You just admitted polling can be wrong and has been. You proved my point.
No, I will try this again. Being off by 1 or 2 % is likely, being off by 13% is not. Is that easier?