If you asked me a year or 2 ago I wouldn't have called it a mistake, but I had Randle and Smart on the same tier, definitely not now.
Really, that's Danny's biggest strength in drafting, when he has a top pick he hits on it, no Smart isn't a superstar, but there wasn't one there to pick and he hit a home run compared to what was available.
Ever since the 2013 draft, which was really a case of Ainge not willing to roll the dice when he should, he's done exactly that. Once you get over the fact that Giannis was still on the board, drafting Olynyk wasn't terrible, as he's 7th in win shares (basketball-reference.com) among draftees, and 26 teams missed on Gobert.
Going by win shares (which I know is a flawed stat in a number of ways):
2014: A very weird draft in that 3 second rounders and one very late first rounder join Embiid atop the leader board. But ignoring those players, who were essentially lottery tickets that hit for their teams, Smart slots right between Gordon (who was picked two spots ahead) and Randle (one spot behind), and 7th overall.
2015: Rozier has generated expected value with the pick. There were a couple of players on the board who have turned out better, but none are generational talents either.
2016: The WS stat will always hate Jaylen Brown, as do the Hield fanboys. Siakam and Brogdon were lottery tickets that panned out, and no board had Sabonis in the same category as Brown, Murray, Hield, etc. I think we can safely say that the C's are lucky they did not end up with the 2nd pick.
2017: The fact that Tatum is 3rd in win shares says more about the stat than the pick or the player.
2018: Hard to complain; the 27 slot has lots of Robert Williams's. Maybe passing on Mitchell Robinson was a mistake, but still too early to tell.