How the Maz Rating Is Built: 80 Ratios, Five Factors, One Sector-Relative Score

Every stock on Maz Vantage carries a Maz Rating, a number between 1 and 5 with a verdict next to it. This article explains exactly how that number is produced: which ratios go in, how they are compared, how they are combined, and what the rating does not try to do. It describes the method; it is not investment advice.

Key takeaways

• The Maz Rating is relative to the sector. A 4.5 means a company ranks near the top of its own sector on the measures below, not that it is a good investment in absolute terms.
• It is built from 80 financial ratios in five factors: Valuation (18), Growth (17), Profitability (21), Financial Health (20) and Momentum (4).
• Each ratio is ranked against the other companies in the same sector. Ratios are combined into factors, factors into one composite, and the result is re-ranked at each step.
• Momentum counts double in the composite. The other four factors count once each.
• The final score is 1 + 4 × the composite percentile, so it always sits between 1 and 5. Verdicts: Strong Buy (4 and above), Buy (3 to 4), Hold (2 to 3), Sell (below 2).
• Ratings are recalculated once a month. Fair-value estimates, the Beneish M-Score, the Alpha Signal and technical signals are shown on the site but never enter the rating.

Step 1: which stocks, which sector

The rating covers common stocks only. Funds, warrants, units, rights, preferred shares and listed bonds are left out. Each company is compared with the other companies in its sector, using the sector its data provider assigns.

A sector needs enough members for a ranking to mean something. Below 20 usable companies, the calculation still runs but the percentiles are coarse, and the rating should be read with that in mind.

Step 2: 80 ratios, in five factors

Each ratio has its own weight inside its factor. In Valuation, for example, the forward PEG ratio carries the most weight; in Profitability it is operating cash flow. Each ratio also has a direction: for a P/E, lower ranks better; for a margin, higher ranks better.

Step 3: every ratio is ranked against the sector

A ratio on its own says little. A P/E of 25 is expensive for a utility and cheap for a software company. So each ratio is turned into a rank within the sector:

1. Extreme values are capped at the sector's 1st and 99th percentiles, so one company with a P/E of 4,000 does not distort the scale for everyone else.
2. The direction is applied, so that a higher rank always means better.
3. The company gets a percentile: where it sits among its peers, from close to 0 (bottom) to close to 1 (top).

When a ratio cannot be calculated for a company (a P/E when earnings are negative, for example), it is simply left out for that company. The other ratios in the factor share its weight. There is no penalty for a missing value.

Step 4: sector adjustments

Some ratios mislead in some sectors, so the method adapts:

• Dropped: a ratio that makes no sense for a sector is removed. Enterprise-value multiples and the Altman Z-score are not used for banks and insurers, whose balance sheets work differently.
• Turned down: a ratio that is only partly meaningful keeps a quarter of its weight. Leverage ratios for financials and asset turnover for energy companies are examples.
• Turned up: a ratio that matters more in a sector gets a heavier weight. Dividend yield and dividend growth count more for utilities, working capital and capital spending more for industrials.

The weights have been checked against historical data in six sectors: Technology, Financials, Industrials, Utilities, Consumer Cyclical and Consumer Defensive. In Energy and Materials the check was only directional. In Healthcare, Real Estate and Communication Services the weights are the defaults, so ratings there should be read as indicative. For real estate investment trusts in particular, EV multiples and P/E are distorted until funds-from-operations measures are added.

Step 5: from ratios to factors to one score

The ratios are combined twice:

1. Ratios → factors. For each factor, the company's ratio percentiles are averaged using the ratio weights. That average is then ranked again against the sector.
2. Factors → composite. The five factor percentiles are averaged, with Momentum counting double, and the result is ranked again against the sector.

Why rank again instead of just averaging? Averages of percentiles bunch up in the middle: almost nobody is top of the sector on every ratio, so plain averages would squeeze most companies between 2.5 and 3.5. Re-ranking spreads the results back across the whole scale, so within each sector the ratings are spread roughly evenly from 1 to 5.

This is also why the overall rating is not the average of the five factor scores shown on a stock page. Take an illustrative company with these factor scores:

Their simple average is about 3.3, yet its overall rating is 4.02, a Strong Buy. Its weighted composite (with Momentum counting double) ranked in the top quarter of its sector, and the rating reflects that rank, not the raw average.

Step 6: the score and the verdict

The score is 1 + 4 × the composite percentile. The same formula gives the score of each factor and each ratio you see on a stock page.

Because a percentile is never exactly zero, the score is always a little above 1. That is why a Strong Sell never appears.

Every month, each sector is scored again and the new rating is saved alongside the old ones, so a stock's rating history builds up month by month.

What the rating is not

• It is not absolute. It ranks a company against its own sector. The best-ranked utility and the best-ranked software company can both be 4.8 for very different reasons.
• It can move without the company changing. If the rest of the sector improves, a company's rank can fall even when its own numbers stay the same.
• It leaves some things out on purpose. Fair-value models, the Beneish M-Score (an earnings-manipulation flag), the Alpha Signal and technical signals are shown on the site as separate tools. None of them feeds the rating.
• It depends on reported data. Ratios are calculated from published financial statements supplied by a market-data provider. Figures are not converted between currencies, and data errors flow through.
• It is not a prediction. A high rating says a company ranks well on these measures today. It does not say the share price will rise, and past ratings do not predict future returns.

Where to see it

On every stock page, the Ratings tab shows the overall rating, the five factor scores and the value of each ratio. With a free account you also see where the company ranks on each ratio within its sector.

---

Coming next: Maz Watch, an email the day a stock on your watchlist changes rating, with the factor that moved it, and a monthly brief of upgrades and downgrades. It is not open yet and nothing is charged. See what it includes and join the waitlist.