Mez does blur very often. It's definitely not 100%. Probably closer to 60 or 70%. Actual blur spells, however, blur more often.
https://wiki.project1999.com/Memory_Blur
Whatever that modifier is gets adjusted significantly. If mesmerize is a 1% chance and works 60% of the time, a 10% chance might work.. 66% of the time. Or perhaps it fails a tenth as often. Perhaps there are calculations taking place that are not immediately obvious. Perhaps there's no perhaps involved, and enchanters have known this for some time.
It's definitely *not* a 10% chance. Nor is Blanket of Forgetfulness a 20% chance.
https://wiki.project1999.com/Blanket_of_Forgetfulness
In fact, I'd go so far as to say that Blanket might apply its blur chance twice by that descriptor, which is why I've had it fail *MAYBE* once, ever.
All blurs are performing well beyond the listed percentage.
Furthermore the blur on Mesmerize might be powerful, but that is often a DOWNSIDE also. Spamming mez to interrupt or lock things down while soloing, for example, can give mobs health ticks. It also doesn't blur mobs that are immune to mez, as far as I can tell.
Memblurs are also beneficial spells. Not hostile. They do not resist, and are significantly more useful as a result.
No, I don't need proof, because I'm right, and so is the server config for memblurs. Evidence from the past likely doesn't realize that the "blur" effect is significantly more noticeable on pacified or far away mobs. The "Higher level less effective" commentary is likely a symptom of larger aggro ranges, or closer quarters.
EDIT: If you want to know the real reason enchanters are batshit broken, see: the exploit known as GCD resetting.