Learn
On reading charts, on how these questions are built, and on the ways practice misleads you. Everything here is written from what we can actually measure — the mechanics are on the methodology page.
practice methods8 min read
Pre-market checklists gather inputs and settle nerves. None has you make a single practice call. We built that step, and the arithmetic decided what it must not do.
methodology8 min read
We have written eight times that the transfer question is open, and never once what would close it. Here is the protocol, the arithmetic, and why we cannot run it.
research7 min read
Two surveys found technical analysis in heavy use among FX dealers and fund managers. At least some universities teach it. The same notes say why a rule fades.
research9 min read
Technical analysis has been tested seriously for thirty years. The reading worth doing is the history of how it was tested, and which results survived out of sample.
methodology7 min read
If you are right 60% of the time, it takes about 194 calls to show that is not luck. At 55% it takes 783. Here is the arithmetic, and why your own record is harder.
price action7 min read
Across 1.68 million bars the opening half hour moves 2.5 times as far as mid-afternoon. The share of moves that go up barely changes all day.
methodology8 min read
We removed a statistical dependence from our question bank. The removal was itself a sampling rule, and it turned a coin flip into 63%. We misdiagnosed it twice.
methodology9 min read
We tested five confirmation rules on 1.13 million bars. All five passed at p below 0.0001. All five were worth about half a percentage point.
price action8 min read
Pattern success rates are almost always scored against 50%. On 1.37 million intraday bars the baseline is 51.73%, which turns several reported edges into deficits.
practice methods9 min read
A trade journal records the setups you were confident enough to enter. That makes it a selected sample, and a selected sample cannot grade the reads you never acted on.
practice methods7 min read
Blocked practice tells you what kind of problem you are facing. Interleaved practice does not — and that is the harder format the research supports.
practice methods7 min read
A five-answer streak is weak evidence: a player choosing at random produces one in 81% of 100-answer runs. We published that maths. The streak counter still lights up.
practice methods9 min read
Is SwipeTA Duolingo for trading? The mechanics look alike, but chart-reading practice has one problem language apps do not: the answer key is noisy.
comparison11 min read
SwipeTA vs TradingView Bar Replay: replay is better for rehearsing a strategy; a question bank can enforce blind selection, balanced outcomes and pre-committed calls.
intuition8 min read
Can revenge trading and trading FOMO be measured? Attention research, Isaac Newton's South Sea mistake, and a tilt metric built on what happens right after a wrong call.
methodology7 min read
The horizon is not a setting on the question. It is the question. We swept 140,319 decision points across five of them to find out what each one actually asks.
price action8 min read
Opening range breakout statistics from 20,650 tested breakouts on US large caps: direction, false-breakout rates by ATR distance, and a base rate to judge them by.
chart reading7 min read
"Tape feel" has a plainer explanation than a sixth sense: fast pattern recognition. The research supports that — and is unusually clear about what it does not.
chart reading8 min read
Cropping the candles at the decision point closes one of them. The other three are a derived indicator, the record the question travels in, and the next question.
chart reading6 min read
Fading the last 5-minute candle is right 60.67% of the time in our question bank and 49.78% across every bar. We traced the gap, and it was not where we said it was.
chart reading4 min read
They get used as synonyms for "practice without risking money", but each one exercises a different skill — and picking the wrong one wastes the hours you put in.
New pieces are announced in the blog’s RSS feed. Corrections to support@swipeta.net — if something here is wrong we would rather know.
