It does seem weird that volumetric density has not been improved. Maybe you are correct, or maybe the number of cells (4416) is actually incorrect. Tesla did a false flag operation to determine who is leaking data (the old catch a spy trick). Only half joking
It only takes one person to open uit the pack to uncover the truth. Not a very clever long term strategy, obviously.
And after boasting best densitty on ttop of lowest cost, why make a product that could be cheaper or better still at zero change in cost? Tesla is well used to charging people for somethig they supplied extra (60/75 packs software restricted). Since they could be thought to keep 20% of energy from buyers, a 95-100 kWh Model 3 pack would in fact be possible, despite Elon claiming75 would be all that would fit. Guess how many customers would LOVE to pay extra for a 100 kWh Model 3 which was already built and sold to them?
Best guess I have, is that a yet cheaper way to make a decent cell was found, and it made the cut to production. So perhaps indeed they have more in store, but are just postponing it. First to put in S and X to keep margins up amidst general market cost and eventually consumer prices to drop. Say, they change the 100D's to 125's at the same price, but in fact the 125's cost them less than tthe 100's anyway. Model 3 was promised at this price, and reservation holders will buy it until Audi or Jaguar make something with a bettter value proposition. No need to discount those cars just yet. Long range cars, those will come under price pressure when the rest of the market gets involved around 2019.
Tesla will want to get cost of Model S and X down asap. So, when both cell production is up there and they've figured out how to let robots assemble the packs to save YET more cost. The $30K marginal price mentioned a good while ago (by Tesla) for a RWD Model S of course will have since crept down, but with 2170's, even of first gen, will be closer to $25K if not less. Remember when batteries were the most expensive part of the BEV? That may still be true, but its share in total car cost has been plummeting. Exhibit A, the Chevy Bolt 63 kWh pack is $15k retail. And car makers are not known to subsidize spare parts, let alone batteries which are in short supply. Every proper cell made on Earth, has a buyer or will spend vvery short time on a shelf. Not such thing as Li-Ion overstock in meaningful quantities. If the product is good and price fair, it's (pre-)sold.
Back to the Roadster. Since basic 5-year-old Tesla motors seem up for the job of powering this car using 3 of them, unless they want exceptional efficiency and/or long burst performance, batteries will be a tiny factor only towards total car cost. For sure by 2020, 2170's will cost Tesla under $20K for the whole 200 kWh car, 10% or less of sale price. As also in fully loaded Model 3's, by the way, $8K in cells towards a $80K car). Let alone Model S/X P100D if that's possible with 2170's (I'm not sure the same power can be drawn for the same amount of time), $10K in cells for a $150K car.
Batteries becoming such a relative afterthought in cost management, if by 2020 Tesla can acquire some batteries of higher gravimetric energy density (say, double), at triple the cost, that'd be TOTALLY worth it. Absolutely no need for bottom of the barrel ultra cheap ass 2170's to make up the 200 kWh.