I just don't see a rationale to tracking total mg at all. Look only at the dosages and their tolerability for each compound in combination for the individual.
Androgens are so disparate in their effects and actions. You could perhaps try to approximate the harms for dosages relative to a testosterone-equivalent dose of 600mg for 12 weeks: e.g., use a 600mg/12wk multiplier, summate all your doses converted to T-equivalent; but then you'd need to factor in an arbitrary multiplier for 17AAs, something like a a multiplier to factor in hepatotoxicity (resistance to hepatic breakdown * potency to activate AR). You wouldn't want to factor in the Hershberger Assay (rat) anabolic/androgenic ratio given all the weaknesses of that. Even this would be very unidimensional: an unwieldy approximation of overall dose-harm based on extrapolation from Bhasin 2001 and a Medical Hypothesis on hepatotoxicity.
Even the basic rationale: that all androgens work via the AR (to justify calculating total androgen dose given the effects downstream of AR activation) seems incorrect.
You can see this most clearly with the binding affinity of Proviron vs. Test vs. 17AAs (Anadrol, Winstrol, Halo). By all metrics, Proviron is a relatively potent anabolic and androgen according to published data; it's a potent AR agonist in rats and rabbits (0.21 RBA in rabbit skeletal muscle vs 0.25 rat prostate) which is 3x greater than Test's RBA (0.07) for the rabbit skeletal muscle cell. We know it's actually very weak (well WE do, someone should ask the gurus like Victor, Team Evil Genius guy, whether Proviron is also a great "anabolic anchor" and does "protein deposition" in muscle as well as anything else).
On the other extreme, the 17AAs like Anadrol, Halo, Winstrol, all have 0.02-undetectable RBAs across the board, in muscle and prostate.
To me, it's clear that 17AAs and many androgens derive much of their activity from non-ligand dependent AR action, membrane AR/GPCR nongenomic action, by modulating glucocorticoids; even aromatization is a myotrophic mechanism, differential effects on intramuscular IGF-I activity.
This diversity of action between drugs makes calculations like even tracking total mg of AAS futile. As is any sort of dosing based on lean body mass/body mass. Start with a minimal effective dose and titrate up based on logic and experience, with some reference to science where most relevant.