Every rating we have published, its date, the price it was published against and what the shares have done since. Nothing is removed from this table. A rating that goes against us stays here with the ones that do not.
One to three trading days is not a track record. Every rating below was published in the last week of July 2026 and none has had time to be right or wrong about anything. This page exists from day one so that it cannot be started later, once the results are known — that is the entire point of it. Judge it in a year, not this month.
| Company | Published | Rating | Price then | Price 31 Jul | Change |
|---|---|---|---|---|---|
| TERTeradyne, Inc. | 30 Jul 2026Unlevered DCF v1.2 | OvervaluedModelled base case $177 | $324.97 | $367.69 | +13.1% |
| KLACKLA Corporation | 29 Jul 2026Unlevered DCF v1.2 | Under reviewSee note below | $169.07 | $182.82 | +8.1% |
| BEBloom Energy Corporation | 29 Jul 2026Unlevered DCF v1.2 | OvervaluedModelled base case $98 | $180.70 | $205.81 | +13.9% |
| AMKRAmkor Technology, Inc. | 29 Jul 2026Unlevered DCF v1.2 | Fairly valuedModelled base case $49 | $45.68 | $49.87 | +9.2% |
| APLDApplied Digital Corporation | 28 Jul 2026Stabilised NOI v1.2 | Fairly valuedModelled base case $27 | $27.00 | $27.39 | +1.4% |
Price then is the quote each report was written against, printed on that report. Price 31 Jul is the closing price on 31 July 2026, the last completed session before this page was last updated. Change is price movement only and excludes dividends. Ratings are not directional bets and a rating is not a price target — overvalued means our modelled fair value sits more than 20% below the market price on the day we published, and says nothing about what the shares will do next.
The KLAC report states a fair value of $110–155, while the interactive model published on the same page returns $71.5 under its base-case assumptions. Those two numbers cannot both be right, and the discrepancy is ours, not the market's. The rating is suspended until the report and its model are reconciled and the corrected figure is published with a dated note. We have left the row in the table rather than removing it.
Once a rating is published it stays on this page permanently, including after we change our mind and including when it has gone badly. Coverage we drop is marked as dropped, not deleted.
A published rating is never quietly restated. Where a rating changes, both the original and the change appear with their dates.
Prices are closing prices from the same source for every name, taken on a single date printed at the top of this page. We do not pick the measurement date per company.
A rating compares our modelled fair value to the price on the day. It is not a forecast of the share price and this table should not be read as a performance record of a strategy.
Ratings are revisited each quarter on results, and immediately on a material contract, a financing event or a price move of more than 25%. This table is refreshed at the same time.
Where we get something wrong, the correction appears on the report with its date and the rating here is updated to match. Suspended ratings stay visible while they are open.