Sunday, August 28, 2016

Clinton vs Trump.
28 August 2016,
72 days from the election

The big news in polling this week was from Ipsos. The company released two huge flurries of polls, 43 on Wednesday night and 40 on Friday. Some of those polls disagree with the conventional wisdom and some in the first batch disagree with some in the second batch. Most of the movement of states from one category to another is because of these polls being included. The general trend is in Trump's favor, but not massively so.

Any state that is underlined and italicized has moved from one category to another. If it is red, the change is in Trump's favor and if it is blue, the change is in Clinton's favor.

Solid Trump (more than 95% Confidence of Victory [CoV]): ND WY UT SD TX TN AK ID IN MS LA OK 
Total: 103

Leaning Trump (between 50% and 95% CoV for Trump): AL KS AR WV MO MT AZ SC NE KY GA  
Total: 88

Toss-up: None
Total: 0
Leaning Clinton (between 50% and 95% for Clinton): WI NC NV IA ME OH NM FL
Total: 93

Solid Clinton (more than 95% CoV for Clinton): OR MI MN PA CT CO MA DE NH WA VA NJ MD IL RI NY VT CA HI DC
Total: 254

The current count 


Ten of the twelve moves this week were from Solid to Leaning, both in the states favoring Trump and the states favoring Clinton. This is The Ipsos Effect, brought on by small samples sizes Ipsos uses in low population states. In the Confidence of Victory system, small sample sizes tend to move the Confidence of Victory numbers closer to 50%. The two moves that are not related to The Ipsos Effect  are Georgia flipping from Toss-up to Leaning Trump and Oregon moving from Leaning Clinton to Solid Clinton. Because Georgia has 16 electoral votes and Oregon only 7, these changes improved Trump's overall Probability of Victory number, though it is still less than 1%, making it four weeks in a row Trump's chances have not climbed above that ignominious threshold.

The fifteen states that have the most sway:

Current probability of victory if the election were held today:

Clinton 99.80%
Trump 0.20%

The Ipsos Effect has put some states in the top fifteen battleground positions that really aren't battlegrounds at all. This is a problem caused by an influx that will be solved by an influx of data, sort of a "lime in the coconut" situation. It changes the numbers in small ways but in no way changes the narrative. Trump is getting badly beaten. It is also true because I decided to use an algorithm that allowed fifteen entries and I'm always using all fifteen slots in my calculations. My best guess right now is the real battlegrounds are Georgia, Arizona and South Carolina, states where Trump now holds a lead, and on the Democratic side North Carolina, Ohio, Florida, Nevada and Iowa. The big problem Trump faces is that if that is correct and he runs the table with those listed stated, he still loses the presidency with just 265 electoral votes. If I'm right about this list and the noise is removed from the signal over the next few weeks, his chances will once again be best measured in parts per million.

Some Democrats worry about complacency in the voters, but I think the more likely problem is the Republican base falling into despair. Let me repeat my message from yesterday. If you are interested in donating money on the national level, it is currently a better investment to put your donations into the Senate races, and that is true for Democrats and Republicans alike. The numbers say what they have said for six weeks now. Clinton is delivering a major ass-kicking to Trump and even the media who dearly want a close contest are coming around to understanding this.

Back next Saturday with more numbers on the Senate.

Saturday, August 27, 2016

The Senates Races
27 August 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 most likely results right now show a tight race for control. My model looks at the 10 closest races are calculates odds for 1,024 possible outcomes checked as of this morning. There is a 35.99% chance of the Republicans having 49 Senators and a 34.27% chance of having both parties caucuses at 50. The next most like outcome is the Republicans having with 48 seats, which has a 14.71% outcome.

The races where the current leader has less than a 90% Confidence of Victory. There are only three races I would consider close, given the history of how often the Confidence of Victory method has predicted the correct winner.

Nevada: Harry Reid is retiring and the race to replace him is close. As of the most recent polls, Democrat Catherine Cortez Masto trails Republican Joe Heck, the Confidence of Victory numbers at 44.1% to 55.9%.

North Carolina: The Republican incumbent Richard Burr has a slight lead over the Democratic challenger Deborah Ross. The Confidence of Victory numbers right now put the race at 66.5% chance for Burr and a 33.5% chance for Ross.

Pennsylvania: Here, the Republican incumbent Pat Toomey is trailing the Democratic challenger Katie McGinty. McGinty's CoV is currently at 77.3%, compared to Toomey's 22.7%.

Slight spoiler alert for tomorrow's numbers. Clinton is still a prohibitive favorite using my model, though not quite as dominant as she was last week, barring polling changes tomorrow. If anyone is considering sending money to candidates on either the Democratic or Republican side, giving to specific Senate races or to the Republican National Senate Committee or the Democratic National Senate Committee is a way to get more bang for your buck. Control of the Senate really is a horse race right now, while control of the White House is more like the beating of a dead horse.

New post on Sunday looking at the race between Clinton and Trump.


Thursday, August 25, 2016

The Ipsos polling flood

The polling company Ipsos, affiliated with the newsgroup Reuters, released information late yesterday on 43 state polls on the Clinton-Trump race. This is by far the largest set of polls I've dealt with in a single day this year, but when stuff gets serious in October, this will be something close to an average day's work. Here are my impressions.

1. Okay, finally a challenge. My polling system does better when there are a lot of polls to work with, but this many polls means maybe an hour of data entry. It kind of amazes how much I don't mind this. At my age, I always look at how long a video is to decide if I want to watch it or not, but with a flurry of data, I don't even blink. When it's done I will know more than I did when I started. That's all the incentive I need.

2. More data is almost always better. There are now only 6 of 51 races that I have to extrapolate from the 2012 race. Three of the unknowns are dark blue races in District of Columbia, Hawaii and Rhode Island, while the others are in deep red Wyoming, South Dakota and North Dakota. These aren't likely to be close and they represent a grand total of 20 electoral colleges. My model always craves fresh data, but these missing contests are small morsels and I don't expect surprises.

3. Almost always better is NOT always better. One problem I have is that Ipsos decided on very small sample sizes in the smaller states. For example, Ipsos decided a sample of 115 New Mexico voters was sufficient. Clinton's lead here was only 3 points, which my system turns into a 63.7% Confidence of Victory value. In contrast, PPP also polled New Mexico, using 1,103 voters, showing a nine point lead, which turns into 99.98% CoV. These are the only two polls that exist, so right now the state looks like a battleground. One more poll in the next few days can either keep it as a battleground state or move it back to Solid Clinton.

4. Ipsos did produce some surprises. As I wrote in the paragraph above, Ipsos thinks New Mexico is close. Ipsos also thinks Pennsylvania is close, but this is a heavily polled state and one poll isn't going to make that much difference. This is what I like about using the median. Almost every surprise Ipsos produced says Trump is doing better than other polling says, including leading in Maine. The one unusual Ipsos result that favors Clinton is a 5 point lead in Missouri. Every other poll in Missouri this month has Trump leading and I doubt the Ipsos result is the start of a trend.

5. This data will be a factor for several weeks to come. Last week's polling results had Trump's chances rounding up to 0.01%, not even one chance in ten thousand. More polls in the next three days could change this, but right now Trump is back up to 0.08%, or about eight chances in ten thousand. Obviously, he has room to improve, but my best guess is that Oregon moving to Solid Clinton will be the death knell for his chances. Let me say that I'm guessing, but I'm guessing based on solid prior knowledge.

Okay, this is a whole lot of words about very few numbers. When I write again about the Senate races on Saturday and the electoral college contest on Sunday, expect a lot more numbers and a lot less words.

Sunday, August 21, 2016

Clinton vs. Trump
21 August 2016
79 days from the election

Numbers first, me yapping about the numbers second. A state that has moved from one category to another will be underlined, italicized and written in blue if the change was towards Clinton and in Red if the move favored Trump.

Solid Trump (more than 95% Confidence of Victory [CoV]): ND WY ID WV OK KY AL NE MS KS UT LA TN SD AR AK MT IN TX
Total: 145

Leaning Trump (between 50% and 95% CoV for Trump): MO SC AZ  
Total: 30

Toss-up: GA
Total: 16
Leaning Clinton (between 50% and 95% for Clinton): NV NC IA OH OR FL
Total: 81

Solid Clinton (more than 95% CoV for Clinton): MN ME CT NM PA MI CO DE NH NJ VA WI WA MD CA IL RI NY VT HI DC MA
Total: 266

The current count 


The big change from last week was a poll that moved Georgia from Leaning Trump to Toss-up. In this election, Georgia is actually a battleground state and Pennsylvania is not.

Let that sink in for a bit.

And then we have the 15 states this week that have the most influence on the race, which is calculated by the number of electoral votes times the chance of victory for the underdog. They are listed in order from most vital to least.

The fifteen states that have the most sway:

Current probability of victory if the election were held today:

Clinton 99.99%
Trump 0.01%

Yes, that's right, Donald Trump's chances of victory in an election held today are 1 in 10,000. They are so low, I will start measuring his chances in parts per million. Currently, he's at 108.5 parts per million away from being a 100% pure loser. 

I just wrote that Pennsylvania is not a battleground state, but it's on the list of states used in the calculation. Notice that Texas is on the list now, too. I don't think either of those states are really in doubt, but further evidence could change my mind. I also have a hard time believing just yet that Missouri and South Carolina are in play, but Trump really is that bad a candidate.

More than that, there are two states I expect to move when more polling comes in. Oregon is in the battleground group because of a single poll produced by a Republican pollster. Utah is in the Solid Trump camp because of a similar polling anomaly. My assumption is that by this time next month, Utah will be a battleground and Oregon will be Solid Clinton.

And now instead of yapping about numbers, let me finish by yapping about other people yapping.

People want to compare this election to others in the past, but no major party have ever had a presidential campaign run by talentless amateurs before. Trump is already talking about rigged polls, which is a great way to make your base feel hopeless and decrease voter turnout among your supporters. 

People in the pundit class on the right are expecting to lose in a landslide and aren't holding their tongues. Dana Perino, who worked as a spokesperson for G.W. Bush, just wrote an article telling conservatives that the polls are very likely correct. She believed the hype in 2012 that Romney would win. Bob Beckel, one of Fox News' lapdog liberals, convinced her that the unskewing of polls was a lie and of course, he was right. She learned a lesson, we will see how many follow her.

I didn't do the Senate numbers today, waiting for next Saturday to make that a weekly event. Right now, it looks like the GOP could keep 50 senators, which will make progress very difficult. This assumes Trump can keep on course to lose by about 200 electoral votes. The thing is, he's a bad spot and has ways it could get worse.

A report on the Senate next Saturday.

Sunday, August 14, 2016

Clinton vs. Trump.
14 August 2016,
86 days from the election

I will leave my commentary for the explanations of the numbers this week. Let's jump straight to the digits. States that have changed position are in italics and underlined. If they are red, they moved towards Trump and if blue, they moved towards Clinton.

Solid Trump (more than 95% Confidence of Victory [CoV]): ND WY ID WV OK AL NE KY KS UT LA TN SD AR TX AK MT MS IN 
Total: 145
Leaning Trump (between 50% and 95% CoV for Trump): SC MO GA AZ  
Leaning Clinton (between 50% and 95% for Clinton): IA NV NC OH OR FL 
Total: 85
Solid Clinton (more than 95% CoV for Clinton): MN ME CT NM PA MI VA DE WA NH NJ WI CO NY MD CA IL RI VT HI DC MA
Total: 262

The current count: 
Clinton 347 
Trump 191

The changes that look temporary: Utah went to solid Trump on the strength of a single poll from the only company putting Trump in the lead. Even Trump on the stump knows Utah is going to be tough. Trump is the stereotype of what Mormons think non-Mormons are like.

Missouri had a couple polls with Trump up by 10 in July, now his lead is 1 or 2. South Carolina is polled for the first time and Trump is up by 2. Prior history says these are comfortable wins for the Republicans. If Trump is truly struggling in those states, Clinton's electoral count could climb past 400, a number I still consider beyond reasonable expectations until serious evidence proves otherwise.

Iowa and Ohio are battleground states that made moves towards Clinton. North Carolina went from solid Clinton to leaning Clinton. They all deserve to be battleground right now. Again, evidence might prove otherwise over time. It should be noted that right now, Pennsylvania is NOT a battleground, but instead a regularly polled state where Hillary Clinton has had a comfortable lead for about a month.

The changes that looks permanent: Wisconsin is solid Clinton now and it will likely stay there. I know I have at least one reader who will be happy about this. New Hampshire is now solidly Clinton based on all the three polls taken in August. I'm starting to believe it's really in the Clinton camp, which is a little odd. Vermont and Hew Hampshire are neighbors and are usually worlds apart. Vermont is the hippie state, while New Hampshire is the "Get off my lawn!" state.

And now the odds if the election were held today, based on 32,768 possible outcomes in the fifteen closest states in terms of expected value. (Note: a lot of these fifteen states are not truly battleground, they are just the closest to being competitive.)

The fifteen states that have the most sway:


Current probability of victory if the election were held today:

Clinton 99.97%
Trump 0.03%

In the words of the comic books, this is not a hoax, this is not an imaginary story. Trump's odds in an election held today are 3 chances in 10,000 or 3,332 to 1. I knew that moving Wisconsin out of the leaning Clinton column could make a difference and sure enough, it did. My best guess is that Oregon belongs in the Solid Clinton column as well.

In closing, don't let anyone tell you this is close, even if his name is Nate Silver. Right now, this is an ass-kicking and Hillary Clinton has room to improve.

Another report next Sunday.

Saturday, August 13, 2016

The Senate races
The way too early edition: Vol. 1
13 August 2016: 87 days to the election

87 days until the election. If you are a political junkie like me - and if you are reading this in early August, face it, you are a political junkie - It feels like a three month sentence in the county jail and none of us actually did anything illegal. And just to pile on a little more bad news, the election won't be completely done come November.

Bear with me. Explanations are coming.

Why it's way too early: There are several stater where the contests aren't completely defined yet, waiting on primary results. Hawaii has their primary today, Alaska is on Tuesday. (All the forecasts I've seen have Hawaii deep blue and Alaska dark red.) On August 30, we'll get primaries in Arizona and Florida and it's not until September 13 when New Hampshire Democrats will decide for sure who faces Kelly Ayotte. All three of those states could be interesting.

And then there's Louisiana. They have the primary on November 8, a free-for-all open primary where the top two candidates regardless of party square off on December 4 in a run-off. This means we might not know which party controls the Senate until then.


The basics: Because the GOP did very well in 2010, they have to play a lot of defense in 2016. 34 seats are up for grabs and 24 of those are currently held by Republicans, only 10 held by Democrats. The current party membership is 54 Republicans, 44 Democrats and two Independents who caucus with the Dems. Let's run down the list so far, again stipulating it's way too early.

The Democratic side
No polls or way too early: CT HI MD OR VT WA
Polls show solid Democratic lead: CA CO NY
Polls show leaning Republican: NV

Harry Reid is retiring and currently the Republican has a small lead in two July polls.

The Republican side
No polls or way too early:AK AL AR AZ FL ID IN KS HY LA ND NH OK SC SD
Polls show solid Republican lead: GA MO OH UT
Polls show leaning Republican: NC
Polls show leaning Democratic: PA
Polls show solid Democratic lead: IA IL WI

Just to be clear, right now the Republicans would hold on to their majority with 51 seats, but it's WAY TOO EARLY. There are still 21 races we know almost nothing about. (Spoiler: There has been polling in New Hampshire, where the Republicans are in trouble, and Florida, where the Republicans would hold on if the election were held today.)

Another report in two weeks, which will still be too early.

Sunday, August 7, 2016

Clinton vs Trump.
7 August 2016,
93 days from the election

It was a bad week for Donald Trump. I fully expect that to be the lede on this blog many times between now and November. The numbers, nuts and bolts are at the end of the post, so if that's all you want, feel free to skip ahead. The beginning of this post is me playing Q&A with myself.

What were the big changes this week? The states that moved from one category to another are New Hampshire (Solid Trump to Leaning Clinton) and Nevada (Leaning Trump to Leaning Clinton).

How was New Hampshire in Solid Trump? A Republican polling company called Inside Sources has Trump winning big. I had never heard of them before, but that's because they re-branded after a disastrous 2012. There were only two polls in July, so they counted as half the median computation and swamped the other poll. A new poll made three polls and the Inside Sources numbers became just a poll to the right of the median. This is the way big swings happen in the median. It gets rarer as more polls get taken every week.

Is there any information that you need to make a better snapshot? The most glaring is the two week hiatus polling companies have taken in Ohio. There is no question that it is still a battleground state until evidence says it isn't, one way or another. 

You said you expect more bad weeks for Trump. Why? Bad things happen when he talks, and he's always talking.

I will admit I am obsessed with the election. I wish I wasn't, because it is not great for my mental health or blood pressure. I report the data I have accurately. A good week for Trump is going to mess me up if it is a major shift, but I don't know if that will happen and I'll be surprised if it is enough to swing the tide now.

Why are you skeptical he can swing the election his direction? Good question, multiple part answer.

1. His personality. Much like with Goldwater, people are pestering psychologists and psychiatrists for diagnoses. It's not that fancy. He's dishonest because lying and scamming have worked for him in the past. He is an egomaniac because he is the CEO of a company with no strong board of directors to reign him in and it is his nature.

I am more of a mathematician than a historian, but I can keep a lot of information in my brain. CEOs without restraining Boards of Directors are like absolute monarchs. Some were timid, some thoughtful, some egomaniacs and assholes. Trump is in the last category. Others of his ilk are Henry VIII, George III, Kaiser Bill and several of the Roman emperors. Trump is orders of magnitude more ignorant than the other people on this list.

2. His opponents understand his weaknesses very well and good information is spreading. He loves to talk in front of big crowds, but he must answer any slight that gets under his skin, which is nearly everything. The Clinton campaign and the media looking into his past are finding stuff that he has to respond to. Some might compare him to Pavlov's dog; I think he's more like a single celled organism.

And then there are the more clever insights. Of everything I have read this week, I give highest grades to Wil Wheaton, whose 21st Century claims to fame are equally as an actor and a board game enthusiast. He is the first person I read who came to the conclusion that Trump's constant plead "Believe me!" is actually a tell, a nervous tic that gives away information. He says it when he's lying. I give Wheaton full marks here because I hadn't seen anyone else say it before, and I'm fuming a little that I didn't see it myself. My sad excuse is that it's been so long since I played poker, my reading skills have gone to crap.

Let me repeat. My excuse is sad.

3. Trump is not running a campaign. It was obvious very early on that Ben Carson was really on a book tour and not a campaign to be President of the United States. Trump is also running a scam, but not a book tour scam.  This week, we found out clear recurring donations to his campaign were nearly impossible to turn off. (As a note of full disclosure, some left wing causes have the same bug/feature in their design, including Equality California, who wouldn't let me out of a monthly payment until the card was discarded for other reasons.) More than money, talking in front of big cheering crowds is a massive turn-on for him. He is making no effort to build his base; he doesn't want votes, he wants strokes, and cheering crowds give him big league strokes. (Aside: some people think he used the non-existent word "bigly" once. I am sure he said "big league". Let's not lie about him, the truth is bad enough.) 

4. Clinton can stumble, but to steal the spotlight from Trump's mouth, she has to fall off the side of a mountain. Bad things can still happen to Hillary's campaign, but to get the spotlight off the egomaniac Trump, it has to be a stumble that is a multiple week story with lasting consequences, somewhere along the lines of Romney's 47% comments, and even that may not be enough to turn the tide.

Enough yammering, now the numbers.

Solid Trump (more than 95% Confidence of Victory [CoV]): ND WY ID WV OK AL NE KY KS LA TN SD MO AR TX AK MT MS SC IN 
Total: 158
Leaning Trump (between 50% and 95% CoV for Trump): AZ IA GA UT
Total: 39
Toss-up (exactly 50% CoV): OH 
Total: 18
Leaning Clinton (between 50% and 95% for Clinton): NH NV OR FL WI
Total: 56
Solid Clinton (more than 95% CoV for Clinton): NC VA ME CT PA NM MI CO DE WA NJ NY MD CA RI IL VT HI DC MA
Total: 267

The current count: 
Clinton 323  
Trump 197
Toss-up 18 

The change from last week is that one Solid Trump and one Leaning Trump moved to Leaning Clinton. That is the direction the wind is currently blowing. More than that, if prior data, both in terms of previous years and the change from July to August can be trusted, Ohio will likely move to leaning Clinton and Oregon and Wisconsin will both move to Solid Clinton. While Ohio would be nice to know, if Oregon or Wisconsin makes the move and everything else remains static, Clinton has enough electoral votes in the Solid range to win.

To repeat: If Oregon or Wisconsin polling looks better for Clinton in the next few weeks, Solid Clinton makes her president, as in "Game over, man!" Trump's odds will shrink to less than a tenth of a percent.

And now the odds of victory. Here are the 15 states with the highest expected value, which makes them most likely to sway the tide. Of those 15, the 11 that really can be seen as battleground currently are written in order of most likely Trump first to most likely Clinton last. Ohio is a toss-up right now, so you can see the cut-off point.


Current probability of victory if the election were held today:

Clinton 99.6%
Trump 0.4%

Yes, that's right. Trump was at 1.1% in last week's snapshot and things got worse. Looking back at my data, Romney was as 0.4% when his worst gaffe completely sunk in, but he got something like a second wind after Obama had a bad performance in the first debate. The way things are going, I'm not sure there are going to be any debates.

I'll be back next weekend, both Saturday and Sunday. The Saturday posts will be about the Senate races.

Tuesday, August 2, 2016

Frequently asked questions about methodology

Where do you get your data? I use two websites, Real Clear Politics and Pollster, the website now owned by The Huffington Post. The articles on RCP lean to the right and the articles on Pollster lean to the left, but both are reliable reporters of polling results.

Are the any polls you don't use? In the past, I used any poll either of those two websites decided to print. I have started a new policy of not using data where the polling sample is only of "adults". I will use polls that ask the question if people are registered or likely voters.

Do you "skew" the results? Not exactly. Back in 2012 when the state polling data showed how much of a disadvantage Romney faced, a fellow named Dean Chambers decided to look at the mix of independents, Republicans and Democrats each poll used and "fix" the ones he thought undersampled Republicans. Doing this, he showed Romney with a comfortable lead. Conservatives loved this and Chambers became a hero. After the election when his methodology was shown to be defective, he admitted his error, but within a few months he convinced himself he wasn't wrong and there had been irregularities at the polls. Most people ignored him.

I do change the results I get, but I don't change who has the lead. For example, a poll in June from Nevada with a sample size of 300 showed Trump with a 47% to 45% lead. Multiplying the sample size by the percentages, this would say 141 people favored Trump and 135 people favored Clinton. I take these numbers and effectively ignore people who are undecided or voting for someone else. Of the 276 people choosing one of the two most popular candidates, Trump has 51.1% and Clinton 48.9%. These are the values I use to get a probability from an assumed normally distributed set. If you are familiar with Excel, the formula I use here is


This gives Trump a Confidence of Victory number of 64.10% in this case.

What about third party candidates? Confidence of Victory does have methods to deal with third party candidates or even more candidates. Unless the third party candidate is within a few percentage points of the two favorites, the CoV number for third place (or fourth place) becomes microscopic.

What if no polls exist? At the time this FAQ is being written, there are still many states that haven't been polled. In that case, I use the 2012 result for that contest and assume a sample size of 200.

What if more than one poll exists? I sort the polls by date and take the median of what I consider to be the freshest polls.

Why median instead of average? Median is not effected by polls that disagree with the consensus by a large amount, what we call outliers. Average does get skewed by these polls.

Define "freshest". In previous years, I usually started poll aggregation after Labor Day, but the conventions were early this year and so I start my collection of data in August instead of September. The polling companies have not ramped up to full frenzy, so many states that are considered battleground still have just a few polls. In ideal situations, I only look at polls no more than a week older that the most recent. In this early going, I stretched that out. In June, I'd use any polls from June and now, I usually stretch the freshness mark to two weeks if there has been a poll in July.

How do you create your probability of victory number? I choose 15 battleground states using this algorithm.

1) If the CoV number in a state is between 50% and 90% for the leader, that state is a battleground.
2) If this method gives us less than 15 states, states where the CoV is between 90% and 95% are added to the mix.
3) If we still fall short of 15, states with CoV over 95% are added on the basis of CoV times the number of electoral votes, which is called the expected value of the contest.

With 15 states, we have 2^15 or 32,768 possible outcomes. Each of these probabilities is tallied in the column of who ever wins the scenario. If a scenario is a 269-269 tie, I assume the House of Representatives will elect Trump, since the House currently has a large Republican majority. If Trump does well enough to tie, it is very unlikely the Republicans will lose the House.

What do you consider a toss-up state? Exactly 50%-50%. Any slight edge puts a contest in the column of the leader.

Does your system ever get predictions wrong? Yes, but in the general election it is rare. In the primaries, it's much more common for polls to get the result wrong. Bernie Sanders big win in Michigan primary is an obvious example from this year, and primaries with a lot of candidates can produce surprises as well. Polls do a bad job of predicting caucus winners, since the number of caucus voters is much smaller than the number who would vote in a primary. American elections are much longer than elections in most of the world and by November, the vast majority of people have chosen their candidate and the get out the vote campaigns are more important than ads trying change people's minds late in the game.

My other assumption about the bad performance of polls in primaries is the amount of time voters have taken to think about the candidates, especially in a multi-candidate race. The differences between the several candidates from the same party are usually not as stark as the differences between two candidates from different parties.

If you have any other questions, please add them in the comments. I'll try to add as many as possible to this list.