Common Advice That Turned Out to Be Wrong
Most bad advice in this area is not fabricated. It is a true observation that lost its context somewhere along the way, got repeated by people who had not tested it, and became something a beginner is expected to accept. What follows are four examples that took an embarrassingly long time to question, held on to mostly because disagreeing with something so widely stated felt like arrogance rather than observation.
Trade Every Signal Because You Cannot Know Which Ones Work
This is presented as discipline and it sounds unarguable. If you cannot tell in advance which setups will succeed, filtering is just an invitation to cherry pick, and cherry picking destroys any edge the rules had.
The flaw is in the premise. You genuinely cannot judge whether an individual break will hold, but you can absolutely judge whether the conditions around it are the conditions the rules were designed for. A range far outside its normal height, or a session where a major release lands moments after the window closes, is not an unknowable coin flip. It is a known bad case. Treating measurable conditions as unknowable because individual outcomes are unknowable conflates two different things, and the cost of that confusion is a long run of trades taken in circumstances that never suited the approach.
Move the Stop to Break Even as Soon as You Can
The appeal is emotional and it is understandable. A trade that cannot lose is a comfortable trade, and the phrase free trade does a lot of work in making this sound like risk management.
In practice it converts a decent number of eventual winners into scratches, because normal movement after a break routinely revisits the entry before continuing. The stop was placed where it was for a reason having to do with the structure of the range, and moving it to a price with no structural meaning at all, chosen purely because it is where you happened to get in, replaces a considered level with an arbitrary one. It reduces anxiety and it reduces results, and the first of those is why it survives.
Backtest Until You Are Confident
Testing is worth doing and the advice is not wrong so much as incomplete. What goes unsaid is that confidence is not the output being sought, and that a long enough search through parameter combinations will always produce something that looks excellent on the data it was fitted to.
The more serious problem is that the test cannot include the part that actually fails. A record of what the rules would have done contains none of the hesitation, none of the skipped mornings, none of the sessions where you would have overridden the entry. Confidence built on that record is confidence in a version of yourself who does not exist, and the first divergence between the two is where the trouble starts.
Find the Instrument That Suits the Strategy
This one sounds sensible and consumed a great deal of time. The search for an instrument whose ranges behave more cleanly is genuinely endless, because there are always more instruments and any of them will look promising over a short enough window.
What was missed is that switching resets everything you had learned. Knowing what a normal range looks like on one instrument takes months of watching, and that knowledge does not transfer. Someone who has watched a single instrument for a year has a sense of when today is unusual that no amount of scanning across many can supply. The instrument mattered less than the familiarity, and the search for a better one was mostly a way of avoiding the slower work.
Why This Advice Persists
Each of these is memorable, easy to state, and comforting in a specific way. They sound like discipline while asking for very little, and none of them can be disproved by any individual session, so the person repeating them is never confronted with being wrong. The only correction available is your own record, kept long enough and honestly enough that it can disagree with something everybody says.