I think a big problem is ground troop AI. Many are choosing lavaloonian attacks because it is easier to predict the outcome. Ground troops you just have no idea where they are going to go, (you can aim them in the general direction most of the time) Supercell has even stated that those troops have a bit of a randomness factor into their coding, which is why you are wondering sometimes as to why your heroes can make the worst possible decisions at times. Its because its random. They stated that they don't want the same attacks to always have the same outcome and thats why they added the randomness factor.
This is reason I prefer lavaloonian (before hounds were even released and I love the addition), because you don't have to deal with the stupid troop AI because I know they are going to target the nearest defense every time and can predict their paths. With ground troops that target anything you have no idea what they are going to do next, you can drop everything perfectly, but it doesn't change the fact that their randomness motions will sometimes cause them to circle the base.
Who wants to flip a coin every time they attack and hope their troops choose to do the right thing? I think the big problem isn't lavaloonian, it is the ground troops AI (especially heroes and wall breakers) at the moment preventing attackers from actually fully planning an attack.




