Of course, ITEM9 had already predicted this way back when we were starting out with the JRSC.
Biggest problem with the "test" video shown above is that our CVT control software is programmed with a bias for economy (and self-preservation), rather than absolute performance. Those are good things in the real world. But not for a drag race.
As long as you don't get into trouble with one of the various other computers in the CR-Z 6MT, you can pretty much do whatever you like with that tranny.
I strongly suspect that the TCM in our cars will stifle torque as it approaches some pre-determined limit, for fear of slipping. Slippage in a CVT is detrimental to the longevity of the transaxle.
The Eng. Explained video is a bit more "theoretical" (as most of his are), and he even says so. Try to put too much power down through ours, and I'm afraid you'll be picking up pieces of CVT all down the highway.
The really superior CVT would be a lot more robust than ours - and heavier, which is not what Honda wanted to do to our cars. It might not yet exist.
In the end, a showdown against a 3-pedal manual transmission car relies on the ability of the driver, too.
Makes the one-on-one video fun, but the outcome is somewhat inconclusive.
No question that it will be an uphill battle trying to sell CVT's to supercar buyers.
It is hard enough to convince purists that an F1-style dual clutch is better than a manual. "Better" at going truly fast, that is!
The one thing that WON'T matter is the CVT whine that most automotive scribes whine about. You'll never hear that over the engine in a 12 cylinder Lamborghini!