The SMA 5/10 crossover explained from zero: a fast signal that learned to doubt itself
One 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.
Key takeaway
- The two lines are one week and two weeks of average prices; which is on top says which way the swing points
- A whipsaw is a buy signal that reverses immediately — the defining cost of fast signals, and what this system counts
- The system trusts sell signals instantly and makes buy signals wait after bad behaviour, and that asymmetry is the design
Two fast lines and what they say
Average the closing prices of the last five trading days and you get one number: what this week has been like. Average the last ten days and you get another: what the last two weeks have been like. Redraw both every day and you have two lines that hug the price closely — far more closely than the slow 200-day lines in the longer-term systems.
When the one-week line is above the two-week line, the last few days have been stronger than the days before them — the short-term swing points up. When it crosses below, the swing has turned down. Because the windows are so short, these crosses happen within days of a turn, not months. That speed is the entire attraction.
The catch: what a whipsaw is
Speed has a price. When the market drifts sideways, the two lines sit almost on top of each other, and tiny wobbles flip them back and forth. Each upward flip says buy; a few days later the downward flip says sell, slightly lower; and then it happens again. Each little round trip loses a little money. Traders call this a whipsaw, and it is the classic way fast crossover systems die — not one big loss, but forty paper cuts.
Here is the observation this system is built on: whipsaws cluster. A market that has just flipped the averages five times in a month is far more likely to serve up another false signal than a market that has been trending cleanly. The signal's recent behaviour predicts the signal's next mistake.
The fix: count the recent lies
So the system keeps score on itself. Every evening it looks at the last 20 trading days — about a month — and asks two questions. How many times did the two lines swap places? And how far did prices actually travel, top to bottom? Many swaps with real travel means genuine reversals: fine. Many swaps with almost no travel — less than 6% from the month's low to its high — means the signal has been crying wolf.
- If the last month was clean, a fresh buy signal is acted on the day it appears.
- If it was moderately choppy — four or five swaps in a narrow band — the buy signal must stay valid for 3 straight days before any money moves.
- If it was severely choppy — six or more swaps in a narrow band — the waiting period is 5 days.
Most false signals cannot survive the wait: the lines flip back within a day or two, no trade happens, and a whipsaw that would have cost money costs nothing. A real trend, by contrast, keeps the lines apart for weeks — losing the first three days of it is a real but affordable fee.
Why selling never waits
The waiting period applies only to buying. The moment the fast line drops back below the slow one, the position is sold — same day, no scorekeeping, no exceptions. At first glance that looks inconsistent. It is actually the most carefully reasoned rule in the system.
If entries were instant in chop
- Every false buy signal costs a real round-trip loss
- Losses arrive in streaks, exactly when the tape is worst
- The forty-paper-cuts problem, at full speed
Because exits are instant, always
- A wrong exit costs little — you rebuy on the next confirmed cross
- A right exit protects the whole trade's profit from the turn
- No winner is ever ridden all the way back down while 'confirming'
Waiting to buy risks missing a few days of gain. Waiting to sell risks giving back the trade. Those risks are not the same size, so the rules are not symmetric — the system doubts good news when good news has been unreliable, and it acts on bad news unconditionally. That principle outlives this particular pair of averages, and noticing it in other systems is a good sign you have understood this one.
Common questions
- Is this a good first system?
- It is one of the easier ones to run — daily closes, ten minutes, unambiguous rules — and one of the harder ones to sit with, because whipsaw streaks test discipline in a way slow systems do not. A beginner who wants crossover logic with far fewer decisions should look at the golden cross / death cross page first; it is the same idea at one-twentieth the speed.
- Do I need software to run the whipsaw meter?
- No. Both averages are drawn by any free charting site, and counting how many times they swapped order in the last month is done by looking at the chart. The 20-bar range is the month's highest close against its lowest. The arithmetic was chosen to be checkable by eye.
- What single habit matters most when running it?
- Acting on the close, once, and not looking during the day. Intraday, the two lines cross and uncross in ways that vanish by the close, and every peek is an invitation to take a trade the system never signalled. The evening routine is the system; the rest is noise.
The ideas behind it
This system assumes you already know these. Each one is explained from scratch in Investing 101.
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.