Just remember Obama and the polls saying he was going to win by 14 points.. he won by 7. Hillary and the polls saying she was winning by 5, she lost by 2. Biden and the polls saying he was winning by 10, he won by 4.
The polls are rigged for dem overconfidence.. be very very cautious!
??? I think you misremember poll results and final election tallies. Either that or you are cherry picking polls.
Polls had Obama up by an average of 7.9% over McCain:
https://fivethirtyeight.com/features/todays-polls-and-final-election/ He won by 7.2% (underperforming the poll average by 0.7%).
Polls had Obama up by an average of 1.6% over Romney:
https://fivethirtyeight.com/features/nov-5-late-poll-gains-for-obama-leave-romney-with-longer-odds/ He won by 3.9% (overperforming the poll average by 2.3%).
Polls had Clinton up by an average of 3.6% over Trump:
https://projects.fivethirtyeight.com/2016-election-forecast/ She got 3.0% more votes than Trump (underperforming the poll average by 0.6%)
Polls had Biden up by an average of 8.4% over Trump:
https://projects.fivethirtyeight.com/polls/president-general/2020/national/ He got 4.5% more votes than Trump (underperforming the poll average by 3.9%).
In half of your examples, the actual final error was less than 1% (not bad for polls that often are +-3% for each candidate). I'd personally call those spot-on. In 1/4 of your examples, Democrats did noticeably better than the polls. In 1/4th of your examples, Democrats did noticeably worse than the polls. But even then, well within the +-3% for each candidate (
remember that is ~+-6% for the overall result since you add the two errors when determining winner or loser*). That doesn't scream rigged in one direction to me.
* if candidate A is 48% +-3% and B is 52% +-3%, then the result could likely range from A 45% and B 55% (B wins by 10%) to A 51% and B 49% (A wins by 2%). The error range is B wins by 4% +- 6%. Notice how 6% is double the 3% of the individual candidate error.