
Capex belongs as a cash outflow. Entered positive, it inflates free cash flow and every multiple that sits on top of it. Row 118 in the live file did not throw an Excel error. It looked industrious. It looked funded. It was wrong in the direction that flatters the deal.
Sign flips survive reviews because the absolute value is often believable. The unit cost matches the bid. The schedule matches the construction tab. Only the sign is inverted, which means the model adds where it should subtract. That is not a formatting issue. That is a cash-flow fiction.
Row 118 has capex entered positive. Flip the sign before this file is treated as a finished view of cash.
Positive numbers look like progress. Negative outflows look like problems. Under deadline pressure, people paste schedules as absolute values and trust a later step to apply the sign. The later step does not always run. The deck still prints.
Do not stop at the cell. Recompute free cash flow, debt draws, and any ratio that used the corrupted bridge. Then re-run the read-back. A single flip can clear while a neighbor remains inverted from the same paste.
Moraton reports a sign flip by row with a plain description. It does not invent market data. It does not certify returns. A person still signs off every output. If you are hunting "capex entered positive" at midnight, flip the sign, recompute the bridge, and only then let the file go upstairs.



