the largest position the sequence has ever held. Every averaging bot with a size multiplier is running this arithmetic. The only question the configuration answers is how many rungs there are before the same thing happens.
- Recorded loss
- $1,300,000,000
What you will be able to answer
- →Scheduled averaging or martingale?
- →What does a 1.5 multiplier ladder cost?
- →How much drawdown does that cover?
- →What do you hold when coverage is exhausted?
Orientation and Year One are open: anyone can read them without an account. From Year Two onward the lessons are for enrolled learners, because progress through the later years only means anything if it is tracked against a record.
It is free. We do not sell the list and there is nothing to buy at the end of it.
Sources and review
- https://www.britannica.com/event/bankruptcy-of-Barings-Bank
- https://www.glynholton.com/notes/barings_debacle/
- https://humanfactors101.com/incidents/barings-bank/
- https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3423101
Confidence high·Volatility low·Reviewed 2026-08-07·Owner unassigned
Contested
R403-01 uses Barings for the absence of a risk-per-trade number, which is the decision that was never made. This lesson uses the same episode for the arithmetic of the doubling sequence itself, being what a geometric capital requirement does against an arithmetic deviation. Two different arguments from one incident and the split is declared. Keep it.
The ladder parameters in part two, being a $100 base order, a 1.5 size multiplier, ten safety orders and 2 percent price steps, are stated inputs chosen so the arithmetic is checkable by hand. Real bot configurations add a step-scale multiplier that widens the deviation as the ladder deepens, which increases coverage and increases the capital requirement further. Per R409-L3 recompute with your own settings.
Scheduled averaging into a long-term holding is a different activity from a martingale safety-order bot and is not criticized here. Per P7 the distinction in part one is the point of the lesson.