how do you actually develop your own strategy instead of just copying one

[ghostknight]

every strategy i trade is someone elses, pulled from a video or a forum, and i never fully trust it because i dont know why it works. i want to develop my own so i actually understand and believe in it. but i have no idea how you go from random market observations to an actual tested strategy. whats the real process for building one from scratch?

[Dean]

it starts with an observation, not a strategy. spend real screen time noticing one repeatable thing, like price tends to react a certain way at a certain level under certain conditions. then form a precise hypothesis, write exact rules for entry, stop and exit, and test it on history honestly with costs. most of your ideas will fail testing, and thats the process working. a strategy you built and watched fail and refine is one you actually understand and trust, unlike a borrowed one.

[Jason]

the observation-first approach is exactly it, and the key is that it has to be your observation from your screen time, not a claim you read. when you noticed the pattern yourself, you understand its context and limits in a way no borrowed strategy gives you. then the testing tells you if your eyes were right. that loop, observe, hypothesise, test, refine, is how every real strategy gets built and why it earns your trust.

[Brandon Lee]

the romantic build-your-own journey usually ends with people reinventing a moving average crossover and feeling proud. theres nothing wrong with trading a well-known edge you understand, the obsession with a unique homemade strategy is ego. understanding beats originality, and most originality is just an inefficient rediscovery of the known.

[mattlive]

the troll makes a genuinely useful correction, the goal isnt originality, its understanding and trust. its completely fine if your developed strategy turns out to be a known edge, what matters is that you arrived at it through your own observation and testing, so you understand why it works and when it wont. you can absolutely start from a known strategy and make it yours by testing it yourself and learning its behaviour deeply. understood-and-borrowed beats original-and-mysterious. the work is the understanding, not the novelty.

[Liam Parker]

how long does this whole observe-hypothesise-test loop realistically take before you have something you can trade, weeks or months?

[wildnorth]

realistically months, and thats normal, not slow. the observation phase alone needs enough screen time to see the pattern recur across conditions, then testing and refinement take more. anyone promising a tradeable personal strategy in a week is selling something. but the months arent wasted, theyre where the understanding and trust get built. a strategy you developed over months you will hold through drawdowns that would make you abandon a borrowed one overnight. the time is the point.

[lowkeysam]

observe, hypothesise, test, refine. understanding over originality. takes months.

[Chris]

one practical tip for the development loop, keep a separate research journal distinct from your trading journal. log observations, hypotheses, and the results of each test. most ideas die, and the log stops you re-testing the same dead idea months later or forgetting a promising thread. building strategies is mostly bookkeeping of failed and surviving ideas, the research journal is what turns scattered observation into compounding knowledge.

[midnightzone]

this reframed everything for me i thought building a strategy meant inventing something nobody had, which felt impossible. observe my own screen, test it, understand it, even if it ends up being something known. that feels actually doable. starting a research journal today.

[Nathaniel]

the research journal will be full of brilliant observations that all evaporate the moment theyre tested with real costs. thats not discouragement, thats just the hit rate. budget for ninety percent of your clever ideas being garbage and youll be pleasantly calibrated.

[coldvision]

the ninety-percent-die rate is real and worth accepting upfront so you dont quit when your first five ideas fail. thats not failure, thats the filter doing its job. the one idea in ten that survives honest testing is worth the nine that didnt, because it survived precisely the scrutiny that will keep you trading it through hard times. expect the graveyard, its where the real strategy comes from.