We shouldn’t look at 2pt% and 3pt% and try to decide whether one outweighs the other when comparing players. There are pretty basic stats that do that perfectly for us, like eFG and TS. We also shouldn’t ignore free throws when trying to conclude whether one player is better than another. TS takes care of that.
If you really want to compare all those wings, why not use better stats? Barrett could be the best among them, but it's not great to base it on eyeballing 2pt%, 3pt% and ast/to.
Barrett:
.533 TS
11.3% Reb
21.6% Ast
1.5% Stl
1.1% Blk
11.5% TO
.223 WS/40
7.5 BPM
24.2 PER
Tatum:
.566 TS
12.6% Reb
12.4% Ast
2.3% Stl
3.2% Blk
15.0% TO
.169 WS/40
7.5 BPM
22.0 PER
Ingram:
.552 TS
10.8% Reb
11.4% Ast
1.9% Stl
3.6% Blk
11.3% TO
.157 WS/40
7.5 BPM
22.5 PER
Winslow:
.572 TS
13.1% Reb
13.2% Ast
2.8% Stl
2.9% Blk
14.2% TO
.196 WS/40
10.4 BPM
22.3 PER
Parker
.558 TS
17.1% Reb
8.6% Ast
2.1% Stl
4.0% Blk
11.9% TO
.205 WS/40
6.6 BPM
28.4 PER
Deng
.551 TS
12.6% Reb
11.8% Ast
no steal% data
2.3% Blk
14.1% TO
.195 WS/40
no BPM data
no PER data
None of this should be surprising, other than maybe the most recent three guys all having the exact same BPM. Barrett is clearly the best passer/playmaker, looking at Ast% and TO%. Rebounding was fairly similar between all of them, except Parker, but rebounding is largely context-based (who played the most time as the biggest guy on the court for Duke, who played the most as the second-biggest guy, how good were the rebounders around them, etc.?).
As for shooting, Barrett is predictably at the bottom of this group judging by TS, but you have to take into account the shots he's trying to take. Do you value an inefficient player more if he takes lots of shots because his team needs him to? What if he's playing with someone much better than him - do you still "reward" him for taking all those shots? Barrett has by far the highest WS/40 in this group, and his PER stacks up favorably against everyone except Parker. This reflects Barrett's usage, raw production per minute, and how much of a burden he tries to carry. WS/40 and PER rewards his and Parker's volume over efficiency. BPM focuses more on efficiency, and even there, Barrett holds his own, but he's not ahead of the pack.
Defense, as always, is the toughest thing to judge objectively. Based on Stl% and Blk%, Barrett looks like an awful, inactive, non-positive player on defense. People who have watched him extensively on defense might say this aligns with their eye test - he's often the weak link in Duke's defense. Defense is somewhat included in WS/40, PER and BPM, and based on those taken altogether, Barrett's defense isn't so bad that it outweighs his offense compared to these other guys. That also makes sense - it's almost impossible for a no-offense lock-down defender to make the same kind of difference a no-defense offensive star makes.
This would be a debate along the same lines as discussing Russell Westbrook's value in the NBA. Reasonable people can have widely varying views. You're going to get a view on Duke-Forum that is strongly anti-Westbrook, because most people here value efficiency more than anything else. If you want to pick one single stat to emphasize when comparing players, I would personally go with BPM, and I think it tells an accurate story here: Barrett is similar in overall quality to freshmen Ingram and Tatum, better than Parker, and not in the same league as Winslow.
There are limits to BPM's usefulness, obviously. Antonio Vrankovic and Justin Robinson are having better seasons than Barrett by BPM, for example. Comparing Barrett to his own teammates with more similar usage and more similar minutes, Barrett's 7.5 BPM is way behind Zion's 21.3 and significantly ahead of Reddish's 6.1. Again, this feels right. Tre is at 8.1, but I wouldn't compare Tre to those other three due to his much different role.