First principles thinking would seem to favor adjusting the set price of each model to manage demand vs. trying to hit a preconceived gross margin estimate. It's undesirable to either have waiting lists of a year or more or to have finished product building up with no willing buyers. Tesla is currently so far away from losing money on each sale that it makes no sense to base selling prices on anything but demand. This makes managements job easier because the game is to simply make as many cars as possible and then adjust pricing as needed to manage sales demand.
This leaves gross margin projections fulfilling only one role, that of financial and product planning within the company. They need a reasonable idea of what their finances will look like a year or two down the road. Tesla publicly links price increases to supplier price increases mostly for optics. It makes sense to people that when suppliers raise their prices, that Tesla should as well.
It's also true that one disadvantage of having waiting lists is that the company is taking on commitments to sell a car a year down the road at a certain price while the buyer can walk away at any time. If demand suddenly drops, as unlikely as that may be, Tesla will lower prices to increase demand, irrespective of their cost to produce or the number of non-binding reservations they once had. When a manufacturer lowers pricing it's never because they are passing on the savings of lower supplier costs to their customers, no, it's to increase demand for one of two reasons: to move unsold product (real or anticipated) or to increase production and ensure large enough demand will be there based on company estimates and projections.
Unprecedented demand for EV's has found Tesla selling them for well below actual market value. That is evidenced by used cars selling for more than new cars. In hindsight, management should have raised prices sooner and more aggressively, but hindsight is 20/20. I can't fault them for this because it's not realistic to expect anyone to be able to forecast future demand perfectly and it's better to err on the side of giving people a good deal vs. creating a situation where they are scrambling to lower prices as reservations are cancelled and inventories build up. Always err on the side that causes less problems if you are wrong.