My take is that there is no value in revving beyond 4500 rpms until higher speeds, where hp is needed to push through the air resistance. Shifting any gear at 4500 rpms keeps the engine above 2000 rpms in the next higher gear. This is the max torque range. If you go above 4500, torque drops off and therefore maximum acceleration is not achieved. I haven't yet fully incorporated wind resistance into my calculations but will post if I get that far.
Thoughts?
I think you'd want to stay in the max
horsepower range, not torque (unless pulling stumps). I don't know for sure but bet Vmax is achieved close to peak hp. F1 engines rev to 20,000 rpm to achieve horsepower, with relatively low torque.
Yes the 718 hp curve is steepest in the 1700-4700 rpm range corresponding to the peak torque plateau, true, but that describes only the rate at which the engine is building hp. You'd still get the fastest acceleration
of the car (as per quarter-mile ET and trap) by maximizing horsepower. In a world of power transmitted to the wheels by fixed gearing, that means maximizing
average horsepower by keeping revs in the narrowest power band possible leading to slightly beyond peak hp 6500 rpm. Ideally shifting slightly beyond peak depending on hp at the bottom of the band (in next higher gear) versus hp at the top of the band (in lower gear) somewhere between peak and redline.
That point will be lower than peak in absolute hp as the curve is trending down, but still higher than the drop-down point in next gear. That's why it's useful (if somewhat counter-intuitive) to rev beyond peak even though the power curve is 'dropping off'. Overall acceleration of the car -- a function of average horsepower in the power band -- is not dropping off, at least not immediately after peak*. Does that make sense?
wrt other posts: my own impression, after 15 years of n/a flat-six engines (but not incl. the 981), is as follows. I think Porsche engineering had a design objective to make the 9A2 (in both Carreras and mid-engine cars) as nearly like their n/a engines as possible, insofar as responsiveness and free-revving to redline qualities. That was their big challenge since huge torque and with it, higher average horsepower, were already givens for turbocharging**. That they succeeded should not be a surprise, given what was at stake (not so much with the B4 but crucially with the B6 Carreras which *had* to feel as good as the n/a 3.8s they replaced) and their experience with racing turbos and also the 911 Turbos and GT2, and given who (on the engineering side) we are talking about.
As for why so many journalists still identify turbo-lag and revving-strongly-to-peak as issues -- well these were the well-understood main drawbacks of turbos forever, just the immutable trade-off for gaining all that power. Thus the resulting perception-lag being hardest to overcome.
I suspect also that turbo-lag can be mistakenly associated with simply being in too high a gear or expecting that big low-end grunt to be felt all the way below 1500 rpm (indeed these are two sides of the same coin).
The perception of constrained free-revving up top is harder for me to nail down as despite 15 years of running n/a flat-sixes to redline, I'd need one in hand again to A/B this phenomenon. Since the turbo redline is 300 rpm lower than its predecessor, and peak 200 rpm lower, I guess the combined sensation could be one of peak power dropping off (slightly) sooner. Still the shape of the curves around peak look very much the same to me on the chart. Could something as simple as the more urgent soundtrack of the n/a at peak contribute to this perception?
*better to get another 10 milliseconds of 345 hp at 6700 rpm -- even though it's begun 'falling off' from peak [email protected] -- then to shift exactly at peak and spend that 10 milliseconds in next gear at 250 hp @ 4500 rpm (even though it's at max. torque and steepest part of the power curve). Indeed it may be best to go all the way to redline, but that depends on the gearing as well as the power curve.
** If I've gained one "aha!" insight in the past year of analyzing 718 performance, it's that high average-to-peak-hp ratio of the turbo power curve answers "how is this even possible" for 350 peak horsepower to accelerate like 385-400 peak horsepower in non-turbo PDK cars of similar weight. 
