Assuming that random factors like "it rained" or "voters got in car accidents and couldn't make it to the polls" aren't a significant factor, there's always a 100% probability of one specific candidate winning since everyone has made up their minds before the day of the election. What polls do is not telling you the real-world probability, it's telling you the likelihood of a given outcome given known data.
Polls always need to be skewed in some way to be accurate, since not everybody will vote. You can't just get a random distribution of the population's preference and assume the more-preferred candidate will win. Polls can never be truly accurate because people will lie about which candidate they're voting for and whether they're planning to vote, and sometimes people who genuinely intended to vote never make it to the polls. There are a huge number of variables to consider when trying to predict the outcome of the election, but it's important enough that it's still worth trying.
Polls always need to be skewed in some way to be accurate, since not everybody will vote. You can't just get a random distribution of the population's preference and assume the more-preferred candidate will win. Polls can never be truly accurate because people will lie about which candidate they're voting for and whether they're planning to vote, and sometimes people who genuinely intended to vote never make it to the polls. There are a huge number of variables to consider when trying to predict the outcome of the election, but it's important enough that it's still worth trying.