The three pillars
Each ring on a stock row is one pillar. This is what each can see, and what it cannot.
Insider conviction
What it measures. Whether people inside the company bought shares with their own money on the open market in the last twelve months, how much, how many of them, and how recently.
Why it carries the most weight. An executive buying with personal money is the hardest of the three signals to fake. There is no press release version of it — the filing is a matter of record and the money is real.
What it cannot see. Most companies have no insider purchases in any given window. Absence is the normal state, so we treat it as Unknown rather than as bad news. It also cannot see why someone bought; conviction and a scheduled diversification look identical in a filing.
Selling tells you almost nothing. Insiders sell for a hundred reasons — vesting, tax, a house, spreading risk — and buy for one. Sell-only activity scores a flat 20, deliberately crude, because a finer scale would suggest a precision that is not there.
Analyst valuation
What it measures. The published consensus rating, and how cheap the stock looks against next year's expected earnings.
What it cannot see. The stocks worth finding are the ones nobody covers, which is exactly where this pillar is weakest. One or two analysts is barely a consensus, so a thin rating is downweighted and marked with an amber dot. Where there is no coverage at all the pillar leans entirely on the forward P/E, and where there is neither, it goes Unknown.
A target price without an analyst is not a target price. The provider occasionally returns one with no rating behind it. We drop it, so that an unsourced number cannot end up carrying a trade setup.
Market behaviour
What it measures. Tradability — average daily volume as a share of market value. Can you get in and out of this position without moving the price yourself?
What it explicitly does not measure. Direction. A stock near its 52-week low can score well here. Whether that is a bargain or a falling knife is a judgement this pillar does not make.
Position against the 200-day average and within the 52-week range appear in the written explanation, because they are useful context, but they do not feed the number.
Reading the rings
- Fill is the strength, 0 to 100.
- Colour is the tier, using the same thresholds everywhere on the site.
- Dashed and empty means no data — which is not the same as zero, and does not count against the score.
- An amber dot means the evidence behind that pillar is thin.
The large ring with the number in it is the composite. Beside it, always, is how many of the three it was able to use.