[Kliment]

As I understand it the landing legs didn't deploy because velocity was too high and angle was wrong. All this stuff gets reported by telemetry because they know exactly how likely it is that a first attempt flight will fail catastrophically (very). So these test rockets are heavily instrumented and networked. Watching it the first time I also thought one of the engines had failed but then that engine had a successful relight and another did not, and the fuel mix ratio appeared to be wrong on one of the engines. This is all consistent with loss of propellant (there was a quite significant leak after the first fire in engine bay) and not consistent with "engines ate themselves alive".

Your greater point about it being better to land than not land stands, but these are experimental rockets. They're expected to fail catastrophically and you plan accordingly.

[nine_k]

The video shows us the wreck site at the end. The rocket is mostly one piece though damaged by hitting the ground too fast and from a (relatively minor) explosion.

It must be possible to find most pieces, and certainly check the engines, even deformed, for overburn signs.

[grey-area]

Bright green flames at the end is a giveaway of engine-rich combustion.

[mhandley]

Indeed, which in turn is an espected outcome if the engine is running oxygen rich, and hence much hotter than the design point. Pretty likely the low fuel tank pressure was the cause of this, and I'm sure SpaceX will have the telemetry data to know for sure if this was the case, irrespective of what pieces remain after the landing.