How this site is made
Every page here is written by Michael Whitman, carries at least one figure traceable to a named research file, and includes a section on what the idea looks like when it fails. Pages are checked by an automated gate before publication and none ships without passing it.
The short version
Every page is written by me, carries at least one figure that can be traced back to a named file, and says what the idea looks like when it fails. A script checks each page before it publishes, and nothing ships that doesn’t pass.
Where the figures come from
Most sites in this space prove their trustworthiness with process — a page describing how carefully they work. I’d rather show the numbers.
- A study of 24,971 trading videos, with view counts, lengths and topic coverage. When a page says how many videos cover a topic or how the median one performed, that is where it comes from.
- My own channel analytics — around 400 published videos, with real view, retention and click-through data.
- Forward-tested screens — picks recorded before the outcome was known, not selected afterwards.
The published datasets, with downloads and the headline figures measured from each, are on the research page.
The rule I hold myself to: a figure ships only if it can be recomputed from the file it came from. Not “sourced from” — recomputed. If I can’t reproduce the number, it doesn’t go on the page. That rule exists because a plausible, on-topic, wrong figure made it into an early draft of this site, and the fix was a rule rather than an apology.
The drawn series, and why its numbers are not quoted
The illustrative charts on this site are drawn from one constructed price series. It is not real market data, and anything measured on it — how wide its bars are, how long its runs last, how deep its dips go — describes an invention rather than a market. Earlier versions of some pages quoted those measurements as though they were evidence. They no longer do, because a figure taken from invented data proves nothing about any instrument a reader could trade.
Where a page needs a number about how markets actually behave, it uses real prices. The main source is the S&P 500’s daily closing level since January 1950, price only, without dividends, with the 16 Sep 1929 close of 31.86 carried in as the record until the index passed it on 22 Sep 1954:
| Measurement (S&P 500 daily closes, 1950 to September 2026) | Value |
|---|---|
| Trading days that closed below the prior record close | 93% |
| Longest wait for a new record | 7.5 years, from 11 Jan 1973 to 17 Jul 1980 |
| Worst fall from a record | 56.8%, from 1,565.15 on 9 Oct 2007 to 676.53 on 9 Mar 2009 |
| First close back above the 2007 peak | 28 Mar 2013 |
| Days that closed higher than the day before | 53.1% |
| Ten-day windows that finished higher than they started | 59% |
| Separate stretches that closed at least 5% below the record | 68, about 0.89 a year |
What these figures are good for: showing the ordinary shape of a real market — that most days are spent below a previous high, that a single day’s direction is close to a coin toss, and that a 5% pullback arrives about once a year.
What they are not good for: predicting anything, or describing a different instrument. Where a page makes a point about sizes — a stop, a spread, a position — and has no real data for it, it uses a worked example that is labeled as hypothetical and whose arithmetic can be checked on the page.
Where the charts come from
The charts on this site are drawn, not screenshotted. They’re generated by a chart engine I built, and they show constructed price action chosen to demonstrate one idea clearly.
That is a deliberate choice, and it is labeled on every chart: illustrative — not real market data. A real chart of a real instrument shows one thing that happened once. A drawn chart can show the pattern and the way it fails, side by side, without either being cherry-picked from history.
Where a page needs a real instrument, it says so and cites the source.
How pages are checked
Each page runs through an automated gate before publication. It checks structure — that the answer comes first, that the sections are present, that internal links resolve — and it checks honesty: no performance claims, no urgency language, no guarantees, and no phrase from a banned list built from the FTC’s own published notices about this industry.
A page that fails does not publish. The gate cannot be argued with, which is the point of having it rather than a checklist.
On AI
I use AI tools to help research and draft, the same way I use a spreadsheet or a charting platform. Every page is written to my own standard, carries my byline, and reflects what I actually think — including the parts where I disagree with the consensus in this niche.
I’m accountable for everything here. If a page is wrong, it’s my fault, and telling me in a comment is the fastest way to get it fixed.
How pages are reviewed, corrected and updated is set out in full in the editorial policy, and every factual correction is listed, with its date, in the corrections log.
What I will not publish
- Win rates, profit figures, or account returns. There is no honest way to present them.
- Testimonials or results claims from anyone.
- Guest posts or sponsored articles. Nobody pays to appear on this site.
- Anything I haven’t checked, including my own older pages when something changes.
This page is educational, not financial advice. Test every idea on your own charts before risking money.