How the test works
Go back one year. Score every stock with only what was public then. Then look at what happened.
For each of 976 of the 1127 stocks on the site, the score was rebuilt from the accounts, quarterly results and share price that were public on the day its older annual report came out (FY2023-FY2025). Nothing from the following year was used. The rules are today's rules.
Then the next audited year was checked for three things that hurt a shareholder: a new loss, a dividend cut, and a big share issue. The test reads business results only. It never looks at the share price after the test date.
What happened next, by last year's score
The lower the score a year ago, the more often things went wrong. All three checks point the same way.
New loss
Made a profit in the test year, a loss the next year. Counted only for stocks that were making a profit.
Dividend cut
Dividend per share fell more than 10% the next year. Counted only for stocks that paid one.
Big share issue
Share count rose more than 10% in a year (placements, rights issues, warrants). Splits and bonus issues are left out.
Strong means 80 or more, Weak means under 40. Only 75 stocks were Strong a year ago, and 299 were Weak.
Dividend cuts were common in every group, Strong included (14 of 40 (35%)). A high score said the dividend was easier to afford. It did not say it would grow.
Which pillar warned about a new loss
Financial Safety hardly told the two groups apart: 8% against 10%.
Business Strength
Financial Safety
Cash & Dividends
Price Fairness
Risk
Risk split them best: a new loss in 3% of stocks it rated Strong, 21% of those it rated Weak.
Why would Financial Safety miss? It measures debt and cash buffers: how well a company could live through a bad year. It does not try to say whether a bad year is coming. A company with little debt can still slip into a loss. So read a high Safety score as “can take a hit”, not “won’t take one”. The score itself was not changed for this article.
What this test can't tell you
- One year only. A single year can be unusual.
- Today's rules were set knowing recent results, so the test flatters the score a little (hindsight).
- Small groups give jumpy percentages, so every number is shown as “X of Y”.
- Dividends are counted by calendar year, so a payment that moved by a few weeks can look like a cut.
- Not a prediction and not advice. A Strong score can still be followed by a loss.
5 share-count jumps were left out because a split or bonus issue fell in the same year.
FAQ
In this test, yes. Of the profit-making stocks that scored 80+ a year earlier, 2 of 75 (3%) made a loss the next year. Okay: 47 of 566 (8%). Weak: 20 of 93 (22%). That is one year of history, not a promise.
Financial Safety. Profit-making stocks it rated Strong turned to a loss 8% of the time, against 10% for the ones it rated Weak. Risk split them best: 3% against 21%.
No. It checks business results only: profit, dividends and share count. It says nothing about whether a share price went up or down.
Each of 976 stocks was scored on the accounts, quarters and share price that were public when its older annual report came out, under today's rules. Then the next audited year was checked for a new loss, a dividend cut of more than 10%, and a share count up more than 10% (splits and bonus issues left out).
Once a year, after the next full year of audited results. Each run is saved as its own edition and never changed. This one was run on 18 Oct 2026.
Next read
Bursa Malaysia's Market Structure, Explained
1,130 companies, 3 markets, the KLCI's top 30, and why banks run the show.
Educational content, not investment advice. Past results don’t promise future ones. See About for how this site works.