Even a simple pan law can - under certain circumstances - affect the two readings, when statistics measures an audio track, and compare it with the master bus levels
Yup, but given the circumstances (metering is done post panner) and if the statistics are created from the final mixdown then there should be zero difference in peak value, regardless of pan law setting.
Right click on an > audio track> , audio > statistics - Peak amplitude
This made me think it isn´t the mixdown, the statistics are taken from. I think I misunderstood though.
I think the real problem is the false reading of the levelbar…
Well I’m not entirely sure either, but since there’s no way do do a summed statistic reading from multiple tracks I assumed it was done either on a (completely unaltered) single track project before mixdown or on the resulting file (using “import to track” option in the mixdown window for example) of the same. Any other method would say nothing.
Well, I thought I just had one track going, so the audio track would/should be reading the same thing as the master bus.
Because the actual audio level on the master fader input is defined by the sum of the single channel faders, and is in no way depending on the master fader. The master fader can attenuate or boost the output level of the master channel, but raising the single channel faders will also raise the input to the master bus, without any relation to the master fader. Master fader at 0.00 simply means, input level = output level - no attenuation, no gain.
So if a signal is fed into the master bus, at -3 dBFs a fader level 0.00 it will go to the oputput with -3 dBFs though the fader is at 0.00
Ok, I still have no idea why the fader being set at 0 looks as though it’s set at -6 next to the level meter, but that’s ok.
If you mean the level meter, then yes - given, you´re monitoring the correct meter point (meter input, post fader, post panner) for your given task
Ok, then that’s all I really need to know. I have a feeling talking anymore about why the fader is actually neutral at -6 is going to make both of our heads explode before I get it. Thanks for your patience.
I did it on a single track that I had imported after I got it back from the mixing engineer. I just realized that I actually had a couple of other tracks going, which is part of why the levels are reading different. 
With those turned off, I still get a different reading in harbal, but it’s not so different that I can’t live with it.
Thanks for the help, guys. I feel so weird having read things wrong in cubase for so long, at least I discovered it before sending a bunch of songs out to mixing and mastering engineers.
OK, so now we have that, just for the records: There are also intersample peak meters, who will again measure diffferent Peak values because they do not measure the actual sample value, but values that can appear between samples. Cubase´s Peak meter does not measure intersample peaks.
Just to point out that this is not always the case.
Due to whats called the panning law on the master buss, default is 3dB, what this means is that a single (mono) channel that peaks at 0dB at the channel meter when panned centre and routed to the master buss will read -3dB.
To see the same level you would have to pan it either hard left or right.
This is designed so that the signal will stay more or less constant in level as the sound comes into the centre. It needs to be turned down as the influence of two speakers comes into play, or the sound would become louder.
Also the dB reading on the peak meters should not be confused with the Fader level setting, the two are entirly independent.
Also just to make clear, Inter-sample peaks are only a (potential) problem in the digital to analogue conversion, Cubases meters accurately give the actual sample levels (as it were)
Just to help you clarify the functionality of the channel fader and the meters: the position of a channel fader has absolutely no direct relationship with the scale of the meter.
What you are doing with the meter is monitoring the incoming signal either pre-fader or post fader depending upon your global meter settings. If you set your global meter settings to ‘meter input’ you will notice that moving the channel fader has absolutely no influence upon the level shown in the meter. However, if you now set your global meter settings to ‘Meter post fader’ or ‘Meter post panner’ moving the channel fader now shows the level changes.
So your channel fader is a volume control. Nothing more, nothing less. If you set it at 0dB (unity gain) it means that you are not adding to or subtracting from the level of the incoming signal. To be perfectly clear:
If you select ‘Meter Post fader’ in the global meters settings and use a test 1kHz mono signal at -20dBFS on a mono audio channel with the fader set to 0dB, you will see -20dBFS registered on the meters. If you now move the channel fader down to -6dB you will see -26dB registered on the meters. If you move it up by 6dB you will see -14dB registered on the meters.
Using a test signal helps clarify what is going on.
Maybe someone can repeat all this one or two times more, just to make sure… 