Who the **** cares what his pass success rate is and how does that make such a subjective discussion so conclusive?
First, you'd need to set the parameters in order to define 'better' then establish the relevant data which would prove or disprove a null hypothesis...
From what I remember of my uni days...
Generally speaking though, the validity of statistical interpretation generally depends on a sound methodology, which nerding on footie sites and making far over-reaching conclusions based on a clear, pre-determined bias certainly isn't.
My issue with xG is precisely that as a concept it is inherently subjective - and there's no standard, nor is it even possible to statistically define a 'good chance' so there never can be an actual standard.
Stats are often a pretty blunt instrument and inherently limited in what can be extrapolated from them.
However, what you can do is point to the 4th best defensive record ( goals conceded) and 10th on form (points accumulated relative to the competition) over 28 games of a season and rebuff the idea that it is rational to predict a bottom half finish based on the manager in question's prior record at the club over an extended run of games.
You can also rebuff the idea that someone is a championship ready 'ball winning DM' based on worse ball winning (successful tackles) stats in the championship whilst attaining fewer senior minutes than a spangly 19 year old winger who only signed from non-league 7 months ago.
One can often make quite broad assumptions, but stats by their very nature will never be qualitative.
The only team stats that really matter in football are goals for, goals against and points accumulated over a significant number of games. Basically, the stats the league table is built on.