There is always base attrition of all equipment, which can be modified by improving reliability stat and is also modified by terrain, weather and supply situation.
You always have to take into account all systems interacting together in plausible scenarios. You can always invent scenarios in which the supply system makes no sense, but these scenarios usually make no sense themselves or would lead to certain defeat given approximately equall skill of players.
One scenario presumes overwhelming superiority on eastern front, so you have minimal losses therefore minimal fuel requirement. But:
1. If supply allows you to have overwhelming strength compared to Russians without suffering supply problems (including reduced fighting efficiency and high attrition due to supply issues), then supply numbers are imbalanced and should be changed during beta.
2. If game allows Germany to produce overwhelmingly better land army compared to USSR so it can have minimal lossess even at the gates of Moscow, then there is tech or production imbalance which should be adressed during beta. All scenarios presented on forum which presume that Germany can prevent oil shortage during war by just tech rushing and producing all tanks/planes they will ever need before war starts just mean Germany has just too many factories and will be nerfed.
3. If Germany can roll mechanized forces to Ural with minimal losses (even presuming minimal opposition), then base (and terrain and weather) attrition should be ballanced to produce something more plausible.
In the end, if all other mechanics will work as intended (and devs want to insure that given the stated vision), oil/fuel availability will limit mechanized forces in plausible ways, end justifying the means and all that. Not immediately, but gradually over time, which I think was their goal, because it would be stupid to have binary states like move full speed one day and stop the other day, because you "just run out of fuel", forget all things like conserving when running low and reserves for critical situations. If we had separate fuel for supply, it would mean producing a system which would not reintroduce stockpiling and produce gradual effects of fuel shortage, reduce (or eliminate) hindsight and last but not least, will be easily visible, understandable and fun, which would be no easy task! Also an extra layer of complexity on top including managing priority if you don´t have enough fuel (on top of managing priority for equipment) and separately balancing effects of fuel shortages, so more things that could contain possible exploits. Given all that, an extra layer of complexity just to simulate fuel consumption would be redundant, and mostly there just for immersion.
Fleets are a special case because they do not have attrition, but as Podcat said, it is not much fun having to watch your grand fleet sit in port while USA takes over Pacific because you have no fuel to intercept them. Much better to challenge them, but not be able to replace lossess because you have no fuel to produce those replacements (just roleplay you saved just enough fuel for those critical missions). Also, looks like we will be producing ships faster than IRL so do not expect that Japan starting fleet "running on fairy dust" will be enough to win the war. In the end, the result is the same (defeat because of lack of fuel), which is what counts.