Crazy with magnetic Variation

As mentioned in other threads seems that MSFS 2020 has not the most updated or reliable value for magnetic variation.
I don’t mind much really but I need to be sure being able to set through SimConnect the proper bearing values I want.

Around england, as example, I get values of MAGVAR that are about 0.2 (documentation says it’s in degree)

The real value should be something around 2 degree.

If i simply check the expression
reading both values by the simulator I get something around 3.7

Now I have a geographical bearing and I need to communicate it to the autopilot I assume converting it to magnetic.

Which of the numbers I should use?

I’m considering making some test in norhtern canada where the value is so big you can’t possible be wrong and still not noticing it…

Anyone stumbled into this problem?

Not directly related to your issue, , but there does seem to be a change in the heading from NAV LOCALIZER. In previous sims, this was always the true heading, but now seems to be using Magnetic. To get around this problem in FSUIPC, I subtract the magvar from the localizer heading to get the true heading.
This was reported in the following thread: ILS localiser inverse runway heading - FSUIPC7 MSFS - The simFlight Network Forums.

Note there is also PLANE_HEADING_DEGREES_GYRO. which is the magnetic heading + drift.
Maybe try with that if the difference is due to the drift.