I can’t extract a whole lot of details from either of those articles, so I don’t know exactly what they were measuring for. But if I had to guess, I would say that both articles are misrepresenting what was actually measured. The whole point of quantum mechanics is that measuring collapses a superimposed state, so you can’t measure both states at the same time: You get one or the other.
BUT if you repeat the experiment many many times, and nothing else changes, then sometimes you’ll get one state, and sometimes you’ll get the other. If you’ve successfully eliminated or accounted for all the variables, all that’s left is quantum probability, so you can show that the state you happen to find when you measure is random, but with predictable odds.
So in this particular case, you take the paddle and reduce it to its ground state (not moving). The first excited (moving) state has some associated energy. We’ll call that energy… 5. In classical physics, you could say that it takes 5 energy to bump the paddle from ground state to its excited state. Any less and it just won’t make it out of the hole.
In QM, though, if you add, say, 3 energy to the paddle, then the paddle isn’t exactly the ground state or the excited state. It’s in a superposition of states with a total energy of 3. But since you can only observe the paddle at one quanta or the other, once you measure it, it will either be in the ground state or the excited state. (Or a higher excited state, but probably with a very low probability)
So you take the paddle, pelt it with 3 energy, and see if it starts wiggling. Then you drop it back down to ground state, pelt it with 3 energy, and see if it starts wiggling. Do this a lot. See what percentage of the experiments lead to wiggling. Compare that to expected percentages. Get misquoted by FOX. The end.
I’d really like to see the actual experiment, though. Has anyone found, say, a paper written by the actual scientists involved?