Modine Manufacturing initiation, July 2026
I published a 32-page initiation on Modine Manufacturing two days before the company reported its quarter.
Published July 27, 2026. Modine reported two days later, on July 29, 2026.
Modine is a share-catcher, and that is the thesis.
Data Centers grew 158 percent in the fourth quarter of fiscal 2026 against the same quarter a year before, and tripled its share of the company in two years, 12 percent of revenue to 35. It does not have to lead this market, only keep taking pieces of it.
The bear case mixes two sets of numbers.
The $88.4 million of corporate cost the bear case quotes is the GAAP number. On the adjusted basis every multiple in the note uses, it is $62.6 million, and I argued the two were being mixed.
Two things cut against me.
Data Centers margin fell through fiscal 2026 even as revenue doubled, 22.1 percent to 19.1, and gross margin gave up 650 basis points, which reads as share bought with price. And no insider bought a share on the open market in calendar 2026 while insiders sold $34.5 million.
Source: Eddy R. Lora, Modine Manufacturing initiation, July 2026, page 3, and the filings it cites.
I wrote the rule before the results
I wrote the rule into the note before it went out. If segment margin came in under 16.0 percent for two quarters in a row, the call would change.
What the first quarter showed
Modine reported on July 29, 2026. Data Centers grew 89.8 percent on the release's own figures, above the top of the range the note set for the quarter.
The margin the rule watches came in at 15.29 percent, the first quarter below 16.0 in the history Modine re-issued on July 24, five days before it reported. The rule says two quarters in a row, and it says two on purpose, so the call held.
Modine reports the second quarter in late October or early November 2026, my estimate from its last four second-quarter dates. That one decides the rule.
Current through Modine's first-quarter earnings.
Note: 15.29% is Data Centers and Commercial HVAC adjusted EBITDA, $51.7 million and $41.6 million, over the same two segments' net sales, $348.6 million and $261.6 million, first quarter of fiscal 2027.
How the model works
I built the model in Excel, by hand and with Python. Python brings Modine's filings into the history, so every historical cell is a filed number or a dated assumption, and the forecast and the valuation I build myself. Before the note goes out, the workbook recalculates and every number the note states is checked against a live cell.
After the first quarter I rolled the model forward and logged each change with what it moved further down the model. Nine of nine reported lines matched within $0.05 million, the limit I set.
Four of the changes to the model after the first quarter. The margin assumption the call rests on stayed where it was, because the rule says two quarters, not one.
| What | Old | New | Why |
|---|---|---|---|
| Corporate cost in the high and bear cases | The GAAP $88.4 million set against adjusted numbers | $62.6 million, from the figures Modine re-issued on July 24 | The same mix I argued the bear case made. The note said so in a source line, and the first update fixed it. |
| The margin the rule watches | Worked out by hand every time | A column in the model that computes 15.29 for the quarter and matches the hand figure to four decimals | So the hand figure and the model always agree. |
| Data Centers growth for the quarter | A growth rate built from two different sets of segment figures | Removed, with the reason written in the cell | Better blank than wrong. |
| The full-year check for fiscal 2027 | Compared one quarter to nothing and showed a difference that was not real | Switched off, with its reason written in the cell | So a skipped check can't read as a pass. |
Source: Eddy R. Lora, model change ledger, first quarter of fiscal 2027.
Read the note
Source: Eddy R. Lora, Modine Manufacturing initiation, July 2026. Each page opens full size.
Ask and I'll send you the second quarter when Modine reports it.
This page shows a research writing sample I published on my own on July 27, 2026. It is not investment advice and not a solicitation. I am not a registered broker-dealer or a registered research analyst. I used software, including Claude, to gather data and check figures. The thesis, the risks and the finding are my own work and judgment.