Those were early numbers, and there were some outliers causing those numbers that had nothing to do with him such as the opposition shooting poorly from the line when he was on, and extremely poor 3PT% that he individually couldn't cause. Like the article itself said:
Opponents are currently only converting 29.3 percent of their 3-point attempts when Teodosic is on the floor compared to 36.5 percent when he’s off it. As we’ve learned over the years, 3-point defense is essentially random, so it’s a good bet variance will kick in at some point and opponents will start making more 3’s when Teodosic is on the floor.
The same effect is happening at the free throw line. Though teams are getting to the line less often when Teodosic is in, they’re shooting 72.8 percent when they get there. When Teodosic is off the floor, that number rises to 77.9. Unless Teodosic is whispering secret, Serbian spells into opponents’ ears as they shoot free throws, this is also bound to change.
With a larger sample size, things have changed just like the article said they would, and while the team is still a solid +5.3 pts/100 possessions when he is on the floor, that is nothing earth-shattering or specifically impressive. For example, we were +19.6 pts/100 possessions when Chris Paul was on the floor last season, +16.6 in 15-16, +20.9 in 14-15, and Paul's "weakest" in the Doc era was +9.1 in 13-14, still a very good number. So with a larger sample size, Milos is no longer objectively the "single biggest difference maker" and certainly not having that crazy on-court impact that we've seen from other players.
That doesn't mean he isn't a nice positive influence on the court though, he can still be a nice positive influence without it being any grand difference.....