We have now published 71 company reports. Each ends with a score out of ten covering the business, its moat, its management, its finances and its price. That is enough to stop guessing and start counting.
Nine reports scored 8.0 or above. Fourteen scored below 6.0. The bulk — twenty-nine of them — landed between 7.0 and 7.9.
That distribution is itself worth sitting with. If you read enough market commentary you would conclude that outstanding companies are everywhere. On our own scoring, roughly one in eight clears eight out of ten.
Seventy-one scores, sorted
| Score band | Reports | Share |
|---|---|---|
| 8.0 and above | 9 | 13% |
| 7.0 – 7.9 | 29 | 41% |
| 6.0 – 6.9 | 19 | 27% |
| Below 6.0 | 14 | 20% |
Note the tail at the bottom. One in five of the businesses we examined scored below six — and we did not choose them at random. These are mostly large, famous companies that somebody asked us to look at, or that we thought would be interesting. The selection is biased towards quality, and a fifth of it still came out weak.
The variation is inside the sectors, not between them
This is the number that should change how you allocate
| Sector | Reports | Average | Lowest → highest |
|---|---|---|---|
| Communication Services | 4 | 7.90 | 7.6 → 8.0 |
| Financial Services | 11 | 7.23 | 6.2 → 8.3 |
| Technology | 13 | 7.19 | 5.2 → 8.3 |
| Consumer Defensive | 9 | 6.74 | 5.4 → 8.0 |
| Consumer Cyclical | 9 | 6.51 | 4.2 → 8.1 |
| Healthcare | 8 | 6.48 | 4.8 → 7.8 |
| Real Estate | 9 | 6.46 | 4.2 → 8.2 |
| Industrials | 5 | 5.86 | 5.0 → 6.9 |
Read the last column, not the middle one. The distance between the best and worst sector average is 2.0 points, from Industrials at 5.86 to Communication Services at 7.90.
The distance inside Real Estate is 4.0 points — it contains both our second-highest-scoring company and our lowest-scoring one. Inside Consumer Cyclical it is 3.9. Inside Technology, 3.1.
The gap between two companies in the same sector is roughly twice the gap between sectors. That is not a subtle effect, and it comes out of our own data without any theory attached.
Real estate contains both our second-best business and our worst. The sector label told you almost nothing.
Three consequences, in order of usefulness
- "I'm bullish on sector X" is a weak statement. Knowing the sector narrows your expected quality by about two points. Knowing the company narrows it by four. If you are going to do work, do it one company down.
- A weak sector is not a reason to skip it. Our lowest-average sector still contains businesses we rate near seven. Industrials averaging 5.86 does not mean industrials are uninvestable — it means most of them are ordinary and a few are not, which is true everywhere.
- The tail is where the money is lost. One in five scored below six. Avoiding that fifth does more for a portfolio than finding the next eight, because the arithmetic of losses is unforgiving in a way the arithmetic of gains is not.
These are quality and price judgements, not forecasts, and they are ours. A 71-company sample chosen by interest is not the market, and a score of 8.2 is not a prediction that the shares go up. Anyone claiming to diagnose an index from seventy-one hand-picked reports is overreaching, and we would rather say so than let the chart imply otherwise.
What the sample does support is the dispersion finding, because that is a statement about the spread of our judgements rather than their level — and spread is far less sensitive to how the sample was chosen.
The two ends of the same sector
The clearest illustration sits inside real estate. One landlord owns the warehouses between the port and your door, with rents below market and the leases to prove it. Another owns properties let to tenants in an industry where a quarter of the rent roll has been in default, and its shares are a long way below their high.
Both are REITs. Both pay dividends. An allocation model that buys "real estate" buys them in proportion to their market value, which is to say it buys more of whichever the market currently likes. A reader who looks at the businesses buys one of them, or neither.
Our second-highest score and our lowest — from the same sector, on the same board.
What to do with this on Monday
If your portfolio is built from sector weights, this data says you are working at the wrong level of resolution — the thing you are choosing explains about half of what the thing you are not choosing would.
And if you own something you have never scored, score it. Not precisely: out of ten, honestly, on the business, the moat, the management, the balance sheet and the price. The exercise takes an evening per company, and its main value is not the number. It is discovering which of your holdings you cannot defend for more than two sentences.
Frequently asked
How many companies score highly in your research?
Nine of 71 reports scored 8.0 or above, and 14 scored below 6.0. The bulk — 29 reports — sat between 7.0 and 7.9. That shape is the honest summary of what serious analysis produces: most good companies are good rather than exceptional, and the exceptional ones are rare enough to count on your fingers.
Which sectors score best?
On our own averages: Communication Services 7.90 across four reports, Financial Services 7.23 across eleven, Technology 7.19 across thirteen. The bottom: Industrials 5.86, Real Estate 6.46, Healthcare 6.48. But the sector average is the least useful number in this article — see the dispersion finding below.
Does sector allocation work?
Our data argues against relying on it. The range between our sector averages is about two points, from 5.86 to 7.90. The range inside a single sector reaches four points — Real Estate contains both our second-highest-scoring company and our lowest. Choosing the sector gets you roughly half the information that choosing the company does.
Is the market expensive right now?
That is not a question 71 individual reports can answer, and we would distrust anyone who claimed otherwise from this data. What they can say is that quality is unevenly distributed and unevenly priced — which is an argument for looking at businesses one at a time rather than for a view on the index.