Wrong advantage gives closer to +4 or +4.5 Eleven accuracy should give closer to +2 or maybe + 2.5 on top of that. Your numbers are low. You cannot simply look at the difference in average dice outcomes and divine anything other than the difference of the average dice roll UNLESS you assume an equal distribution for all ACs/chance to hit values which in 5e is actually a terrible assumption. In 5e the more extreme the AC the less likely you are to encounter it as there just are very few enemies with those extreme AC's. In other words the majority of what you fight will usually fall between 30% to 70% chance to hit (the -5 +10 feats change this). Your number doesn't account for this.
I agree with you that it is fairly strong. One way to assess it is to look at the combinations on the dice.
misses/out-of
10/20
10/20*10/20 = 100/400
10/20*10/20*10/20 = 1000/8000
1/20
1/20*1/20 = 1/400
1/20*1/20*1/20 = 1/8000
19/20
19/20*19/20 = 361/400
19/20*19/20*19/20 = 6859/8000
Say a group plays once a week for two years, making one check of each type per session. How many missed checks of each type might groups like that experience?
50% = all experience up to 40 missed checks, half experience up to 52
25% = all experience up to 16 missed checks, half experience up to 26
12.5% = all experience up to 6 missed checks, half experience up to 13
5% = all experience up to 1 missed check, half experience up to 5
0.25% = about one-in-five groups experience 1 missed check
0.0125% = about one-in-eighty groups experience 1 missed check
95% = all experience up to 92 missed checks, half experience 99
90.25% = all experience up to 86 missed checks, half experience 94
85.7375% = all experience up to 81 missed checks, half experience 89
For reference at 35% (the average chance of missing against CR appropriate foes) all experience up to 25 missed checks, half experience 36
All means >99/100 such groups,
half means >50/100.
Maybe that can be narrated by saying that most groups will hardly notice the change in frequency for critical hits, but groups should notice the change in frequency for ordinary hits. For example, missing every fortnight improving to missing twice per three months, to missing once per four months.