Yes, I did take that into consideration. Pennsylvania was the best case for a voter in 2020 because it had the highest likelihood of being the tipping point state in that election. But I also tried to estimate the expected value of a vote in California and got a number around 60 cents. So the expected value can can vary by nearly four orders of magnitude depending on the state you are in.
The main factor in this discrepancy is not so much the number of electoral votes, but the partisan lean of the state. Am election that ends up being decided by a single vote in California would imply that there were extremely unexpected results in other states. But your expected value in Florida is much higher because it would not be unusual for the election to be decided by an extremely close result in that state.
Interesting - I went through the calculation based on "chance the election would be decided by one vote; my vote" and it's pretty damn vanishingly small. (I assumed any election decided by a single vote would be decided by my vote, which isn't entirely fair, and ignored appeals, etc), but didn't value it on election spending but instead on "changes to me".
The main factor in this discrepancy is not so much the number of electoral votes, but the partisan lean of the state. Am election that ends up being decided by a single vote in California would imply that there were extremely unexpected results in other states. But your expected value in Florida is much higher because it would not be unusual for the election to be decided by an extremely close result in that state.