Of course you are right, that is why I only gave figures for when the airplane was on the ground and at that time the book says there is no position error correction, so it is the only time the altimeters can be accurately checked. For that matter, they cannot be checked against each other apart from in normal level flight, so comparison on final is pointless.
I actually simplified the figues, by reducing them to the same datum. The actual PFD altitude for the first check was 20 feet while the standby was 100 feet, for example. This is in fact outside tolerance for the altimeter system, according to the FRM, but I would be willing to bet it has always been this way and not one other pilot in the company would be aware of this. Of course when the temperature was at a more normal value the standby altimeter was again within tolerance.
The company I work for says the correction for low temperature should be made, but it does not say how it is to be done or where. That allows the more pedantic among us to complicate the procedure to their heart's content. Most, however, do not have a clue and leave it alone. This does not worry me, since we do not operate old airplanes in the Frozen North and neither do we fly non-precision approaches in poor terrain. Very few of our guys have ever flown a circling approach, another time when corrections would be very important.
Most of our guys do not even know that the auto Antice system does not operate on the ground, which is way more scary.
I have an agenda of course, and that is to simplify the operating procedures. I see the altimeter correction as an unnecessary complication IF it is in fact not needed. If it IS, then we should all be doing it, and the right way.
Unfortunately, nobody has posted here who has the real technical poop for me to decide this for myself. And so far, my requests to the manufacturer for information deeper than what they have published in their technical manuals have been ignored.
Nevertheless I have been educated way to hell and I hope others have too.