Ioandis' original study was in Santa Clara County, CA, where they estimated 0.17%. I don't see how New Jersey, for example, materially differs in a relevant way in terms of population statistics from Santa Clara.
I would also note that New Jersey STILL has a high COVID case rate, even with 47% of the population vaccinated. They did not achieve herd immunity with their 0.3% death rate, because far less than 100% of the population has gotten COVID, so the IFR is very obviously significantly higher than 0.3%.
They made a prediction, the prediction was wrong. It's clearly motivated reasoning, because the authors on these studies have had policy prescriptions since the very beginning that depend on COVID not being very dangerous.
Unfortunately reality is not required to adhere to a particular set of political views, and arguing with a virus is not effective.