What to expect
A method that cuts its losses short and lets gains run doesn't look like a straight line upward. This page describes the shape it does have, using made-up numbers to show the arithmetic, and then a real example.
Winning less often than losing, and still paying
Swing trading signals are mostly entries at a turning point, with a stop close by. A close stop is stopped out often. A move that works can be several times larger than the stop distance.
That combination means a method can win a minority of its trades and still come out ahead. Here's the arithmetic with hypothetical numbers. Suppose a method makes 10 trades, each risking $2.00 a share. Three of them make $6.00 a share, a total of +$18. The other seven are stopped out at -$2.00, a total of -$14. Net result: +$4.00 a share across the ten trades, from winning only three in ten.
Change one number and the picture reverses. If the winners made only $2.00 each, the same three winners would give +$6 against -$14: a net loss of $8. What matters is not how often the method wins but how large the winners are compared with the losers.
The actual proportions for the real signals will be shown from our backtests. This page doesn't state them, because a figure belongs next to the evidence that supports it.
Profit factor: what the winners pay for
Profit factor is a way to put that comparison in one number. It's total gains divided by total losses.
In the hypothetical example, gains are $18 and losses are $14, so the profit factor is 18 ÷ 14, about 1.3. A profit factor of 1.0 is break-even. Above 1.0, the winners pay for the losers and leave something over. Below 1.0, they don't.
It isn't a promise of anything. It's a way to read a list of trades and ask: did the winners pay for the stops?
Losing streaks
If a method wins three trades in ten, then by chance alone it will sometimes lose several in a row. With made-up odds of 70% that any one trade loses, a run of five losses in a row has about a 17% chance of starting from any given trade. Over a hundred trades, the longest losing run typically falls somewhere between seven and twelve.
Nothing is wrong when that happens. It's what the odds produce. The risk per trade from page 6 is what keeps those runs survivable.
Long quiet stretches
Markets spend long periods moving sideways in narrow ranges. In those stretches, swings are too small to pay for their stops, and the method makes little progress. Marc Rivalland describes the swing trader's account as a staircase with long treads. The risers come when a market moves.
Patience is part of the method. Taking extra risk to fill a quiet stretch is a different strategy.
A worked example: Tesla (TSLA)
Below, every trade the method would have taken in Tesla (TSLA) over the last two years, with the entry, the stop, any stop move, the exit and the result in dollars a share. The losers are there with the winners.
[Placeholder: the complete Tesla (TSLA) backtest report goes here. It's built after the algorithm's backtests are regenerated: a performance summary, then period-by-period charts with their trades.]
An example, not a forecast. Results vary.
Where this leads
The last page turns the method into a routine.
Next: Putting it to work.