Lessons.
Trading concepts explained from measurements, not definitions. Every lesson opens with a real figure from our own scan of roughly 1,700 stocks, with the date and the sample size attached, and works outward from there.
Why the numbers come first
There is no shortage of pages explaining what RSI is. There is very little telling you how often it actually fires, across a real universe of stocks, on a stated date.
We run a scan across roughly 1,700 equities every day, so we can answer that second question. It turns out to change the first one: a threshold that triggers on under one percent of stocks means something quite different from a threshold you were told to trade around.
Every figure is published with the date it was measured and the number of stocks it came from. A statistic without a sample size is an assertion, and this section would rather be checkable than confident.
What a lesson contains
The same four things, in the same order. A measurement taken from our own scan on a stated date. The number of stocks it was computed across, which is usually not the whole universe and is never rounded up to look tidier. The assumption sitting underneath it, named out loud. Then the concept the number is actually about.
The sample size matters more than it sounds. A stock needs roughly two hundred sessions of history before it has a 200-day average at all, so a figure about long-term trend covers fewer names than the scan does. Reporting it against the full universe would make the percentage look different and would be wrong. So each figure carries the sample it came from, and where two figures on one page have different denominators, the page says why.
Every published figure is archived
When a lesson quotes a number, the scan that produced it is saved to a dated file and committed alongside the page. That sounds like bookkeeping and it is, but it is the difference between a statistic you can check and one you have to take on faith.
It also means we can find our own mistakes. Two figures we published for the same day once disagreed by a tenth of a point, and the reason turned out to be that they were not the same measurement: one used an exponential average and the other a simple one, over slightly different lists of stocks. The pages were corrected on the page rather than quietly, and the snapshot tool now captures both numbers together so it cannot happen the same way twice.
Two a week, and no more
There is a version of this section that publishes forty pages a month by asking a model to explain every indicator in turn. It would rank for nothing, and a run of near-identical educational pages is one of the clearer signals that a site is producing content for search engines rather than readers.
So the cap is two lessons a week, and it is a cap rather than a target. A week with one good measurement in it gets one lesson.
What this section is not
It is not a definitions site. Pages explaining what an indicator is already exist in their thousands, written by people who have been at it for twenty years, and adding another would help nobody. It is not signals, and it never tells you what to trade. Nothing here is advice.
What it is, is the one thing a site with a daily scan across roughly seventeen hundred stocks can offer that a general reference cannot: what the market actually looked like on a particular Thursday, and what that does to a rule you have probably been told to follow.
Lessons are educational. They describe what we measured and do not offer advice, recommendations or forecasts. See the risk disclosure.