I agree, but as I tried to state, there is simply more to it. Giving bad interviews doesn't mean anything. Accidentally beating the world champion doesn't mean anything. Spending all night studying a rarely played chess line that just happens to be the exact line played the next day doesn't really mean anything either. Not really being able to analyze like a GM doesn't mean anything too.

But when you have all these factors happening during one game, statistically it is not probable.

You should see the sequence. He was totally unable to explain any lines he had in mind, stating some positions were « obviously winning » (where it was absolutely not obvious, and in fact the engine marked it as loosing), etc. A total disaster.