But now you are completely disregarding both the direct and indirect costs of that extra damage.
(Being the -5 to attacks and not-taking-an-ASI-or-other-feat, respectively)
Such shallow analysis makes me regard your take as an emotional outburst, nothing more.
If you want to participate, how about giving your complete take on -5/+5? Most forumists seem to agree this is worthless in regular combat. The "+5" part is obviously good, but not good enough to compensate for the "-5" part. Not to speak of the other goodies the feat brings. And, to repeat myself, not to speak of the fact that you just did not take one of the other feats, or +2 to Strength or whatever.
Since most forumists also agree -5/+10 is way too much, I conclude that if the feat lay somewhere in between, it could otherwise be left untouched, and thus we would have a minimal change, which is often the best change.
It doesn't take much to realize that at level 4 when most PCs can first take a feat, it's not that hard to negate most of the -5. Even comparing it to +1/+1 of taking a stat boost, it's fairly easy to illustrate that without any attempt to minimize the penalty, these feats are stronger than a stat boost.
Simple example:
75% chance to hit normally (AC 11), 16 Str.
Level 3 DPR with greatsword: .7*10+.05*17 = 7.85
Level 4 DPR with +1/+1: .75*11+.05*18 = 9.15
Level 4 DPR with GWM (not using -5/+10): .7*10+.05*(17+7.85) = 8.2425 (not including the bonus attack 7.85 DPR if a foe is dropped but not criticaled, maybe one round in two to four so +1.9 to +3.9 more DPR)
Level 4 DPR with GWM (using -5/+10): .45*20+.05*(27+.45*10+.05*27) = 10.6425 (not including the bonus attack 5.85 DPR if a foe is dropped but not criticaled, maybe one round in three to six so +1.0 to +2.0 more DPR)
50% chance to hit normally (AC 16), 16 Str.
Level 3 DPR with greatsword: .45*10+.05*17 = 5.35
Level 4 DPR with +1/+1: .5*11+.05*18 = 6.4
Level 4 DPR with GWM (not using -5/+10): .45*10+.05*(17+7.85) = 5.7425 (not including the bonus attack 5.85 DPR if a foe is dropped but not criticaled, maybe one round in two to four so +1.5 to +2.9 more DPR)
Level 4 DPR with GWM (using -5/+10): .2*20+.05*(27+.20*20+.05*27) = 5.6175 (not including the bonus attack 5.35 DPR if a foe is dropped but not criticaled, maybe one round in three to six so +1.8 to +2.7 more DPR)
So GWM is stronger than +1/+1 in all cases except when not using the -5/+10 against a solo (or in a scenario where the melee PC cannot get to another foe with his movement in the same round). Every +1 to hit that an ally can give the melee GWM PC makes both GWM using and not using the -5/+10 even stronger than the +1/+1.
As for the -5 cost, that's fairly trivial come level 4 with spells like Bless around. Sure, Bless could make the +1/+1 case better, but the DPR would still be worse in that case. This is not very different then Sneak Attack. By himself, the Rogue rarely gets a spike of damage. It's mostly with the assistance of others. The same applies to GWM and Sharpshooter. The GWM PC does good without other's assistance, but when they do help him, he goes nuts.
And at level 5 with multiple attacks and +1 proficiency, the DPR delta becomes even greater. Monster AC does not keep up with proficiency, spells, stat boosts, and magic items. GWM just keeps getting stronger and stronger. Just with all of the discussions on buffing and flying the GWM PC, it should be fairly obvious that it's a tactic that works and works (too) well. If it's stronger than +1/+1 in the single attack per round case, it's even stronger in the multiple attack per round case.
Now, if you want to find the break even point for -5/+x at various levels, do the math yourself. I am not interested in a -5/+x solution.