I saw an article
today suggesting Donald Trump is in deep trouble this election. They
make some good arguments that this is the case, citing a general lack of
campaign staff and advertising on the part of the Trump campaign, but
it also ignores key variables like third party candidates and the
electoral college. That being said, is Trump really as bad off as the
article implies? Let’s take a detailed look.
Trump’s Numbers
The article correctly cites that Trump as poor poll numbers, hovering in the upper 30s. They compare this to previous campaigns’ numbers, and currently, Trump’s support is reminiscent of McCain’s numbers in 2008, when Obama and the democrats basically destroyed the republican party that year. However, as we know, this is not your average election year. Voter dissatisfaction with the two party system is very high, and a lot of polls that include Gary Johnson and Jill Stein have them getting a lot of support. I’ve seen Johnson score as low as 4-5 points but as high as 12. I’ve seen Stein score between 2 and 5%. As such, third party candidates are poised to take a much larger share of the vote than in previous elections, which can screw with the numbers. Still, all things considered, Clinton is, on average, 6 points ahead in the polls, with her reaching 45 and Trump only reaching 39. Feel free to look at Real Clear Politics’ aggregate polling data and make your own decision. I do think it spells good news for Clinton and bad news for Trump, but third party numbers and undecideds also vary a lot in some cases and could reduce the margin somewhat. In some polls, up to 20-30% of the electorate is unaccounted for. Some of these guys will decide closer to the election, some may support a third party, and some may stay home. I would keep an eye on the situation. This is not a conventional election year and a lot of unknowns are out there. Things may not be as they seem and things can theoretically change between now and then. We need to look at the electoral college to get a better idea of what to expect for this election.
The Electoral College
As we know, the electoral college is what really matters. As we know from Bush v. Gore, you can win the popular vote and lose the election. That being said, let’s get a good idea of what the electoral college looks like. There’s a lot of missing data, a lot of unknowns for some states, but here’s what we know. There are 50 states and a few other territories and the like. Many of these will definitely vote one way or another. Many small states in the south and midwest will support Trump and the republicans. Period. A lot of states with large urban populations will support Clinton. Out of the 50 states in the unions, only 12 or so are actually where the real battles for president will take place. Looking at RCP’s electoral map, we start out with the republicans having a near guaranteed 164 electoral votes and the democrats being guaranteed 211. This means the GOP needs to win 106 electoral votes and the democrats need 59. If we are looking at this as a literal race, imagine the democrats having a head start half way to the finish line. That’s what the situation looks like without even looking at the state of the swing states. Here’s what the actual swing states look like.
Arizona
Arizona is surprising. I would suspect that it would lean heavily republican as it generally goes republican in most recent presidential elections, but this puts it at having 1 point in Clinton’s favor on the aggregate. Still, with one poll being responsible for the majority of that lead, given its history, it’s very questionable who will
win here and we will need to keep an eye on it.
Colorado
This state voted for Obama in 2008 and 2012, but the one poll conducted seemed to learn heavily toward Trump. I would say this state is difficult to call because there’s only one poll from 7 months ago and it seems to deviate widely from recent voting behavior. Still, it did have a conservative lean previous to Obama, so maybe it will go toward Trump. I would like to see more data here too.
Florida
Florida is the biggest swing state at 29 electoral votes, and it could go either way. It seems to be 1.6 points in Clinton’s favor with most polls favoring Clinton slightly, but it could easily swing to Trump at this point.
Georgia
I don’t know why this is a swing state. Trump is consistently ahead and leading by four points here. It also has a heavy recent history of supporting republicans.
Iowa
Iowa doesn’t have any recent data, but it seems to lean toward Clinton. I find this plausible based on previous voting behaviors.
Michigan
This is another state I don’t know why is considered a swing state. It has been heavily polled and leans heavily toward Clinton. If Trump wins here I’ll be shocked, just as I’ll be shocked if Clinton takes Georgia.
Nevada
Nevada has no polling data at all. We have no idea what’s happening here. It has a history of voting for both sides so this is a wild card.
New Hampshire
New Hampshire seems to be leaning toward Clinton. I’d say she’s going to take it.
North Carolina
It’s pretty close here with only a 1 point average toward Trump, but given its voting history, I think a Trump victory is likely here. The only year I’ve seen it go blue recently was in 2008 when Obama won.
Ohio
It seems to be leaning toward Clinton but Trump has won some polls here. It will likely go toward Clinton but I consider Ohio to be very much in play.
Pennsylvania
Clinton, like most democrats in my lifetime, has a significant advantage. I’ve heard some people suggesting it’s in play, but I know from history living here, that it’s very hard for the republicans to win here. I’m inclined to believe she will win here based on the poll numbers.
Virginia
Virginia is leaning toward Clinton by a heavy margin too, which is surprising since I consider it to be very much a swing state.
Electoral Scenarios
Okay, now that we’ve examined the field, let’s plug all this data into some 270 to win maps and examine the most likely outcomes.
In my first scenario, I’ll give all states with a margin larger than two points to the candidates that are slated to win them. That being said, I’ll give Trump Colorado and Georgia, and Clinton Iowa, Michigan, New Hampshire, Pennsylvania, and Virginia. The other states I will classify as too close to call for this scenario. This leaves Arizona, Florida, Nevada, North Carolina, and Ohio as toss ups.
Scenario 1
Well, looks like Clinton has already won. She has 270 electoral votes and a solid advantage in the electoral college. Trump could win every other state and still lose the election. He’s screwed here.
Now, if we give Arizona, Florida, and Ohio to Clinton, with Trump taking North Carolina, the map looks like this.
Scenario 2
Now we have Clinton at 328 and Trump at 204. I did not give Nevada to either side since I don’t know how to call that, but even if we gave it to Trump for pity points, he would only have 210 electoral votes. If we flipped Arizona, it would only bring him up to 221.
Now, what would it take for Trump to win? Well, first of all, let’s go back to scenario 1. Trump would need to take every state still up for grabs there and then some. Let’s bring up the map all filled in there.
Scenario 1.1
This scenario involves Trump winning all states in which he is within striking distance. This involves Trump carrying the rest of the map, and Clinton losing all states she has below a 2 point lead in. Unlikely in itself, but possible. Even this would still make him narrowly lose the election. In order to win, he would need to carry all that, plus one state in which Clinton has a significant lead like Iowa, New Hampshire, Virginia, or Pennsylvania. If he loses Arizona, which he may, and Colorado, which is possible since Trump’s lead is derived from a single old poll, that puts him an extra 20 electoral votes in the hole, meaning he would need to win a combination of Iowa, Virginia, and New Hampshire. Maybe Pennsylvania as well, since that is just as easy as any of the above states would be. Here is what the Iowa, Virginia, and New Hampshire combination would look like.
Scenario 1.2
Take note that with those three states accounting for 23 votes, just winning Pennsylvania would not be enough. It’s only 20 votes. He would need to win a combination of the above states or somehow keep Arizona and/or Colorado.
Conclusion
Taking into account the 6 point lead in the popular vote and the various electoral college combinations, I would say Trump is screwed. It’s not impossible for him to win, but he is fighting a serious uphill battle and only wins in the most optimistic scenarios out there, which pretty much ignore many statistical hurdles which make such an outcome very unlikely. I would say Hillary Clinton will almost definitely win the election, with about 90-95% confidence. It is possible that she can lose, but you pretty much have to be denying reality to think it is extremely likely. That being said, yes, based on the evidence presented, Donald Trump is as screwed as some say he is. He’s six points down in national polls. While this is not devastating in and of itself, the electoral college, which already favors democrats, skews even more toward Hillary when one considers the poll numbers in the battleground states themselves. Trump could win every state where Hillary has less than a 3-4 point lead and still lose the election. It’s that unlikely. Trump make significant gains between now and November if he has the slightest of hopes of realistically winning the election.
For reference, I do plan on reviewing this information closer to the election and seeing how things change at least once, if not more. There are still a lot of unknowns here and the election can swing in unexpected ways in the next 4-5 months. It would be a good idea to look over this data some time in October or the beginning of November to see what’s going on.
Trump’s Numbers
The article correctly cites that Trump as poor poll numbers, hovering in the upper 30s. They compare this to previous campaigns’ numbers, and currently, Trump’s support is reminiscent of McCain’s numbers in 2008, when Obama and the democrats basically destroyed the republican party that year. However, as we know, this is not your average election year. Voter dissatisfaction with the two party system is very high, and a lot of polls that include Gary Johnson and Jill Stein have them getting a lot of support. I’ve seen Johnson score as low as 4-5 points but as high as 12. I’ve seen Stein score between 2 and 5%. As such, third party candidates are poised to take a much larger share of the vote than in previous elections, which can screw with the numbers. Still, all things considered, Clinton is, on average, 6 points ahead in the polls, with her reaching 45 and Trump only reaching 39. Feel free to look at Real Clear Politics’ aggregate polling data and make your own decision. I do think it spells good news for Clinton and bad news for Trump, but third party numbers and undecideds also vary a lot in some cases and could reduce the margin somewhat. In some polls, up to 20-30% of the electorate is unaccounted for. Some of these guys will decide closer to the election, some may support a third party, and some may stay home. I would keep an eye on the situation. This is not a conventional election year and a lot of unknowns are out there. Things may not be as they seem and things can theoretically change between now and then. We need to look at the electoral college to get a better idea of what to expect for this election.
The Electoral College
As we know, the electoral college is what really matters. As we know from Bush v. Gore, you can win the popular vote and lose the election. That being said, let’s get a good idea of what the electoral college looks like. There’s a lot of missing data, a lot of unknowns for some states, but here’s what we know. There are 50 states and a few other territories and the like. Many of these will definitely vote one way or another. Many small states in the south and midwest will support Trump and the republicans. Period. A lot of states with large urban populations will support Clinton. Out of the 50 states in the unions, only 12 or so are actually where the real battles for president will take place. Looking at RCP’s electoral map, we start out with the republicans having a near guaranteed 164 electoral votes and the democrats being guaranteed 211. This means the GOP needs to win 106 electoral votes and the democrats need 59. If we are looking at this as a literal race, imagine the democrats having a head start half way to the finish line. That’s what the situation looks like without even looking at the state of the swing states. Here’s what the actual swing states look like.
Arizona
Arizona is surprising. I would suspect that it would lean heavily republican as it generally goes republican in most recent presidential elections, but this puts it at having 1 point in Clinton’s favor on the aggregate. Still, with one poll being responsible for the majority of that lead, given its history, it’s very questionable who will
win here and we will need to keep an eye on it.
Colorado
This state voted for Obama in 2008 and 2012, but the one poll conducted seemed to learn heavily toward Trump. I would say this state is difficult to call because there’s only one poll from 7 months ago and it seems to deviate widely from recent voting behavior. Still, it did have a conservative lean previous to Obama, so maybe it will go toward Trump. I would like to see more data here too.
Florida
Florida is the biggest swing state at 29 electoral votes, and it could go either way. It seems to be 1.6 points in Clinton’s favor with most polls favoring Clinton slightly, but it could easily swing to Trump at this point.
Georgia
I don’t know why this is a swing state. Trump is consistently ahead and leading by four points here. It also has a heavy recent history of supporting republicans.
Iowa
Iowa doesn’t have any recent data, but it seems to lean toward Clinton. I find this plausible based on previous voting behaviors.
Michigan
This is another state I don’t know why is considered a swing state. It has been heavily polled and leans heavily toward Clinton. If Trump wins here I’ll be shocked, just as I’ll be shocked if Clinton takes Georgia.
Nevada
Nevada has no polling data at all. We have no idea what’s happening here. It has a history of voting for both sides so this is a wild card.
New Hampshire
New Hampshire seems to be leaning toward Clinton. I’d say she’s going to take it.
North Carolina
It’s pretty close here with only a 1 point average toward Trump, but given its voting history, I think a Trump victory is likely here. The only year I’ve seen it go blue recently was in 2008 when Obama won.
Ohio
It seems to be leaning toward Clinton but Trump has won some polls here. It will likely go toward Clinton but I consider Ohio to be very much in play.
Pennsylvania
Clinton, like most democrats in my lifetime, has a significant advantage. I’ve heard some people suggesting it’s in play, but I know from history living here, that it’s very hard for the republicans to win here. I’m inclined to believe she will win here based on the poll numbers.
Virginia
Virginia is leaning toward Clinton by a heavy margin too, which is surprising since I consider it to be very much a swing state.
Electoral Scenarios
Okay, now that we’ve examined the field, let’s plug all this data into some 270 to win maps and examine the most likely outcomes.
In my first scenario, I’ll give all states with a margin larger than two points to the candidates that are slated to win them. That being said, I’ll give Trump Colorado and Georgia, and Clinton Iowa, Michigan, New Hampshire, Pennsylvania, and Virginia. The other states I will classify as too close to call for this scenario. This leaves Arizona, Florida, Nevada, North Carolina, and Ohio as toss ups.
Scenario 1
Well, looks like Clinton has already won. She has 270 electoral votes and a solid advantage in the electoral college. Trump could win every other state and still lose the election. He’s screwed here.
Now, if we give Arizona, Florida, and Ohio to Clinton, with Trump taking North Carolina, the map looks like this.
Scenario 2
Now we have Clinton at 328 and Trump at 204. I did not give Nevada to either side since I don’t know how to call that, but even if we gave it to Trump for pity points, he would only have 210 electoral votes. If we flipped Arizona, it would only bring him up to 221.
Now, what would it take for Trump to win? Well, first of all, let’s go back to scenario 1. Trump would need to take every state still up for grabs there and then some. Let’s bring up the map all filled in there.
Scenario 1.1
This scenario involves Trump winning all states in which he is within striking distance. This involves Trump carrying the rest of the map, and Clinton losing all states she has below a 2 point lead in. Unlikely in itself, but possible. Even this would still make him narrowly lose the election. In order to win, he would need to carry all that, plus one state in which Clinton has a significant lead like Iowa, New Hampshire, Virginia, or Pennsylvania. If he loses Arizona, which he may, and Colorado, which is possible since Trump’s lead is derived from a single old poll, that puts him an extra 20 electoral votes in the hole, meaning he would need to win a combination of Iowa, Virginia, and New Hampshire. Maybe Pennsylvania as well, since that is just as easy as any of the above states would be. Here is what the Iowa, Virginia, and New Hampshire combination would look like.
Scenario 1.2
Take note that with those three states accounting for 23 votes, just winning Pennsylvania would not be enough. It’s only 20 votes. He would need to win a combination of the above states or somehow keep Arizona and/or Colorado.
Conclusion
Taking into account the 6 point lead in the popular vote and the various electoral college combinations, I would say Trump is screwed. It’s not impossible for him to win, but he is fighting a serious uphill battle and only wins in the most optimistic scenarios out there, which pretty much ignore many statistical hurdles which make such an outcome very unlikely. I would say Hillary Clinton will almost definitely win the election, with about 90-95% confidence. It is possible that she can lose, but you pretty much have to be denying reality to think it is extremely likely. That being said, yes, based on the evidence presented, Donald Trump is as screwed as some say he is. He’s six points down in national polls. While this is not devastating in and of itself, the electoral college, which already favors democrats, skews even more toward Hillary when one considers the poll numbers in the battleground states themselves. Trump could win every state where Hillary has less than a 3-4 point lead and still lose the election. It’s that unlikely. Trump make significant gains between now and November if he has the slightest of hopes of realistically winning the election.
For reference, I do plan on reviewing this information closer to the election and seeing how things change at least once, if not more. There are still a lot of unknowns here and the election can swing in unexpected ways in the next 4-5 months. It would be a good idea to look over this data some time in October or the beginning of November to see what’s going on.
No comments:
Post a Comment