You seem to believe that small numbers invalidate a poll.
https://www.scientificamerican.com/article/howcan-a-poll-of-only-100/
Now I believe that polls have some issues but it has nothing to do with sample size but more to do with how they select their samples and how they phrase their questions.
There is also some selection bias coming from those polled. For whatever reason some that agree to participate are not giving the pollsters truthful answers. In a small sample size that can really skew the results.