Community / general technical analysis
The SMA 5/10 adaptive crossover: a fast cross that measures its own whipsaw
Fast moving-average crossovers catch swings early and get shredded in sideways markets — that is their whole reputation. This system keeps the speed and attacks the shredding directly: a 20-bar meter scores how choppy the recent tape has been, and the choppier it was, the longer a fresh golden cross must hold before the system buys it. Exits take no confirmation, ever, and the reason for that asymmetry is the best lesson in the system.
- Style
- Trend following
- Approach
- Mechanical
- Difficulty
- Beginner
- Horizon
- Swing (days to weeks)
- Holding period
- Days to weeks
- Time needed
- 10 minutes after the close
- Markets
- Liquid ETFs · Index futures · Large caps
The rule set
- Entry: buy when SMA(5) crosses above SMA(10)
- Whipsaw meter: count how many times the two SMAs swapped order in the last 20 bars, and measure the 20-bar close-to-close range
- Clean regime (3 or fewer flips, or a range of 6%+): buy on the cross bar itself
- Moderate chop (4–5 flips inside a sub-6% range): buy only after SMA(5) holds above SMA(10) for 3 consecutive bars
- Severe chop (6+ flips inside a sub-6% range): require 5 consecutive bars instead
- Exit: sell as soon as SMA(5) crosses back below SMA(10) — exits are never delayed for confirmation
What makes it distinctive
- Two moving averages and one chart answer everything — the simplest trend system there is
- Measures its own worst enemy: a 20-bar whipsaw meter decides how much proof a golden cross needs
- Confirmation scales with the chop — 0 extra bars in a clean trend, 3 after moderate chop, 5 after severe chop
When it works
Markets that swing in multi-week directional legs — liquid ETFs, index futures and large caps with clean trend structure.
When it fails
Extended sideways markets are still its worst case: the filter cuts the whipsaw losses but cannot eliminate them, and the delayed entries give up part of every real move. Fast V-shaped reversals also hurt — the exit cross always lags the actual top.
How a decision moves through it
Input
Daily bars, about a year of them
Daily closes are all the calculations use. The averages themselves need only ten days of history; the whipsaw meter needs twenty.
Measure
SMA(5) and SMA(10)
One week of prices against two. The pair reacts within days of a swing turning, which is the appeal, and crosses on every hesitation, which is the problem the rest of the system exists to manage.
Measure
The whipsaw meter: flips and range over 20 bars
Two numbers describe the recent tape. Flip count: how many times the two averages swapped order in the last 20 bars — a trending stretch scores 0–1, sideways chop scores 4 and up. Range: how far the closes travelled top to bottom, as a fraction of the low.
Decide
Golden cross, with confirmation scaled to the chop
In a clean regime — 3 or fewer flips, or a 20-bar range of 6% or more — the cross bar itself is the buy. After moderate chop (4–5 flips inside a sub-6% range) the 5-day must hold above the 10-day for 3 consecutive bars first; after severe chop (6 or more flips, sub-6% range), 5 bars.
Act
Buy confirmed crosses; sell every death cross immediately
The exit is the mirror cross with no ladder attached. The bar SMA(5) closes back under SMA(10), the position is gone — whatever the meter says.
Fast crossovers have one famous disease
A 5-day average crossing a 10-day average is about the fastest trend signal that can be built from daily closes. In a market swinging through multi-week legs it is early to almost every one of them. In a market going sideways it is a machine for small losses: the two averages sit nearly on top of each other and swap order on every wobble, and each swap is a paid round trip.
The classical responses are to slow the averages down — which surrenders the earliness that was the point — or to bolt on a second indicator as a referee. This system does neither. It keeps the 5/10 pair and asks a narrower question before each buy: how badly has this exact signal been behaving lately?
The whipsaw meter, and the ladder it drives
Two numbers, both computed over the last 20 bars: how many times the two averages swapped order, and how wide the close-to-close range was. Many swaps inside a narrow range is the signature of chop — the signal has been firing without the market going anywhere. Swaps across a wide range are different: a market that travelled 6% or more was genuinely reversing, not oscillating, and its crosses were information.
- Clean regime — 3 or fewer flips in 20 bars, or a range of 6% or more: a fresh golden cross is bought on the cross bar, no questions asked.
- Moderate chop — 4–5 flips inside a sub-6% range: the 5-day must hold above the 10-day for 3 consecutive bars before the buy.
- Severe chop — 6 or more flips inside a sub-6% range: the hold requirement rises to 5 consecutive bars.
The confirmation bars are a toll on entry that is only charged where whipsaws live. In a trending tape the system is exactly the naive 5/10 cross; after a sideways mess it demands that the cross survive the very failure mode the mess just demonstrated.
Entries wait; exits never do
The death cross is not filtered. The bar the 5-day closes back under the 10-day, the position is sold — even if the meter reads severe chop, even if the cross looks like noise. The asymmetry is deliberate and it follows from what each delay costs. A delayed entry costs a slice of one move, and only when a real move is starting. A delayed exit costs a slice of every reversal, because every winning trade ends in an exit cross, and it buys almost nothing — the false exits it would filter are small, but the real ones it would slow down are where the money leaves.
Five ways into this system
- The cross, the meter, and the 0/3/5 ladder — every number stated exactlyFour numbers run the whole system: a 20-bar lookback, a 6% range threshold, and confirmation holds of 3 and 5 bars. Here is what each one does and why the exit side has none of them.7 min read
- Sizing a trade whose stop is a moving targetThe exit is wherever the averages next cross, which is not a price you can mark in advance. Sizing therefore starts from an estimate of that distance and a budget for losing streaks, not from a stop level.5 min read
- A daily-bar swing system for clean, liquid instrumentsThe system wants instruments whose trends come in clean multi-week legs and whose spreads do not tax a frequent trader. That describes index products and large caps, and very little else.5 min read
- What the filter fixes, what it merely discounts, and what it cannot seeThe whipsaw filter is a discount, not an immunity. The system still pays in long ranges, still gives up bars to confirmation, and still exits every top late — and the meter itself has a memory that misreads regime changes.6 min read
- The SMA 5/10 crossover explained from zero: a fast signal that learned to doubt itselfOne idea carries this whole page: a signal that has been wrong a lot lately should be made to prove itself, and a signal that says 'get out' should never have to.6 min read
The ideas behind it
This system assumes you already know these. Each one is explained from scratch in Investing 101.
Compare with
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.