D&D 4E did not have any mechanical problems that caused its relative failure. As I said above, its problem was not mechanically inept design: it was a failure to understand the allure of the older system, and an attempt to fix it that instead aggravated a lot of players.
See, playing a Fighter in D&D has never been rocket science. “I hit it with my sword” is about the maximum amount of investment you need to play a Fighter in original, advanced and 3rd edition. 4E changed that by making Fighter just as complicated to play as the Wizard. This meant that for a large portion of the fanbase, D&D 4e was a giant step up in terms of complexity. Ironically, 4E also made the Wizard much less complex to play at the same time. This failed to appeal to the Wizard fans, because Wizard fans liked playing insanely complex characters with lots of fiddly powers. That was the allure to them. So you get one set of fans calling 4E too complex and another set calling 4E a dumbing down and both were right and both were wrong at the same time.
4E was an example of the Golden Mean Fallacy in action. Sometimes, the right answer is not a compromise in the middle between two extremes.
5E is an attempt to address this, by allowing players to choose their own level of complexity within their class. You can play a straightforward “I hit it with my sword!” fighter, or a complex Warlord with action dice and all sorts of shenanigans, as you please. It’s an attempt to appeal to those opposing fans who want both a simple and complex game at the same time. Whether it succeeds… eh, we’ll see. But at least it is an attempt to address the problem with 4E. For a game the size of 4e, there is no one size fits all.