My Photo

Ordering Information

Tomi on Twitter is @tomiahonen

  • Follow Tomi on Twitter as @tomiahonen
    Follow Tomi's Twitterfloods on all matters mobile, tech and media. Tomi has over 8,000 followers and was rated by Forbes as the most influential writer on mobile related topics

Book Tomi T Ahonen to Speak at Your Event

  • Contact Tomi T Ahonen for Speaking and Consulting Events
    Please write email to tomi (at) tomiahonen (dot) com and indicate "Speaking Event" or "Consulting Work" or "Expert Witness" or whatever type of work you would like to offer. Tomi works regularly on all continents

Tomi on Video including his TED Talk

  • Tomi on Video including his TED Talk
    See Tomi on video from several recent keynote presentations and interviews, including his TED Talk in Hong Kong about Augmented Reality as the 8th Mass Media


Blog powered by Typepad

« Panic Reset: Is There A Tightening Race or Is The Race TOTALLY Settled? Hillary vs Trump is.. Totally Settled | Main | Live Poll Decypher Blog: Day Before Election (and Election Contest) (updated 4x) »

November 04, 2016



This is how you use big rallies:

Clinton's concert tour isn't just about big crowds

"And already, the campaign is seeing signs that the concerts are having their desired effect — including when artists are paired with traditional political surrogates: Ahead of President Barack Obama's appearance with James Taylor, for example, the local operatives handed out tickets across the street from an early voting site in Fayetteville, North Carolina. Voting turnout there jumped 80 percent compared to the previous day, making it the single largest voting day there so far — and bumping up county-wide turnout 16 percent by itself."

I cannot get rid of the eerie feeling that the Hillary campaign has a hand in the recent media onslaught that Trump is winning.


Allowing people to vote is a sure sign the elections are rigged, according to Trump & co

Nevada GOP chair: Polling locations open late so 'certain group' can vote

"Taking the stage here, Trump then suggested that the polling location's extended closing time to allow voters to cast their ballots is a sign of a "rigged system" pitted against his campaign.
"It's being reported that certain key Democratic polling locations in Clark County were kept open for hours and hours beyond closing time to bus and bring democratic voters in. Folks, it's a rigged system. It's a rigged system and we're going to beat it. We're going to beat it," Trump said Saturday."


"Nate Silver 'correcting' poll results based on how well different polls predicted previous elections is both good and bad. Firstly, there would have to be systematic error in each pollster's method."

The real problem is that you have to assume that things did not fundamentally change. But if in all earlier elections the party members loved their candidate, and now a considerable fraction hates their candidate, that introduces a whole new skew.

All these polls and "corrections" seem to falter on voter turn out. Without a sane model for voter turn out, all will fall apart.



" and now a considerable fraction hates their candidate, that introduces a whole new skew."

Not only that, but that Trump followers and Republicans are groups with in large parts do not overlap. How that will play out is nearly unpredictable. It's very hard to guess how much hard-core Trump fans which are not already Republican voters there really are.

Tomi T Ahonen

Hi Everybody

Lets go back into two items in history. There was 2012. If you remember, after the election, Obama's team showed their internal polling vs the public polling. The public polling was wild, all over the place. The Obama polling looked like a old-style hat with a large brim - totally flat for a long time. Then jumped up, was flat, jumped down, was flat, till election day. And it hit the election result perfectly. It was that SAME election result already in their internal model in June.

The one up-tick, which was huge, and seemed to last initially, was the 47% comment by Romney. That was a game-changer. Obama would have won the election by 10% if they would have held to that - and remember, their internal polling held that level steady for about 2 weeks, no erosion at all. But this 'hat shape' then had the instant fall - back to the original level. Why was that? It was the first debate, that Obama lost.

Nothing else in 2012 moved the race, and Romney was the most flip-floppy politician ever measured up to that point, there was the long fight to get to see his tax returns, there were the last 2 debates that Obama easily won, and all the other stuff about women in binders etc. Nothing mattered. It was totally flat for months. One jump up (47%) then totally flat towards election day, except Obama lost his first debate, and the election dropped down to the same level it had been before, to 5%. And held that through till the end of the election. NO FLUCTUATION OTHERWISE.

It has to be a HUGE event to change an election, not the daily noise. And polling of 1,000 people has HUGE amount of NOISE. The race is not collapsing or jumping. It is steady. And if we take larger data sizes - like Reuters/Ipsos daily tracker, but run their polling up to levels of 5,000 people - its rock-steady no movement, 5% race. Very very very gradually inching up towards a 7% race. Totally steady. Yes, the 3 debates did create a gain for Hillary but the larger concurrent damage was to Trump about p*ssygate just before the second debate. And even those four events combined (3 debates + the sex story) only shifted the race by 2 points, if we take the large sample size and maximum view.

The daily story is the wild range of the nonsense polls like ABC/WaPo daily tracker that has gone from +12% Hillary to -1% Trump to now +5% Hillary in two weeks. Its nuts. That is not reality. But it DRIVES THE NEWS and the narrative.

So one, we KNOW from history, that the reality of the race is stable. The electorate gets to know the candidates - and in 2016 the two candidates were better known than ever in history, as a pair of rivals. The voters knew where this race is/was. It hasn't changed. What changed was the number of contestants. THAT is rare for a race, that we had third party candidates taking at some point up to 15% of the votes, combined (and today taking about 8%). That introduced genuine uncertainty to the election, but even with that, the race between Hillary and Trump remained utterly stable. Its primarily a function of the ceiling that Trump built, that he cannot penetrate. The 40% level. When we calculate women voters, black voters, Hispanic voters, and Trump's catastrophic voter performance in those three groups, he will do worse than Romney. That is inevitable. That is simply math. And Romney got to 47%. Trump will be lucky to get 40% on Tuesday. If we remove the propaganda pollsters like Rasmussen and IBD/TIPP and Gravis, that all have a systematic pro-Trump bias, and note, at least Rasmussen with enough data points, we can see they have INCREASED THEIR BIAS towards Trump consistently from 0% bias in July to 5% bias now - then Trump is at 40% today.

And whether Hillary gets 45% on Tuesday or 53% on Tuesday, depending on the undecided voters and any vote-shifters, if Trump is stuck at 40% he obviously is never going to win this election. And what is he doing? He was in Reno Nevada yesterday, saying its a bad thing that polls were kept open so Hispanics can vote! His campaign reacted to a court ruling about how the Trump Campaign has to stop intimidating voters - by saying they are going to APPEAL the court order AGAINST INTIMIDATION !!! The Trump campaign wants to go to court and insist they have a right to intimidate voters ????????

This is not going well for Trump. There was a path for Trump to get to 270 votes without Pennsylvania, but that meant among many miracles, to win Nevada. The early vote in Nevada has been so huge, in those parts of the state where the population lives and the election is decided - and driven by record Hispanic vote, that Nevada is done. Its won by the Democrats. Yes, there is still the actual election day on Tuesday and yes, the Democrats have to show up still on Tuesday, but this race is over. The only viable alternate map of extreme wishful thinking, of winning via bypassing Pennsylvania - is now destroyed because Nevada went with a Hispanic wave - a wave driven by Trump - that he still now is working to BOOST THAT WAVE. And that wave hits Arizona, Florida, Texas, Colorado as well.

So the main point is. The race is typically flat-even and isn't fluctuating anything nearly as wildly as the news tells us nightly - and when those pollster results are used in larger numbers, they typically tell a flat even non-bouncy story - which is less exciting for news. So they don't tell us that. Now lets take the other historic example we have. This cycle, we had that fantastic example of the Washington Post poll, that was given to 5 different professional pollsters to analyze - they got the RAW data before the poll was officially published, and each of the 5 pollsters gave their 'result' of the SAME data. They were off by 5 points at the extremes !!!

These are professional pollsters using the IDENTICAL NUMBERS and one can say, this is a 4% lead for Hillary, and another equally professional pollster can look at the SAME DATA and say, this is a -1% lead for Trump. Before they knew what others had said, just examining the data alone, professional pollsters can be as far apart at 5 points! Which is a very VERY revealing way to look at this year and see why someone can say, its a 5% race, and someone else can say, its a 0% race.

Do I know its a 5% race and not a 0% race, no. I can't know this. I don't do ANY polling myself (anymore haha). And I don't have any access to any of the raw data. But as I've said, for months, we have the NEAREST THING to access to the actual data, via Reuters Daily Tracking Poll. We can use their filters and calculate out the exact number of registered voter white women with no college degree; or the exact number of likely voter black men who earn minimum wage, etc etc etc. And Reuters/Ipsos daily tracker does over 6,000 interviews per month. Its FAR FAR more stable, on larger time-windows, and far FAR more likely to be close to accurate, than the various polls we get out every day that have a sample size of 800 people or 1,200 people. And the DATA of Reuters/Ipsos says, as I showed you in the diagram a few days ago - a TOTALLY FLAT RACE. A billiard-table. No contest AT ALL. Even the three debates that Hillary won by massive scale - and its rare for one side to win all 3 of the debates - yet the race didn't jump. The p*ssygate which seems to us, that it should be a more disqualiying thing than Romney's 47%, has moved the race much less than Romney's 47% did. Because why? Because Trump was stuck at 40% and his supporters will not desert him. He cannot win with 40% but he can't lose his supporters either.

Now. We know there is NOISE in all polling. The noise can be enormous as we saw in 2012 when the Obama data was overlayed on the polling data published by public polling. Obama internal data had only two shifts, they were immediate and big - the 47% comment helped Obama/hurt Romney, and then the first debate hurt Obama/helped Romney. And the net result was the race returned to form and was flat till the end of the election. And that end result was what they were polling in June before the conventions and before the debates. This race, as a 4-way race, had been about a 5% race before the conventions. It was about a 5% race in August, in September and in October. In very very rough terms, its a 5% race. If it has had any shift, that shift has been in Hillary's favor, ie seems to be about 7% now.

Now the Reuters/Ipsos data when run out to larger sample sizes, is consistent with a story that a, its a VERY flat race, and came from a 5% race in June, to a 7% race now. Its almost perfectly what Reuters/Ipsos data tells us too. The latest R/I gives us today, is daily data up to Thursday, Nov 3. If we take the last 5 days they have run 2,021 interviews (they have been growing the size of their daily interview polling as we get nearer to election day, to give even more accuracy) which gives the race as 43.6% Hillary vs 39.4% Trump with a 4.2% race. If we take the last 21 days to about the start of p*ussygate they cover 4,435 interviews and have the race at 43.2% for Hillary and 38.2% Trump and a 5.0% race. This is no movement. That is totally noise much inside the margin. That is a flat 5% race. If you really really want to dream its a tighter race, then yeah, at this rate, Trump would catch up to Hillary - IN MARCH of 2017 !!!!

No, a 'shift' of 0.8% in three weeks is NOT MOVEMENT. Its noise. Its far more likely that the race has not shifted. And both findings are consistent with Reuters/Ipsos data saying, .. its a 5% race.

Now to the Nate Silvers calculation of pollster bias, yes, there is certainly that, and it can be measured. We can make guesses about it now, we will know more on the actual result on election day (which is why some cowardly pollsters like Gallup and Pew have now decided not to dare to put their reputations on the line and they are not doing this test). But most elections are not wildly bouncing around. This election even more than most, seems exceptionally stable. And the data has consistently shown.. a 5% race. If anything, it has a slight opening up of the race towards the end, to about a 7% race, but suddenly, in the last 2 weeks, there is a rush of Republican propaganda pollsters (only) who insist, hey this race is tied!

It may be so. Its HIGHLY unlikely to be so. And these pollsters are VERY dubious to deliver that message. If we heard it from a reputable pollster that gosh, its a 0% race, like say CBS News/NY Times or Economist/YouGov or USA Today/Suffolk - then we could believe that might be happening. But none of the reputable pollsters have said this. Meanwhile what is our reliable daily tracker telling us? That its a 5% race (ok a 4% race). Meanwhile, what were these Republican propagandists telling us PREVIOUSLY? Rasmussen said it was a -2% race for Trump in early October and as massive as a -5% race for Trump in September! If you 'believe' Rasmussen now saying its 0%, - then Rasmussen has seen Trump LOSE FIVE POINTS in the last two months. Not gain anything! Rasmussen has found Trump stumbling into the last weeks...

Ok. I do think we'll see something in the 5% to 7% to 9% range in polls coming out today and tomorrow, the higher the better for Hillary but this is not a tight race at all. And her final election-day result will be better than these final polls, because of the several reasons we've discussed here. Now all we can do is wait. Wait first on Sunday/Monday for the last polling. And then wait, again, on Tuesday, for the actual election result. A little bit of a window into the direction of where the winds will blow, can be determined from voter turnout data in early voting which seems to be breaking records in many states.

Tomi Ahonen :-)

Tomi T Ahonen

Hi Everybody

And the first pollster is out. Its NBC News/Wall Street Journal and they are nailing their colors onto the wall at 4%.

They think Trump will hit 40% and Hillary will get 44%. Their poll two days before election day finds 8% undecided (which to me seems far too high).

Note that if we assign the undecided 8% proportionately to the 4 candidates, Hillary leads by 4.3%. If we leave Trump out of the undecided, and split the 8% into exact thirds, for Hillary, Johnson & Stein, then this poll would suggest Hillary's final margin to be 6.7%. If you want to believe in Trump's chances, you can see this result as a 4.3% election estimate. If you believe Trump has a ceiling he can't breach, then this suggests a 6.7% election result.

Also finally, remember, NBC News/Wall Street Journal has had a pro-Hillary bias this season. These numbers may over-count Hillary's actual performance. But we have the first numbers and NBC News and the Wall Street Journal believe the election will be a 4% election (should I remind you, that on election night 2012 the polling average said its a 1% race for Obama when it then after GOTV turned out to be a 5% race).

Lets see what other numbers we get

Tomi Ahonen :-)

Tomi T Ahonen

Hi Everybody

Ok and do please ignore that poll by NBC/WSJ. It is utterly totally batshit-crazy bullshit poll. It finds 4% result based on likely voters - which they define as 'having voted in 2012 or 2014, or being under age 24 and expressed willingness to vote'

The NBC/WSJ poll EXCLUDES ALL surge voters !!!! Any woman who now comes to vote for Hillary who hasn't voted for some years (or ever); plus any older Hispanics than 24 year olds who didn't vote last time, are EXCLUDED from the count. This vote will under-count the final result MASSIVELY.

Ok. They said 4% with those assumptions? Ok. Add 1% national vote gain for Hillary out of Hispanic surge, and add 2% for female surge (at least) and its consistent with a 7% polling result. Maybe more. Maybe a lot more...

Ok, we go back to poll-watching :-)

Tomi Ahonen :-)

Tomi T Ahonen

Following up with NBC/WSJ poll

Yea. Women same percentage as previous elections 53%/47%. This is wrong. There is already a measured increase in female vote.

Then demographics. White vote is 73%. Thats way too high. Why is it? Hispanic vote is too low. Duh. They have Hispanics at 10%. Thats the SAME LEVEL as in 2012. We have record Hispanic turnout. So this polling instrument is very crooked. It is SEVERELY under-estimating Democratic vote.

They exclude the women's voter surge. They have a voter imbalance that is nothing better for women than all past elections. This is the female vote year, and we have evidence of a female voter surge. That means they under-estimate Hillary's vote

They have excluded any voters who didn't vote in the 2012 Presidential election as not being a likely voter (unless too young to vote, in which case they asked). So any surge of any 'irregular' or unusual voters is excluded. That may not be a big factor most years, this year when there is a measured Hispanic wave, that is a vast error in their method. And they have yes, the Hispanic vote share to be only same as the Hispanic vote was in 2012. While we know its surging, surging big (and going to Hillary).

Yeah. On quick math that tells me add 1 point out of Hispanic and at least 2 points out of female voter surge, to this final result. The female voter surge may end up even bigger but Hispanic voters are such a small part of the total electorate that it probably wouldn't get to a 2 point change to this result. Women, could easily be 3 or even 4 points more. But its safe to say, this tool under-counts Hillary support by at least 3 points. When they say 4% race, they actually measured 7% but didn't know how to read their thermometer correctly..

Tomi Ahonen :-)


Here's an interesting, if logical, development:

"After spending 2016 trying to outmaneuver each other and deliver the next big break, hundreds of newsrooms are now engaged in unprecedented reporting partnerships to uncover barriers to voting and debunk fake news that can cause chaos and confusion on Election Day.

The biggest of the new alliances is Electionland, a project involving more than 400 newsrooms across the country casting aside competitiveness to share real-time data and tips on everything from reports about long lines and voter intimidation to hoax tweets suggesting stuffed ballot boxes."

Tomi T Ahonen

Hi Everybody

Florida early vote is analyzed rather deeply by Politico. And yet they kind of fudge the real issues by talking of a lot of percentages but not breaking out the absolute numbers, several which are stunning.

In 2012 Florida cast 8.4 million total votes (50% to Obama, 49.1% to Romney). By Saturday with one more day of early voting to go, Florida has already cast 6.1 million early votes! And of those votes, 1.5 million didn't vote in 2012. A small number of that would be young voters who were too young to vote last time, but most are voters who were not bothered before.

The Hispanic vote has traditionally strongly lagged their proportion of the electorate nationally and in Florida. They are 16% of the Florida electorate. In 2012 only 14% of Hispanics voted. In the early vote Hispanics have now voted at nearly 16%. And they have generated 900,000 votes. Of those Hispanics who voted now, 36% had not voted last time in 2012, ie 325,000 votes. If we assume this new vote splits by the recently-polled Florida Hispanic vote average, then it suggests 1% of the total Florida vote would now be gained by Hillary over 2012. Note, this is an unfair assumption because a lot of those who now joined are not traditional Cuban voters who are most Republican of all Hispanics in any state, but for example the Puerto Rican voters who are strongly Democratic. So it will be better than 1% in the total state vote, but less than 2%.

The race that was truly razor-thin in 2012, is now blown more open, into Hillary's favor mostly due to this strong Hispanic wave.

Note, that is only the early vote. But now an interesting tidbit to it that relates to our GOTV interests. The article also said that 25% of those Hispanics who voted early, had voted on election day in 2012. That means there is LESS WORK for Democrats to try to do, with a revised and stronger machine with more volunteers. I counted that it means of the total target voters that might be left for Democats to call on Tuesday, they are down at least 9% less to call because these 25% have voted early. Its likely a bigger gain, could be about 12% but I was using very conservative numbers. At least 9% less work to do, with a more powerful machine, and more volunteers this time. Yeah, I think its pretty safe to say, the Democrats will outperform in Florida this year....

And in very rough terms, this is the type of impact we should also see in other states that have large Hispanic minorities, ie Nevada, Colorado, Arizona, Texas. Nevada and Colorado are out of reach for Trump. Arizona is razor-thin, Texas is in play (but may be lost)

Tomi Ahonen :-)


FBI: Review of new emails doesn't change conclusion on Clinton


If Kurt Eichenwald and David Farenthold don't get Pulitzers outta this, we should just declare the U.S. a press-free zone.


And a bit of good news for Trump

A Transit Strike In Philly Could Lower Turnout, Especially Among Black And Poor Voters

What's wrong with this union? Do they want to make sure Trump gets elected? RCP shows only a 2.4% lead for Clinton in PA. This strike could really tip the balance Trump's way.


Early Voting Results: Who Leads State-by-State on Nov. 6?

I cannot really say what the numbers mean.


As long as they count voters by party affiliation alone, I don't think they mean anything. Due to the candidates' natures there will inevitably be more crossover voting than ever before. How, for example does a 'tight race' in Florida mix with 28% voter bleed among Republicans as was reported elsewhere? It can't be both, one of these has to be bogus.

It's also strange that the emails are getting played up this much again as a deciding factor - and the way they get played up indicate some suggestive polling being used to get the intended results.

And why have the polls 'tightened' everywhere when some more consistent polls Tomi provided show that this apparently does not happen? Is it again some propaganda polls influencing the results? Or is it just the news networks' desire to declare the race tight to have better headlines? After all, just applying different analysis can be enough to doctor a poll's result to one's liking.

All very hard to say. The only good thing in the article is the picture that shows that Trump virtually has to score a perfect victory in all relevant states to win the election, he may afford to lose one, but with two he'd be toast.

Wayne Borean

Talking Demographics, the military vote has skewed Republican for years. This year?

Donald J. Trump, Military and Political genius may have driven away some military voters. Here's another example of how HE knows better than the military. A private with one year under his belt would know why Trump's most recent statement is wrong.



FiveThirtyEight story has been updated to say that the strike has been settled as of this morning.


Trump is reported to be losing his mind. I always thought that the rumors about Hillary's health were diversions from Trump's own state. There are good reasons he did not share his medical records. I would not be surprised when they contain information about his mental health.

Limping to Election Day: Donald Trump is losing his Twitter account, and his mind
Donald Trump's lost his Twitter account. Maybe it's time to worry beyond just Tuesday

Inside Donald Trump’s Last Stand: An Anxious Nominee Seeks Assurance

"Aboard his gold-plated jumbo jet, the Republican nominee does not like to rest or be alone with his thoughts, insisting that aides stay up and keep talking to him. He prefers the soothing, whispery voice of his son-in-law.

He requires constant assurance that his candidacy is on track. “Look at that crowd!” he exclaimed a few days ago as he flew across Florida, turning to his young press secretary as a TV tuned to Fox News showed images of what he claimed were thousands of people waiting for him on the ground below."

Tomi already wrote about the possibility that Trump sanity would not survive a loss. That fear might become all to real.

Isceald Glede

Even before the election, Melania is delivering on her campaign promise to fight on-line bullying. Removing Donald's direct access to twitter is an amazing achievement, and I hope she able to continue with equally impressive results.


And more good news from Florida. The black turnout is back in business, thanks to Obama.

"Black turnout jumped substantially, boosted by Obama's multiple Florida visits and Sunday's "souls to the polls" voting drive led by black churches.
Black turnout in Florida will end up higher than in 2012, Daniel Smith, a University of Florida political science professor and election data analyst, predicted on Sunday."

Florida’s Early Vote Ends With Record Turnout In Democratic Strongholds

The comments to this entry are closed.

Available for Consulting and Speakerships

  • Available for Consulting & Speaking
    Tomi Ahonen is a bestselling author whose twelve books on mobile have already been referenced in over 100 books by his peers. Rated the most influential expert in mobile by Forbes in December 2011, Tomi speaks regularly at conferences doing about 20 public speakerships annually. With over 250 public speaking engagements, Tomi been seen by a cumulative audience of over 100,000 people on all six inhabited continents. The former Nokia executive has run a consulting practise on digital convergence, interactive media, engagement marketing, high tech and next generation mobile. Tomi is currently based out of Helsinki but supports Fortune 500 sized companies across the globe. His reference client list includes Axiata, Bank of America, BBC, BNP Paribas, China Mobile, Emap, Ericsson, Google, Hewlett-Packard, HSBC, IBM, Intel, LG, MTS, Nokia, NTT DoCoMo, Ogilvy, Orange, RIM, Sanomamedia, Telenor, TeliaSonera, Three, Tigo, Vodafone, etc. To see his full bio and his books, visit Tomi Ahonen lectures at Oxford University's short courses on next generation mobile and digital convergence. Follow him on Twitter as @tomiahonen. Tomi also has a Facebook and Linked In page under his own name. He is available for consulting, speaking engagements and as expert witness, please write to tomi (at) tomiahonen (dot) com

Tomi's eBooks on Mobile Pearls

  • Pearls Vol 1: Mobile Advertising
    Tomi's first eBook is 171 pages with 50 case studies of real cases of mobile advertising and marketing in 19 countries on four continents. See this link for the only place where you can order the eBook for download

Tomi Ahonen Almanac 2009

  • Tomi Ahonen Almanac 2009
    A comprehensive statistical review of the total mobile industry, in 171 pages, has 70 tables and charts, and fits on your smartphone to carry in your pocket every day.

Alan's Third Book: No Straight Lines

Tomi's Fave Twitterati