If you have put calories/HR/exercise data against a chest heart rate monitor, how does the overall count and distance measurements stack up? It appears you are quite the PRO and it is very interesting for me to read about your experience!
I'm not sure the exact comparison you're asking for, but I'll offer a couple.
I also have a Bluetooth Polar HR chest strap that I can pair with my Apple Watch. I bought it early on, because I wasn't sure of the accuracy of the wrist-based HR data. But at this point, I seldom use it except when I'm rowing (and that's because the Concept2 ErgData app can't pull HR data from the Watch). The chest strap produces slightly better HR data--more frequent readings, so slightly faster picking up quick changes--but the overall readings are basically identical. (I've compared the data from the Watch and the ErgData app during rowing workouts, and the average HR from the two is consistently within one beat per minute.)
At this point, I would only use the Polar strap for running if I were running a HR-based interval workout where I wanted the best HR data possible. On a typical run, I only need a general indicator as to how hard my heart is working, and the Watch is fine at telling me if I'm at 140 or 150 bpm.
I haven't seen any evidence that using a HR strap changes calorie calculations on the Watch. The Watch always has HR data, so substituting data from a strap is not going to make much of a difference. (Also, calories burned per mile running doesn't depend much on HR. You burn calories faster running harder, but you also cover the distance faster. The end result is a fairly consistent calorie burn per mile given the same runner on the same terrain. As an example, on Wednesday, I ran 8 miles at 8:16/mile with an average HR of 147--it was hot and humid. The Exercise app says that I burned 108.7 calories per mile. Last night, I ran 12 miles at 7:27/mile with an average HR of 155--it was much less humid. I burned 108.7 calories per mile. The runs were on different stretches of the same flat bike trail.)
In the past I've used various Garmin Forerunner devices (202, 215, 315xt, 910xt). I still have the 910xt, and it works just fine. I hardly ever use it anymore either. The GPS accuracy of the Apple Watch seems comparable to the Garmin devices. The advantage of the 910xt, at least when it was new, was 20-hour GPS battery life compared to 5-hour GPS+music life on the Apple Watch 4. If ever I decide to run another ultramarathon, the longer battery life would be handy. For marathons and marathon training, the Apple Watch battery is more than sufficient.
I did notice in the past that I got very different HR estimates from my Garmin 910xt depending on whether I was wearing the HR strap. Without HR data, Garmin estimated much higher--inflated--calorie burns than with the HR data. I haven't tried turning off the HR sensor on the Apple Watch to see if the same thing would happen.
[The only poor GPS device I've ever owned was a Garmin Fenix 3 bought before the Apple Watch was released. The GPS performance was so terrible, that Garmin sent me a replacement after collecting data from it. (They had me go on runs and bike rides wearing both the Fenix 3 and the 910xt.) The replacement was better, but still incredibly inaccurate compared to the 910xt. I ended up returning the Fenix 3 for a refund.]