Here is an an academic paper draft manuscript about this topic. Apolitical and unbiased.
"Public Election Polls are 95% Confident but only 60% Accurate"
Quote:
ABSTRACT
In this paper, we investigate the calibration of public election polls. We present a definition of poll accuracy based on whether the poll’s margin of error spans the true election outcome. Most polls provide a 95% confidence interval along with the poll results, we sought to find whether their accuracy is as high as their confidence levels claim. Furthermore, we also wanted to see how this accuracy evolves over time as polls are conducted closer to the actual election. We find that even a week away from the election, polls claiming 95% confidence are only accurate 60% of the time. Moreover, we conclude that these polls would in fact need margins of error twice their reported width in order to be truly 95% confident. This provides a unique insight into the adjustment polls need over time and quantifies the systemic error polls contain beyond what the traditional statistics captures.
CONCLUSION
...
Our findings highlight the need for those communicating the results of election polls to adjust downward the level of confidence they claim, or to adjust upward the size of the reported confidence interval. If they wish to report well-calibrated confidence intervals, pollsters or reporters ought to report substantially wider confidence intervals. Current practice, which reports confidence intervals assuming that the only source of error is sampling error, overestimates the accuracy of polls.
...
|
Source:
https://psyarxiv.com/rj643/ -- worth a read - only 2 pages, if you don't understanding polling, margin of error, confidence levels, etc. this is a good start to jump right into it.
Authors: Don Moore (Decision researcher at UC Berkeley's Haas School of Business) and Aditya Kotak (Quantitative Trading Strats Analyst @ Goldman Sachs)
tl;dr
Either
- Change their methods so their polls WORK, and keep a high confidence interval (95%) and low margin of error
- Adjust their confidence interval DOWN so we know how unreliable the poll is
- Adjust their margin of error UP so we know their poll might be "right", but not super meaningful because the range of outcomes is so wide
2 & 3 don't help us get more accurate information about how people will vote. And (most) pollsters haven't changed their methods.
The 2016
STATE POLLS were OUTSIDE the margin of error in many instances, and sometimes by a multiple of the margin of error. The polls were...
WRONG