Larry Connors & Cesar Alvarez
Double Seven: three rules, one number, and the week's worst close
This is the shortest complete rule set in the library: one filter, one entry, one exit, and the only parameter in any of them is the number seven. It comes from the same 2008 book as RSI(2), makes the same bet with a simpler trigger, and inherits the same uncomfortable property — the published rules contain no stop loss, and the exit is a level that a falling market may never print.
- Style
- Mean reversion
- Approach
- Mechanical
- Difficulty
- Beginner
- Horizon
- Swing (days to weeks)
- Holding period
- Days to a few weeks
- Time needed
- 10 minutes after the close
- Markets
- Index ETFs · Sector ETFs
- Source
- Short Term Trading Strategies That Work — Larry Connors & Cesar Alvarez
The rule set
- Only consider the trade while the close is above the 200-day moving average
- Buy when the close is the lowest close of the last seven trading days
- Sell when the close is the highest close of the last seven trading days
- Published as a no-stop system — the trend filter is the stated risk control
- A holding cap is a sensible addition, so a reversion that never comes does not become a permanent position
What makes it distinctive
- Three rules and no parameters to tune beyond the number seven — there is almost nothing here to curve-fit
- The 200-day filter is what makes it work: it only ever buys dips inside an established uptrend
- Designed for index ETFs, where a dip is usually noise and a single-name blow-up cannot happen
When it works
Index ETFs in a bull market, where shallow pullbacks resolve upward within a few days and every dip finds buyers.
When it fails
The absence of a stop is not free. In a market that breaks down while you are already long, the only exit in the rules is a new 7-day closing high — and in a persistent decline that level may never print. Never run it on single stocks that can gap away.
How a decision moves through it
Input
Daily closing prices
Closes only. The filter, the entry and the exit are all computed from the close, which is why the whole routine fits in ten minutes after the bell.
Measure
200-day moving average
The regime gate. It decides whether the instrument is eligible at all, and it is the only thing standing between this system and buying every leg of a bear market.
Measure
Rolling 7-day extremes of the close
The lowest and highest close of the last seven bars. Two rolling extremes replace every oscillator: today either is the week's worst close or it is not.
Decide
New 7-day closing low, inside an uptrend
Both together. The window includes the current bar, so 'at or below the 7-day low of the close' means exactly 'today is the week's lowest close'.
Act
Buy the close, exit at the first 7-day closing high
The exit is the entry's mirror image. The product's graph also carries a 20-bar holding cap, an addition the published rules do not have, so a failed reversion cannot sit open indefinitely.
The whole system fits in three sentences
Is the market above its 200-day moving average? If it is, and today's close is the lowest close of the past seven trading days, buy. Sell the first time a close is the highest of the past seven days. That is the entire system — there is no oscillator, no threshold to pick, no pattern to recognise.
One parameter means there is almost nothing to optimise, and therefore almost nothing to fool yourself with. Most systems fail in backtesting because their settings were tuned to the past; this one barely has settings.
The trade it produces is specific: the market has drifted down for a few days inside a longer uptrend, everyone who wanted to sell the dip has sold, and the position is a bet that the drift resolves the way dips in bull markets usually do. Holding periods run from a couple of days to a few weeks, and most of the time the account is in cash.
The RSI(2) sibling: one bet, two triggers
Double Seven comes from the same book as RSI(2) — Short Term Trading Strategies That Work, 2008 — by the same authors, behind the same 200-day filter, with the same deliberate absence of a stop. The two are not merely similar; they are the same underlying bet, that short-term weakness inside a long-term uptrend gets bought.
| RSI(2) | Double Seven | |
|---|---|---|
| Regime filter | Above the 200-day average | Above the 200-day average |
| Pullback detector | 2-period RSI below a threshold | Today is the lowest close of 7 days |
| Parameters to choose | RSI period, entry threshold, exit rule | The number seven |
| Exit | Close above the 5-day average | First 7-day closing high |
| Stop loss | None published | None published |
The missing stop, and an exit the market has to grant
Connors's argument against stops is the same here as in RSI(2): the system buys weakness, so a stop below the entry sells at the moment the setup is most extreme, and every honest backtest shows stops reducing the results. What Double Seven adds is a sharper version of the problem — its only exit is a new 7-day closing high, which is a level the market has to produce. In a persistent decline it may simply never appear.
A system that wins on the large majority of its trades and carries no stop has moved its entire risk into the trades you never see in a good year. The sizing page is about carrying that honestly.
Five ways into this system
- One gate, one entry, one exit — and the number seven three timesEvery rule is computed from daily closes, and the entry and exit are the same object mirrored — a rolling seven-day extreme.6 min read
- What the trend filter does and does not do for your riskThe published answer to 'where is the stop?' is 'the trend filter'. That answer gates new trades; it does nothing for the one you are already in.6 min read
- Built for index ETFs, and honest about nothing elseThe premise — a bad week inside an uptrend is noise — is a statement about diversified indices. On a single stock it is frequently false, and the system has no defence for when it is.5 min read
- The decline with no bounce, and the exit that never printsThe system's losses are structural, not accidental: every one of them comes from the same place, a dip that turned out to be the start of something.6 min read
- Double Seven for beginners: the simplest honest system to study firstIf you want to understand what a mechanical system is, this is the one to take apart: every rule is visible, every failure is explainable, and there are no dials to hide behind.5 min read
The ideas behind it
This system assumes you already know these. Each one is explained from scratch in Investing 101.
Compare with
- RSI(2) Mean Reversion (Connors)Inside a long-term uptrend, buy the sharpest short-term pullbacks and sell into the bounce a few days later.
- Turtle Soup Failed BreakoutWhen price breaks a 20-day low and closes back above it, the breakout has failed — buy the recovery with a stop just under the low, and be out within days.
These are documented methods described for study. Nothing here is investment advice, a recommendation, or a claim about future returns — every system on this page has losing periods, and the pages say where.
Reading about a system is not having one.
Plutux is where you write your own rules down, test them against real data, and keep the record your memory would otherwise rewrite. Join the waitlist for early access.