This is why we are having this discussion... I'm not saying you are stupid, but I think you lack practical knowledge as well. When I look at his monthly option returns tables, for both puts and calls, the first thing I see is that calls return positive and puts negative. There is only one way this is possible on average/consistently... and that's market up. Also, the mean returns show bigger loss (whether excess return or not) for OTM puts, and bigger gain for ITM calls. That means to me, that the loss of put premium is mainly time value plus upward movement. For the calls, it's loss of time value, but gain in value because delta long. Further, monthly average return of S&P500 is not that big, so you also see a loss in OTM calls, but a gain in k=1.06... probably because low purchase value and that few times the market rallied in one month catapulted the call value. That's my 2 cents of practical options trading experience (10+ years professional) together with masters degree in financial markets... I'm not sure how to interpret the confidence interval data... probably because my lack of statistical knowledge ps. he did mean a negative 95% excess return on OTM puts... and it really doesn't matter that it's excess and not normal, since it's that high. Can't be higher than 100... it's just high...
I will say that I buy a lot of deep otm puts and am happy when they go out worthless. There are other benefits to being net long units on the downside of a portfolio including PM margin control.
For my PA I am a happy seller of index puts (covered by strike cash) - after all, you can think of them as getting paid for dollar averaging. I also think that front end crash is crazy rich these days due to silly regulations so a private investor might as well figure out a way to sell them (you know, in IDB 80/70 daily crash cliquets for 50bps per year, which is silly). However, as a career trader you don't want that exposure