Well, honestly the description sounds like the training wasn't completed. The end goal is to be able to call the dog away from anything it is guarding. In that situation, I would want the training to be complete so that if the dog guarded something unexpected, the protocol would be the same…over here dog, and good stuff will happen.
The longer a dog has practiced guarding, without intervention, the harder it is to retrain with any protocol. The idea is to catch it early, and change the way the dog thinks about guarding. If the dog thinks he is trading up anytime he has something he values as good...then he won't feel the need to guard.
I know a lot of people feel this is backward; when a dog growls at you, you must correct it...but the new thinking tells us that these methods just work better.
I agree with a lot of the critique, though....yes, it probably has a genetic component...yes, if a dog starts with little to no bite inhibition, it is unlikely that will change; however you *can make your dog bite harder, by ignoring the initial growl/snap/muzzle punch that leads up to a real bite. It is an escalation the dog uses to get it's message across. And, yes...of course I agree with strong management or euthanasia for dogs who do serious damage to family members. This is not Chance, though. He is trying it out, and using quite a lot of bite inhibition...so now is the perfect time to start the protocol.