Clinton vs Trump. 25 September 2016, 44 days from the election

Another good week for Trump, but important states are moving both towards his column and away from them. More on that after the numbers. RED indicates a state moving from one category to another in Trump's direction, while BLUE means the state moved in Clinton's direction.

Solid Trump (more than 95% Confidence of Victory [CoV]): ND WY AL ID TN UT MT SD KY LA WV OK MS MO AK SC ME-2 KS AR IN TX  
Total: 160

Leaning Trump (between 50% and 95% CoV for Trump): IA NE GA OH NV AZ NC

Total: 77

Toss-up: FL
Total: 29
 
Leaning Clinton (between 50% and 95% for Clinton): CO MI PA WI ME  RI NH NJ

Total: 79

Solid Clinton (more than 95% CoV for Clinton): VA WA MN DE NM OR IL ME-1 CT DC NY VT MA HI CA MD
Total:193

 The current count 

CLINTON 272
TRUMP 237
TOSS-UP 29

The big changes this week were Nevada and North Carolina moving into the Leaning Trump camp and Florida leaving the Trump total and going back to toss-up.  With Clinton at 272, she cannot afford to lose a single state that now supports her, a daunting proposition. My model is a snapshot and it gives Clinton much better odds than many others do, but right now these are very small slivers of hope. I will explain below why those slivers are a little larger than you might think.

Here are the odds of victory if the election was held today, using the 15 battleground states according to my algorithm.

Trump's non-battleground count of electors: 122

The battleground states ranked by pqn: 
(Clinton%, Trump%, electors)

1. FL: 50%, 50%, 29
2. NC: 45%, 55%, 15
3. PA: 82%, 18%, 20
4. OH: 18%, 82%, 18
5. MI: 79%, 21%, 16
6. AZ: 37%, 63%, 11
7. CO: 71%, 29%, 9
8. GA: 10%, 90%, 16
9. NV: 34%, 67%, 6
10. TX: 3%, 97%, 38 
11. WI: 87%, 13%, 10
12. NJ: 94%, 6%, 14
13. NE: 10%, 90%, 5
14. IA: 7%, 93%, 6
15. RI 91%, 9%, 4


Current probability of victory if the election were held today:

Clinton 84.8%
Trump 15.2% 

It should be expected that Clinton is still favored, but it might seem counter-intuitive that my model still has her as a solid favorite. It's all about playing defense. Yes, Trump only has to steal a state and has several to choose from. Colorado is easiest right now, but Michigan, Pennsylvania and Wisconsin are in the realm of possibility. His odds are made longer because he also has to play defense, most especially in North Carolina and Florida, and to a lesser extent in Nevada. At the moment, Ohio doesn't look much like a battleground, Trump having the lead in eight of ten polls conducted in September. That said, Clinton has those kinds of advantages in the four states where she must play defense.

State: # of polls with Clinton leading in September of total # of September polls
CO: 6 of 8
MI: 5 of 5
PA: 6 of 7
WI: 5 of 5

There are two analogies I can point to right now, and they do not point in the same direction. Obama in 2012 had a bad first debate and his numbers fell for awhile, but after about a week and a half the numbers climbed back to their previous levels. The other is 2004, where it really came down to Ohio and Florida, and Bush only won re-election because he swept them. 2012 is an analogy in Clinton's favor if her numbers bounce back and it should be noted Obama's worst day had him still winning 278 to 260, then slowly climbed back over 300 electoral votes. The 2004 analogy should favor Trump, but Ohio and Florida in 2004 were both closely contested while none of the states he needs to steal currently are close.

Though my numbers are more confident than Nate Silver's, I'm still anxious seeing how close the Electoral College numbers are right now.  I'll be back next week after the first debate, which has every chance to be the strangest presidential debate in living memory.

It will also be October when next I write. There were times this year I thought it would never come.
  
  

The Senate Races 24 September 2016

This week's polling has been very good for the Democrats in the Senate. Last week, the Republicans looked to be favorites to keep their majority, but good news for the Democrats in North Carolina, Wisconsin and Pennsylvania has turned the numbers around. Last week, the single most likely event was a 51-49 advantage for the GOP. Now the most likely event is a 50-50 tie.

Here are the changes from last week.

Probability of Republicans holding the Senate outright: 37.5%, down from 54.3%
Probability of a 50-50 split: 32.1%, up from 29.4%
Probability of the Democrats taking the Senate back: 28.6%, up from 16.4%

Races that look close.
1. In North Carolina last week, incumbent Richard Burr was ahead of challenger Deborah Ross by 87% to 13%. This week shows a 50% to 50% tie.

2. Missouri is now the second closest race, though the numbers there did not move. The odds right now are 57% to 43% in favor of incumbent Blount over Democrat Jason Kander.
 
3. In Pennsylvania, challenger Katie McGinty is now ahead of GOP incumbent Pat Toomey 63% to 37%. Last week, this was a 50% to 50% call. McGinty looked like a favorite in August.

4. Kelly Ayotte still leads her Democratic challenger Maggie Hassan in New Hampshire, but the odds have slipped to 69% to 31%, marginally down from 72% to 28%.

5. Nevada is next closest and it favors the Republican Joe Heck over Democrat Catherine Cortez Masto, vying for the seat being vacated by Harry Reid. This would be a gain for the GOP. The probabilities of victory in an election held today are 74% for Heck and 26% for Cortez Masto, which is a gain for Heck over last week's 72% to 28% numbers. While the numbers have been relatively close, Heck has had the lead in five of the polls taken in the past three months, while one poll showed a tie.

6. Two new polls in Illinois show challenger Tammy Duckworth still leading incumbent GOP senator Mark Kirk, but the race got somewhat closer, now at 82% to 18% chance for the Democrats to gain a seat here.

7. In Indiana, incumbent Todd Young leads trails challenger Evan Bayh. This would be a loss of a seat for the GOP. The current odds are 14% to 86%, respectively, unchanged from last week.

In a race that looked close but no longer does, Russ Feingold's probability of victory has moved back over the 99% mark, a big improvement over last week's 89% to 11%. This would be a gain for the Democrats.

These races would say the Republicans are most likely to lose three or four seats, which would put them at either 51 or 50 seats, as shown in the line graph above. If the election were held today, it's in the realm of possibility the GOP would be the minority at 49 seats. Even that number could improve for the Democrats over the next month and half. It's not uncommon to see the public change their views in October in Senate races, when larger numbers begin to focus on these down ticket contests in a presidential election year.

Back tomorrow with the numbers in the presidential race.

Clinton vs. Trump. 18 September 2016, 51 days from the election

The long and short of it was that Hillary Clinton fell down, both literally and figuratively. Trump had his best week ever and is now in striking distance. The big question for the last seven weeks is if this is a trend or if Hillary can get back on track. Here are the numbers.

Solid Trump (more than 95% Confidence of Victory [CoV]): ND WY AL ID TN UT MT SD OK KY WV MS LA MO AR AK SC KS IN TX  
Total: 159

Leaning Trump (between 50% and 95% CoV for Trump): ME-2 IA NE GA OH AZ FL

Total:86

Toss-up: NC
Total:15
 
Leaning Clinton (between 50% and 95% for Clinton): NV CO MI WI RI NJ NH

Total:63

Solid Clinton (more than 95% CoV for Clinton): ME PA VA MN DE NM OR WA IL ME-1 CT DC MD MA NY VT HI CA
Total: 215

The current count 

CLINTON 278
TRUMP 245
TOSS-UP 15

Almost all the movement this week was in Trump's favor, gaining the median lead in both Ohio and Florida and bringing North Carolina into a 50-50 tie. Unlike 2004, where winning Ohio and Florida was just enough for George W. Bush to beat Kerry, the demographics have changed and Trump will need to get other wins in state that currently do not favor him. 

Here are the odds of victory using the 15 battleground states according to my algorithm.

 Trump's non-battleground count of electors: 122

The battleground states ranked by pqn: 
(Clinton%, Trump%, electors)

1. FL: 39%, 61%, 29
2. NC: 50%, 50%, 15
3. OH: 23%, 77%, 18
4. AZ: 36%, 64%, 11
5. MI: 85%, 15%, 16
6. CO: 71%, 29%, 9
7. GA: 10%, 90%, 16
8. NV: 66%, 34%, 6
9. TX: 3%, 97%, 38 
10. WI: 87%, 13%, 10
11. PA: 96%, 4%, 20
12. NJ: 94%, 6%, 14
13. VA: 96%, 4%, 13
14. IA: 9%, 91%, 6
15. NE: 10%, 90%, 5


Current probability of victory if the election were held today:

Clinton 89.7%
Trump 10.3% 


Instead of using the algorithm and looking at paths to 269 electoral votes, my choice of 11 states for battlegrounds going from most likely Trump to least likely would OH, AZ, FL, NC, NV, CO, MI, WI, NJ, PA and MN. This only brings his numbers up to 12.1%, not much of a difference and this is as generous a list as I can create in good conscience.

How can his odds be so bad when he only needs to get North Carolina in his leaning category and pull away one of several states from Clinton? The reason for this seeming contradiction is that he could lose his new won advantages and he has so many states in the leaning column he still has to defend.

Why would I think his gains of this week aren't certain? I looked at the numbers I compiled in 2012. After a disastrously low energy first debate, Obama's numbers against Romney sank in a similar way. After about a week and a half, they returned to their previous levels. A bad debate difference is not the same as a health scare, but if they are both temporary, I expect the effects will be temporary as well.

Earlier in the campaign, I wrote the race looks like an anaconda swallowing a pig: long, slow, ugly and inevitable. Right now it's doesn't look as inevitable, but I will wait a week to see if his current trend continues. With Hillary back on the trail, Trump defending his retreat from birtherism and his return to Twitter attacks, I have am not willing to call this race close yet.

Back next Saturday with information on the Senate.

The Senate Races 17 September 2016

There has now been polling in Arizona and John McCain looks comfortably in control. The changes this week that have made the most difference are in New Hampshire, where incumbent Kelly Ayotte has become the favorite, and Missouri, where Roy Blount's re-election in no longer a sure thing. 

Here are the changes from last week.

Probability of Republicans holding the Senate outright: 54.3%, up from 45.5%
Probability of a 50-50 split: 29.4% down from 34.2%
Probability of the Democrats taking the Senate back: 16.4%, down from 20.1%

Races that look close.
1. In Pennsylvania, GOP incumbent Pat Toomey was trailing challenger Katie McGinty in August, but is now in a flat-footed tie at 50% to 50%. These numbers are the same as last week's.

2. A surprising poll in Missouri says Jason Kander is now in a race with one term senator Roy Blount. Articles are crediting Kander's ad in which he assembles a firearm blindfolded, one of those valuable skills every senator from Missouri needs, I suppose. The odds right now are 57% to 43% in favor of Blount. Prior to this week it was about 91% to 9% in Blount's favor.

3. Nevada is next closest and it favors the Republican Joe Heck over Democrat Catherine Cortez Masto, vying for the seat being vacated by Harry Reid. This would be a gain for the GOP. The probabilities of victory in an election held today are 72% for Heck and 28% for Cortez Masto, which is a gain for Heck over last week's 56% to 44% numbers.

4. New Hampshire is the third closest and Kelly Ayotte has pulled ahead of her Democratic challenger Maggie Hassan. The odds here are 72% to 28%, respectively. This change is the biggest reason for the overall improvement in the Republican chances in the upper chamber.

5. In Indiana, incumbent Todd Young leads trails challenger Evan Bayh. This would be a loss of a seat for the GOP. The current odds are 14% to 86%, respectively, unchanged from last week.

6. In North Carolina, incumbent Richard Burr leads challenger Deborah Ross. This would be a hold of a seat for the GOP, so no net gain. The current odds are 87% to 13%, also unchanged from last week.

7. In Wisconsin, Russ Feingold is a solid favorite to beat Ron Johnson at 89% to 11%. This would be a gain for the Democrats, and the number is yet again unchanged for last week.

These races would say the Republicans are most likely to lose three or maybe four seats, which would put them at either 51 or 50 seats, as shown in the line graph above. I have set 95% as the arbitrary mark for a solid lead, but in my experience it's the races that are closer than 75% to 25% that are actually contests in doubt. Pennsylvania, Missouri, Nevada and New Hampshire would count by that standard.

Back tomorrow with the numbers in the presidential race.

Clinton vs Trump. 11 September 2016, 58 days from the election

OMG, it really is less than two months until the election. It's feels like it's been an eternity and it might never end.

First things first. It was a good week for Trump and he now has his best numbers since July. Those numbers aren't good by any rational standard, but he currently has the momentum.

And now the numbers, state by state, on in the case of Maine, district by district. (I'm still waiting for such precise polling from Nebraska, which also splits its electoral votes by district.)

Solid Trump (more than 95% Confidence of Victory [CoV]): ND WY AL ID TN LA MT SD WV MS OK TX KY UT AK IN KS MO SC AR  
Total: 159

Leaning Trump (between 50% and 95% CoV for Trump): NE GA ME-2 AZ IA   
Total: 39

Toss-up: FL
Total: 29
 
Leaning Clinton (between 50% and 95% for Clinton): OH NC NV CO WI RI MI NJ PA NH
Total: 116

Solid Clinton (more than 95% CoV for Clinton): ME VA MN CT OR DE NM WA IL DC ME-1  MA MD NY VT HI CA
Total: 195

The current count 

CLINTON 311
TRUMP 198
TOSS-UP 29

Grown-ups will be made nervous by Florida being a toss-up and Ohio being the weakest leaning Clinton state. All I can say to calm those worries is that right now, Trump has to run the table of all the states where he currently leads, and then win in Florida, Ohio, North Carolina, Nevada and Colorado.

Mathematically, that's hard to do, which is why my system now counts him as being far behind. I personally would hate to see him win Florida and Ohio, because it would give his candidacy
legitimacy it does not deserve, but winning those two states alone won't make him president.

A change this week. I not only list the 15 battleground states, but also the probability Clinton will win, the probability Trump will win and the number of electors, which in statistical terms are p, q and n. I also give Trump's count of electoral votes that are not in the battleground mix. This gives any keen nerd a chance to check my work.

  

Trump's non-battleground count of electors: 160

The battleground states ranked by pqn: 
(Clinton%, Trump%, electors)

1. FL: 50%, 50%, 29
2. OH: 54%, 46%, 18
3. NC: 55%, 45%, 15
4. AZ: 34%, 66%, 11
5. GA: 11%, 89%, 16
6. CO: 79%, 21%, 9
7. IA: 39%, 61%, 6
8. NV: 66%, 34%, 6
9. WI: 87%, 13%, 10
10. PA: 94%, 6%, 20
11. MI: 94%, 6%, 16
12. NJ: 94%, 6%, 14
13. NE: 10%, 90%, 5
14. RI: 91%, 9%, 4
15. VA: 98%, 2%, 13

Current probability of victory if the election were held today:

Clinton 96.5%
Trump 3.5%  

I set up my system to have 15 battleground states, but if I were using my own judgment instead of an algorithm, my current view of the true battlegrounds would be Florida, Ohio, North Carolina, Arizona, Colorado, Iowa and Nevada. If Georgia becomes closer for Clinton or Wisconsin closer for Trump, they would be on my true list as well. Using the shorter list, Trump's chances improve to 94.6% to 6.4%, still much longer odds than what Nate Silver is broadcasting to his much larger audience.

Fun stuff for data nerds only.
1. As I said earlier, I'm waiting for more info on the district by district numbers in Nebraska. Often, the Omaha district goes Democratic and the rest of the state goes Republican strong enough to overwhelm the "big city".


2. There are now three companies polling at least 40 states each, Ipsos, Google Consumer Surveys and SurveyMonkey.  Ipsos posts every week, but I decided my system wouldn't let any one company have more than one poll in the mix at any time, so only their freshest poll gets counted. Ipsos and Google have the unfortunate tendency to take very small polls in low population contests, which messes up my Confidence of Victory (CoV) method. For example, a 5 point lead in a sample of 500 voters give a CoV in the 90% range. If the lead is 5 points but the sample size is 200, that shrinks to 80%. Due to these discrepancies, I have made the arbitrary decision that the lowest acceptable data set size is 256, which is two to the eighth. I'm an OP (Original Programmer), and that's how I roll. 

3. The three big companies don't show a major bias in either direction, sometimes showing Trump closer than you might expect, sometimes doing the same for Clinton. On the other hand, there are pollsters at the state level, notably the newcomer Emerson and the once reliable Quinnipiac, who now skews Republican by assuming white people always vote and non-whites not so much. I take them at their word, reporting the numbers they publish, figuring the market will correct for them over time. Nate Silver's 538 has decided instead to "adjust" numbers for pollsters they don't trust completely. I am morally opposed to doing this. I can read cross-tabs as well as anyone who understands addition, subtraction, multiplication and division, but I don't adjust the numbers a company produces as long as the sample sizes are large enough.

In any case, I'll be back next weekend with more info on the Senate races and the electoral college. Thanks for stopping by.
 

The Senate Races10 September 2016

The polls this week mainly favored the Republicans. The race I would like to see most is Arizona, where McCain faces a stronger than usual challenger.

Here are the changes from last week:

Probability of Republicans holding the Senate outright: 45.5%, up from 26.1%
Probability of a 50-50 split: 34.2%, slightly down from from 34.3%
Probability of the Democrats taking the Senate back: 20.1%, down from 36.8%

Races that look close.
1. In Pennsylvania, GOP incumbent Pat Toomey was trailing challenger Katie McGinty, but is now in a flat-footed tie at 50% to 50%.
2. Nevada is next closest and it favors the Republican Joe Heck over Democrat Catherine Cortez Masto, vying for the seat being vacated by Harry Reid. This would be a gain for the GOP. The probabilities of victory in an election held today are 56% for Heck and 44% for Cortez Masto.
3. New Hampshire is the third closest and Kelly Ayotte is trailing her Democratic challenger Maggie Hassan. This would be a loss for the GOP. The odds here are 33% to 67%, respectively.
4. In Indiana, incumbent Todd Young leads trails challenger Evan Bayh. This would be a loss of a seat for the GOP. The current odds are 14% to 86%, respectively.
5. In North Carolina, incumbent Richard Burr leads challenger Deborah Ross. This would be a hold of a seat for the GOP, so no net gain. The current odds are 87% to 13%.
6. In Wisconsin, Russ Feingold is a solid favorite to beat Ron Johnson, but the current Confidence of Victory number is slightly under the arbitrary 90% mark at 89% to 11%. This would be a gain for the Democrats.



These six races would say the Republicans are most likely to lose three or maybe four seats, which would put them at either 51 or 50 seats, these are the two most likely results, as shown by the line graph above.

Back tomorrow with the numbers in the presidential race.

Clinton vs. Trump. 4 September 2016, 65 days from the election

A lot of polls this week, some favoring Trump and some favoring Clinton. The one big switch from last week is Iowa moving from Leaning Clinton to Leaning Trump. Several states that became Leaning Trump with the first flood of polls from Ipsos have moved back to where they belong in the Solid Trump camp, but some others move form Solid Trump to Leaning Trump, and there's a good chance that's where they belong. The states that I think are currently misplaced are Kentucky and Montana, that are only in the Leaning Camp because of single polls with small sample sizes. On the Clinton side of the ledger, Oregon, Michigan and Connecticut move to under 95% Confidence of Victory, but all of them could easily move back up over that threshold in the next few weeks.

Solid Trump (more than 95% Confidence of Victory [CoV]): ND WY MS TN SD ID LA AL AK KS IN WV TX MO OK 
Total: 127

Leaning Trump (between 50% and 95% CoV for Trump): GA AR AZ NE KY UT MT SC IA   
Total: 70

Toss-up: None
Total: 0
 
Leaning Clinton (between 50% and 95% for Clinton): NC ME NV OH NM WI FL OR MI CT
Total: 117

Solid Clinton (more than 95% CoV for Clinton): PA CO NH MN WA IL DE NJ MA VA MD NY RI VT DC CA HI
Total: 224

The current count 

CLINTON 341
TRUMP 197

My system has very specific rules for toss-ups and those rules make them rare. During the week Iowa went from Leaning Clinton to Toss-up to Leaning Trump. I have no idea who will be favored to win that state two months from now, but if I was forced to put up a wager, I'd put a small amount on Clinton.


And now we get to the odds of winning. My system takes the current numbers and chooses the 15 states most likely to sway the outcome, so the list changes from week to week.

The fifteen states that have the most sway:

NC FL OH SC IA AZ WI KY GA UT NV ME MI NE OR
 
Current probability of victory if the election were held today:

Clinton 99.95%
Trump 0.05%   

I am rounding the probabilities to the nearest hundredth of a percent because if I rounded to the nearest tenth, Trump's chances would have rounded to zero three times in the past five weeks, including today, when he barely rounded up to 0.05%.

How can Trump gain a state and get much worse odds than last week? In the words of The Mighty Monarch "Let him take all the pawns he needs!" Right now, the straightest path for Trump to victory would be to hold on to every state where he has a current lead - which is not a sure thing - and then to win North Carolina, Ohio, Wisconsin, Florida and one of three smaller states, either Maine, Nevada or New Mexico. Iowa has six electoral votes, so as the race stands right now, it's not a vital component. If Wisconsin changes and becomes more difficult to win than Florida is, even the pawns will be important, but that is not the case right now.

Short rant: Nate Silver's website is making this race seem much closer than it is and it seriously pisses me off. The Senate is close, Clinton-Trump isn't. I understand the need for clicks, but I'd rather the political journalists spent more time explaining how vital the Senate is to both parties and focusing on those contests, because that really is an exciting horse race that could go either way two months from now.

Clinton-Trump is exactly what it looks like. A professionally run campaign against a gang of corrupt goofballs who have no idea how to expand their voting pool. In other words, a major ass-kicking. Maybe something will change in the next two months, but it is extremely unlikely to change because Trump becomes a better campaigner. He loves ranting in front of adoring crowds, with not even the slightest clue he's not helping his cause. Sanders loved the crowds as well and he had actual professionals working for him. He didn't beat Clinton and Trump doesn't look likely to beat her, either.

Back next Saturday with another look at the very tight race for control of the Senate.
 
 
 

The Senate Races3 September 2016

We start with the basic numbers. The Republicans have 54 Senators as of today. A total of 34 seats are up for grabs, 24 of them currently held by Republicans and 10 held by Democrats.

Republican seat overview: 12 of the 24 races have polling data. Of those 12, the Democratic challengers are favored to win 5 and the Republicans to hold the position in the remaining 7.

Democratic seat overview: 4 of the 10 races have polling data, Of those, 3 look good for the Democrats and Nevada looks close, currently favoring the Republican candidate.

The overall odds looking at the ten closest races. The information in the first three paragraphs is exactly the same as last week. The important changes that happened are in the probabilities of victory in two races, Wisconsin and Ohio. Rob Portman's lead in Ohio looks stronger now, so the race is no longer one of the 10 closest and Ohio will very likely stay in Republican hands. Instead, the race in Wisconsin is the tenth most closely contested, with Russ Feingold still favored to unseat the Republican incumbent Ron Johnson. The chances for how many seats the Republicans will hold now look like this.

Here are the changes from last week:

Probability of Republicans holding the Senate outright: 26.1%, up from 12.6%
Probability of a 50-50 split: 37.0%, up from 34.3%
Probability of the Democrats taking the Senate back: 36.8%, down from 53.1%

Races that look close.
1. Nevada is closest and it favors the Republican Joe Heck over Democrat Catherine Cortez Masto, vying for the seat being vacated by Harry Reid. This would be a gain for the GOP.
2. New Hampshire is next closest and Kelly Ayotte is trailing her Democratic challenger Maggie Hassan. This would be a loss for the GOP.
3. In North Carolina, incumbent Richard Burr leads challenger Deborah Ross. This would be a hold of a seat for the GOP, so no net gain.
4. In Pennsylvania, GOP incumbent Pat Toomey is trailing challenger Katie McGinty. This would be a loss for the GOP.
5. In Wisconsin, Russ Feingold is a solid favorite to beat Ron Johnson, but the current Confidence of Victory number is slightly under the arbitrary 90% mark I count as the threshold between contested and not contested.

Back tomorrow with the numbers in the presidential race.