# What is the minimum percentage of the popular vote required to win the electoral vote in the USA?

Asked by ETpro (34360) December 3rd, 2012

Small states have a decided advantage in the number of electoral votes per capita they cast in US presidential elections. If the electoral vote winner won every tiny, winner-take-all state by just one vote, and big states went for 100% for the loser of the electoral vote, how skewed could the election results get? I heard a commentator on a TV show claim that it would be theoretically possible to win the electoral vote with just 22% of the popular vote. Is our current Electoral College system really that crazy? If it is, how great a margin do you think would be needed before the public demanded pure representative elections?

Observing members: 0 Composing members: 0

I can’t give you a number but it is pretty small. And one does not need 50%+1 in the states, just a plurality in many of them. Many of the states in 2000 were won by plurality because Nader took enough that nobody could get 50%.

But despite the quadrennial grumblings about the electoral college, going purely popular vote would just shift the dynamic from the farm states to the industrial states. Whichever party wins the Presidency is generally okay with the electoral college, and won’t support changing it.

zenvelo (29440)

The Founding Fathers didn’t want to have people elected to office by pandering to the large states, and ignoring the small ones. Yes, the small states wield more power per capita, but no candidate can afford to ignore either the small or large states.

filmfann (43215)

It also depends on how many candidates there are. So it could be 50% plus 1 in enough states to get 270, and not a single vote in any other state. But if there were 3 candidates and their votes were split evenly, then you’d only need 32.33% plus 1 in those states, and not a single vote in any other state.

These are theoretically possible scenarios, but they are very unlikely. If there were five candidates, the amount would be even less.

It is important not to get hung up on the overall vote percentage. There are other factors involved. The reality of the matter is that anyone who wins is going to get a significant plurality of the votes, if not the overall majority.

wundayatta (58484)

@wundayatta Except in 2000, when Al Gore had 543,816 more votes than GW Bush.

zenvelo (29440)

Yes, @zenvelo. In that election George Bush got a significant portion of the votes, along with a majority of the electoral votes. He did not get an electoral majority, but he got in the high 40s, percentagewise.

wundayatta (58484)

I’ll try and figure it out.

So the electoral college benefits states with low population moreso than those with high population, in the sense that each state gets two “free” electoral votes that do not depend on population. 2 votes are worth much more to a state with a low population, like Montana, where those two votes are 67% their electoral weight, than in a state like California, where 2 votes are a drop in the bucket. This means that the worst case scenario you’re looking for would happen when a candidate wins a lot of small states rather than a few large ones. An easy way to understand this is to realize that if a candidate wins many small states, he is gaining a lot of those “free” 2 points, points which don’t actually have percentage points of the popular vote attached to them.

A candidate needs 270 electoral votes to win, so adding up starting from the smallest values until we reach 270, this means the prez would need to win MT, WY, DC, AK, ND, SD, VT, DE, HI, ID, NH, RI, ME, WV, NM, NE, IA, KS, UT, NV, AR, MS, OK, OR, CT, KY, LA, SC, AL, CO, WI, MO, MN, MD, AZ, IN, TN, MA, WA, VA, and NJ. This is a total of 272 electoral votes, if I did everything right.

So of course the candidate needs only just above half in each state to win it. I’m going to ignore the special rules in NE and ME for now. Now based on this census information, this is a total of about 71596373, out of a total of 312,913,872 people in the US, a percentage of about 22.9%.

WOW, what an interesting question! I had no idea it could be so low.

Mariah (22836)

I found a link that I can’t get to post that shows the formula for working it out and the answer is 28.51. It is from this site: math.arizona.edu/~voting-theory/chapter04.doc

bkcunningham (18419)

@bkcunningham, your link was very interesting, but I don’t like the methodology used. I don’t understand their reasoning for choosing to use the 11 most populous states, when the true worst case scenario would be for a prez to win 40 of the least populous states. This is why my percentage turned out smaller; I do not think the percentage calculated in your link is the true absolute minimum.

Edit: Oh wait, I saw the number you cited (28.51%) and stopped reading. If you read a tiny bit below that, they go on to say, as I did, that using the least populous states produces the true minimum, and by their calculations with that method they say the true minimum percentage is 21.84%. Interesting to say the least! I think I know why there is a discrepancy between this number and mine, too. Since the total electoral votes that I used was 272, probably some strategic trading could be made to get that down to exactly 270 and find an even lower number than 22.9%.

Mariah (22836)

@Mariah Your total people is not a valid number for the equation, because the total includes people under 18 and people who are not citizens (they can be legal, but still not citizens and then also we have some illegal aliens of course).

JLeslie (52557)

Ahh of course silly me. I did this in a bit of a rush. If I find the time later I will amend my calculations.

Mariah (22836)

@Mariah & @bkcunningham Amazing. So the guy who said on TV that in a 2-person race, a candidate could win the electoral vote with 22% of the popular vote was right in the ballpark given rounding for ease of discussion. I greatly appreciate the effort you both put into that. I searched without success before asking.

@filmfann Otherwise, what are question details for. If all the question is contained in the question, no need for details, no?

ETpro (34360)

@ETpro No, Mariah’s number is incorrect, see my answer to her. 23% of the US is under 18 accordng to what I read, so we math in my head that reduces the Total voting population to about 230 million and then reduce it again by non-citizens which I think is around 30 million, not sure, so if my numbers are close that would be 200million people can vote in the US making the percentage just over 35%. Again, all in my head, and tons of rounding, so someone should really check my math and verify the stats, but even if I am off by some it is still higher than the 22%.

JLeslie (52557)

@JLeslie You guys are going to make me go carefully run the numbers myself, aren’t you. Good grief, I thought Fluther existed so I wouldn’t have to do that. :-)

ETpro (34360)

No @JLeslie, keep in mind it wasn’t just my denominator that was wrong, my numerator did not take into account voting age/citizenship either. So the number will not necessarily go up.

In fact, I’d bet that numerator and denominator are probably off by a similar percentage, changing the final answer very little.

Mariah (22836)

@Mariah Of course you are right! Why didn’t that occur to me? If we were taking into account just a few states I would do further research into exact numbers, because some states have many more new immigrants, and some states have lower or higher than the average of young citizens, but since it is so many states making up the equation it probably all evens out.

JLeslie (52557)

Probably it does, but also it did occur to me that the 40 states included in my estimate include pretty much all of the southern states, where illegal immigration might be disproportionately high. Even so, I’m guessing the discrepancy is quite small; I’d put money down that 22% is a pretty accurate estimate.

I did notice that in my rush, I used as my denominator a number in the table that is not only not the most recent census data, but included Puerto Rico et. al. The denominator should be 311,591,917, but this still doesn’t correct @JLeslie‘s objection that I’m including under 18 citizens and other non-voting people.

This link could possibly be useful for getting a more accurate estimate.

Mariah (22836)

@Mariah & @JLeslie Argh! Now I’ve got to factor in felons and which states prohibit them form voting. If I want to get super accurate, I have to include those states that prohibit released felons from voting unless they meet certain requirements, and I must research how many felons have done so in each state. Also, how many of any given state’s adult US citizens have actually registered to vote. This is complicated. I’ll take care of the shopping.

ETpro (34360)