Yes, really. I have a dual-motor performance model, and I have not seen any material range reduction in the extreme cold that we recently had in the Northeast U.S. Our ICE vehicles did not handle the cold as well, having a myriad of different issues caused by the extreme cold.
As the AAA study noted, it all depends on how you're driving: if you drive at low speeds, it can be a lot; if you drive at highway speeds, only a little**. Likewise it depends on the air temperature, whether you're doing a full trip or not (higher energy consumption at the beginning), road conditions, wind, etc, etc, etc. But for a "typical" winter day vs. a "typical" summer day - 20°F/-7°C vs. 75°F/24°C - you lose about 18% if you drive US06 (typical US-style highway driving); 29% if you drive at moderate speeds; 38% if you drive EPA-combined (mix of slow and moderate-speed driving); and 45% if you drive UDDS (slow speeds). But geez, you'd be driving a long time trying to exhaust a pack with UDDS-style driving
** Only a little reduction relative to the same sort of driving in the summer. Obviously, if you drive at fast highway speeds during the summer, you won't get the EPA range (and certainly not the WLTP) then either. But we're talking about a reduction relative to your summer range, not relative to some arbitrary range standard.