The verdict card — is anything wrong with this machine

Every other view answers a question you already knew to ask. This answers the one a mechanic actually arrives with: is anything wrong with this machine, and what should I do about it.

Plan note: the verdict card is a Pro feature, and it reads the Pro views.

Where the rows come from

The card runs no analysis of its own. Every row is produced by one of the other surfaces — the deviation model, trip analysis, stage performance and its sensor checks — and carries a “Show me” that plots the channels behind it, moves the chart to the moment in question and opens the view that produced it. A verdict you cannot check is just an opinion.

How rows are ranked

The distinctions it is careful about

Everything is computed over the visible time range. Zoom the chart and every row is recomputed for that window — the same machine can be clean over a month and interesting over one Tuesday.

When it finds nothing

It says so, and then lists what it actually checked: how many channel families were compared member against member and how many against a stored baseline, how many recorded trips were read back, whether stage ratios were available, and the sensor sweep. “Nothing found” is only worth reading when it comes with the list of things that were looked at.

Privacy

As with every other view, all of it is computed in your browser. Nothing is uploaded.

FAQ

What does Overtrace check when it says a machine looks fine?

Every numbered channel family compared member against member and, where a steady window exists, each member against its own baseline; every recorded shutdown read back half an hour; stage ratios and the work split where pressure channels exist; and every channel checked for dead sensors, frozen transmitters and physically impossible readings. The card lists exactly that under “What was checked”.

Why is a cylinder that runs cold only marked “for the record”?

Because it has run cold since the log's first steady running window. A member that has been different since the machine was known-good is duty or staging — most often a throw on another stage that shares a channel name — rather than a developing fault. If it had been in line at the baseline and moved out since, it would be at the top of the card instead.

Why are sensor problems listed above machine problems?

Because a verdict built on a bad reading is worse than no verdict. If a transmitter is frozen or a pressure reads below the one feeding it, that gets resolved before anything is concluded about a cylinder.

Does the verdict card use AI?

No. Every row is a rule over numbers computed from your log, and each one shows the numbers it used and links to the view that produced them.