I have not read it, or seen more than just the graph he shows, but:
It would be interesting to see breakdowns of the types of suspected adverse reaction, and the demographics of vaccine recipients (elderly vaccinated early, etc).
It would be interesting to see a control of remaining early doses (if available) used at later times.
For the larger batches, it would be interesting to see a breakdown of suspected adverse reactions over time within usage of one specific batch.
It would be interesting to see confirmation that criteria for the term “suspected” remained stable over the time of this retrospective (I had “flu-like symptoms” for 24 hrs, maybe this tended to be excluded later? Dunno).
It would be interesting to know the official composition of all doses and, if possible, an independent analysis of their composition (qualitative and quantitative) by, say, mRNA seq and mass spec.
It would be interesting to see a comparable study for vaccines from, say, Moderna used in a parallel population.
It would be interesting to see some [Bayesian] statistical analysis?
In short, in the rush to publish which has blighted the COVID field somewhat, it seems a bit of a teaser: So as it stands, perhaps endless debate is possible. Maybe he is right and this vaccine is super-safe now, or maybe reporting criteria (or some other factor) has/have simply changed.