This is highly inaccurate. The author is using Plausible via a proxy but Google Analytics directly (a very biased way to do things).
Secondly, PiHole, uBlock origin and most other adblockers also block Plausible analytics (there are discussions on their own Github Issues regarding it). If you are proxying Plausible to get accurate readings, same should be done with GA.
I don’t read this as Plausible vs GA but server-side anything (including a GA proxy) vs client-side GA. in that case, it doesn’t seem like these figures should be wildly wrong.
Secondly, PiHole, uBlock origin and most other adblockers also block Plausible analytics (there are discussions on their own Github Issues regarding it). If you are proxying Plausible to get accurate readings, same should be done with GA.