Guy scores 19 points once, and he's a good pick over guys who are legit stars in the league? There's a term for this: rationalization.
That's not an argument that anyone here is making. No one is saying that taking him was better than taking the Greek Freak. Nor are people saying that just this one game is what makes him a good pick.
Pretty much every non-top 10 draft pick (and many in the top 10) has a better player selected after them. This shouldn't be a surprise: most later picks are either going for a safe bet (like Olynyk) or swinging for the fences (like the Greek Freak or Bruno Caboclo). With 47 picks after Olynyk for potential "swing for the fences" picks, some are going to be hits. That's just basic statistics. There's no guarantee on picks like Giannis. Sometimes they work out, but most of the time you end up with a bust. You can't compare a safe pick vs a risky one without considering the chance of them being a bust. Danny chose to play it safe instead of taking a risk, and got a good player out of it. It wasn't the best possible outcome, but that doesn't mean it wasn't a good outcome.
And nobody is saying that Kelly is good based on this one game. Last year, he shot 40.5% on 3.0 3PA/game, good for 14th in the NBA, and best among big men (unless you could Omri Casspi or Doug McDermott as big men, but they mostly play SF). Like him or not, his 3 point shot is a definite weapon, and when you combine it with his decent perimeter defense (which is rare for a stretch big, although his lack of interior defense is pretty typical), you have a player that, while not a star, is a reliable rotation piece that any team would love to bring off the bench, and that quite a few teams picking ahead of us (Cleveland, Sacramento, Utah, Philly if they hadn't traded MCW) would likely have preferred to their own picks. In a redraft, Kelly would have gone at the same spot or higher, which is the definition of a good pick