You're assuming I want to choke the AI. That doesn't always work if you have a lot of AIs all around you, and even when it does, it feels kind of cheap.
I'm sure the Quechua is an even better choking unit.
Rushing/choking is a personal play style choice. If you don't like doing it, than a uu that's good at it certainly won't be the most appealing for you. However, saying that the dog soldeir is a poor uu based on personal
preferences isn't the best argument(e.g. I
prefer not to use the praet because I believe it is overpowered; however I do not deny that it is a superlative UU).
Also, when I said your logic was flawed, I was referring to your argument that dog soldiers are used predominantly on deity, and since most of the community doesn't play on deity/immortal, their usefulness is lower than that of a ballista elephant. You stated this here specifically:
Again, the specific logical fallacy is that the deity tactic is not useful to the community at large because it plays on lower difficulties (where said tactic would be more effective in fact) and therefore the ballista crapshoot is more useful.
Besides that, it's not "overwhelmingly hammer efficient" either. Against archers, it's strictly inferior to a normal Axeman, whether you're facing them on offense or on defense. This is markedly apparent if the AI attacks you early with Archers, and sometimes it does.
While certainly not hammer efficient, the Dog Soldier has one major advantage against archers as compared to axemen: They can attack before there are a lot of archers. In all fairness, regular axeman aren't that great at attacking heavily fortified archers anyway. You need 2 to 3(if protective/uu/hill/bad luck with rng) axes per defending archer + barracks which is pretty poor efficiency in and of itself. If I'm going for a well defended city, I might as well wait till cats/swords or spies.
Any advantage it has in dealing with barbarian Axemen are, I deem, merely compensatory for the disadvantages it gets against barbarian archers.
Since when are barbarian archers a problem for anyone? Put a dog on a hill/forest between the barbs and your city and archers will never be a problem. Or just fortify them on a regular plain tile. You could use archers for this just as well. Although dogs are nice for this in that you can hit bronze working and have a counter to barb archers available directly as opposed to then teching hunting archery in the event that you don't have copper in your bfc (as are holkans and quecha).
They're slower, yes, but again, unless you're playing the highest difficulty levels wherein you have to consciously exploit every AI programming weakness you can manage (even going so far as to consider choking), then it's simply not very relevant to most games being played community-wide. Very few people routinely play Deity for fun.
Again, you're attempting to tie the usefulness of the dog soldier to deity. Fallacy. As difficulty decreases, it takes the AI longer to get a force of archers sufficiently large enough to prevent a rush. While it may not be
necessary to rush on lower difficulties, it can be extremely
profitable (becoming more profitable as difficulty decreases). It can be more profitable to rush sooner with dogs than later with axes.
It's like saying the Praetorian is useless on settler because almost every player can beat settler with their eyes shut and their hands tied behind their back using any strategy. While that's certainly true, it does not remove heavy preat spam as a viable strategy on settler. It will work there just like it works almost everywhere. By the same token, Dog rushing/choking is a viable strategy on low difficulties as well as high with the added advantage of being able to ignore copper for a while. On the other hand (running with your earlier ballista to dog comparison), the ballista elephant will almost never be useful to anyone on any difficulty or setting. Almost.