I would warn against relying on exit polls as they are very inaccurate, even less than normal polls. The exit polls in 2016 predicted a Clinton landslide (Trump 48-47 for instance) for example
Early exit polls are trash (for obvious reasons) but the ones later are OK. I did the math and most of them check out.
Por ejemplo, in Arizona, the exit polls showed a 75% white electoral with Trump getting 54% of whites, with 32% of non-whites (25%). This adds up to 48.5 and Trump got 48.1 in the state.
Not 100% accurate, but that's not what polls should be used for - just an idea. It's one of the biggest problems people have with polls. They think polls should be 100% accurate when that's not possible (not saying this about you, but for others).
Also, exit polls are for showing why people voted, not who they voted for. Topline numbers in exit polling are crap but the crosstabs are valuable
