What Is the Election Needle? And Why Will We Have 4 of Them Tonight?

Feb 03, 2020 · 109 comments
SG (NY, NY)
Once again, after the Hillary debacle, the needle was completely wrong even after 70% of results came in. The disorganization in Iowa is partially to blame, but clearly the statistical model used for the needle is flawed and suffers from inherent bias. Nate Cohn is not unlike the DNC in this respect. I agree that the needle should be retired.
Tom (Washington)
If you insist on using the needle please include a way to see the needle’s history. The needle is worthless unless it can be judged on how it moves - with increasingly precise certainty (I’ve never seen this), or wild out of control fluctuations reminiscent of a 737 MAX shortly after takeoff (Never to be trusted again).
Kaori (Tokyo,Japan)
You know,I (We all except Trump supporters)have trauma for this needle. I have reservation to believe this because I remember How this needle changed 180 degree (this is maybe too much,I don't have record)from Clinton to Trump in short time,on Nov.2016.
Paul (Brooklyn)
The Iowa caucus used to be a beauty contest of minimum worth of predicting who would win the nomination let along who would win the election. A few times due to the esoteric nature of the thing, it wasn't clear who really won. Now it is a complete joke and laughing stock of the country. They should change the name to the Iowa carcass. It would be more fitting.
tom (midwest)
Here we sit, still waiting the next day. Congratulations Iowa.
mmk (Silver City, NM)
The Iowa Democratic Party has had four years to fix up their caucus reporting procedures and this monumental mess is the result? Trump is going to campaign bigly on their incompetence. Add to it the refusal to release the latest big poll and you have to wonder what is going on? This is a terrible way to start the primary season and will hurt the Democrats in the general election. I hope state party leadership and Tom Perez get the sack.
paul (New Bedford, MA)
Please get rid of your Election Needles, "Who Will Win?" In 2016, people saw this likelihood of Hillary winning (did it get as high as 80%) and thought that gave them a bye to not have to vote for her, because the meter says its so sure she'll win. So how many voters in Wisc, and MI voted Democratic down the ticket, but left the President blank? Enough to carry the margin on one of those states. I find the meter image to be a very deceiving way to present predictions. I had an awful flashback when I signed on to see Iowa results. I truly admire the graphics at the NYT, but the meter should not be used. Just report votes and percentages when reported from actual locations. You can't predict and it could work against Democrats again. Thank you for considering my request seriously.
Rob (West)
Sanders will never win- I do not understand the idealism of supporting candidates that cannot win.
Marien Ett (OR)
Tradition relies on history. Change relies on strength. Trust relies on truth. Passion relies on conviction. Perspective relies on point of view. Reality relies on what is real. Is this real?
jb (ok)
I was traumatized by the needle in 2016. It makes me queasy still.
Chicago Guy (Chicago, Il)
An election needle? Really? From the people that told us on the day of the 2016 election that Clinton had a 89% chance of winning? I'll pass.
Sarah (San Francisco)
@Chicago Guy There's an 86% chance that a random day is not a Tuesday. 89% is not a high enough probability to conclude there was a probability error based on one data point.
Gary FS (Avalon Heights, TX)
The so called "needle" is irresponsible journalism at its worst. It is not based on scientific sampling and statistics. In fact it doesn't rise above the level of a Vegas craps table or racetrack futures. It wasn't the polling that was so wrong in 2016, it was all the infernal "needles" predicting a 93% chance of a Clinton victory. It had a devastating negative impact on Democratic turnout, and it gave voters the illusion that they could safely cast a "protest" or a joke vote on Trump. Nate Cohn is nothing more than a modern soothsayer. We'd have been better off sacrificing an unblemished he-goat in 2016 than listen to him. Shame on the NYT for running this shabby excuse for reporting.
Bernie (LA)
The 2016 NYT needle lulled me into thinking that Hillary Clinton would win. Sometimes the needle said she’d win by a little, sometimes the needle said she’d win by a lot. You printed it, and I trusted it. I’m happy to say that I have learned from my mistake. Pity that the NYT has not.
RN,PhD (NYC)
Iowa, this is so irrelevant as to the outcome of the general election. Iowa is not a representational cross-section of the population as a whole or even the voting population in general. The fact that this is a fluid and dynamic situation with regards to Trump and his 40%, Iowa is disproportional to the overall election. The democrats need to step up and put forth a candidate that will win. A candidate that is representative of the nation, society in general, and an individual that has both the tenacity and charisma to beat trump. This is a new type of Politics, not the same old game. This is not a normal election. Another four years of Trump will have dire consequences to the United States. The needle is a sad presentation of a non representative citizenry population and if the media continues the type of Hilary chanting and hype this is a lost cause. The needle … like that’s going to help beat trump.
Jon (San Diego)
Everything I read and relearn this time of every fourth year seems to be less about ALL the candidates and the issues, and more about guiding voters to the "right" candidate. The NEEDLE is but one more of the tools used to "inform" voters. We need a shorter Campaign Season (6 months), a national Primary Day (1st Tuesday following the 1st Monday in May - General Election rule language), and finally 6 months later the General Election Day, this year November 3rd. The current election process is a waste of our time, natural resources, and money.
Peter (Colorado)
I’m staying away from the needle(s). I’ll always remember opening the phone in 2016 and seeing Clinton’s advantage evaporate as the mid West states reported and then asking my wife at 11:30 mountain time if she wanted to know what was happening. She knew by my question what had happened. No more needles.
mchristiekroll (putnam valley, ny)
Oh, please. The needle is the work of the devil. Stop now.
LynnBob (Bozeman)
Ranks right up there with NYT's "Hillary will win by whatever" predictions of 2016. Both it and today's Election Needle should have been/should be scuttled.
badubois (New Hampshire)
This needle apparatus worked so well for the NYT on November 8, 2016... no wonder it's back!
Dotconnector (New York)
Even one needle, let alone four, is one too many. It may be a minority viewpoint, but this reader continues to believe that this sort of gimmick is beneath the dignity of The New York Times. If memory serves correctly, what "The Needle" told us as late as Election Night 2016 was that the Clinton probability of victory was 85%. Wishful thinking perhaps, but nevertheless ridiculous.
Ken Nyt (Chicago)
I realize ya gotta do what ya gotta do, NYT. But I gotta do something, too. After 2016 I gotta stay the heck away from news media returns predictions. I just don't have the heart for it any more. And my attention won't move the "needles". So thanks but......
JB (New York NY)
I remember this election needle had a slight malfunction in November 2016. I hope you’ve been able to fix it since then ;-)
AutumnLeaf (Manhattan)
Love the needle, thanks for bringing it back
Elizabeth (Hudson)
Not The Needle!! I just felt my cortisol levels jump. #NototheNeedle
Anonymous (The New World)
I have no hope for our republic. We are now an Oligarchy and the Senate sealed our fate.
AY (California)
Yes, I hit 'refresh' every 10 secs when watching the Obama results in 2008. Perhaps misremembering, but I think that was before the sites stated "results updated in 10 seconds" or similar. But I think we should get rid of these guesses, as I believe several other countries do. Sure, in these polarized times we've all made up our minds. Besides that, however, as books like Carr's The Shallows & Powers' Hamlet's Blackberry have argued, this sort of instant gratification isn't good for us, and probably isn't good for our democracy-republic. So: "The needle gives many readers the piece of information they want more than anything else on election night: It tells them who is on track to win the election. And when the needle doesn’t know, it tells them that, too." And, "For those who wonder whether the world really needs the election needle, we realize the actual results will emerge soon enough. But we also think that the millions of people who follow election night results online ought to have the context to understand them as well as experts do." Now, that puzzles me. *Context*? I think it would be wise to use a little deferred gratification and not make predictions until at least half the votes from PT states are in. Thanks for considering this in the next election. (Can't help adding, "If there is one"....)
Lynn in DC (Here, there, everywhere)
The last "election needle" said Hillary had an 80+% chance of becoming the next president. The needle didn't move until 9pm election night. Why remind the public of that fiasco?
badubois (New Hampshire)
@Lynn in DC Exactly right!
Michael (San Anselmo)
@Lynn in DC An 80% chance of winning means a one in five chance of losing, which is hardly no chance, and not a fiasco.
Ana (New York)
Last time you did the needle you gave Hillary a determined win. Don’t do it. Don’t do it. Don’t do it.
Jennifer L. (Boston)
My two most traumatic election nights: In 2016 I was home alone and I had my own 50-state spreadsheet that I was filling in while watching the election returns. I hoped that once Clinton won a sufficient number of eastern and midwestern states I could just project that California's 55 electoral votes would go for her and rest easy for the remainder of the evening. So I knew there was trouble as soon as some of those key states were in question. I had another traumatic election night experience when I was in my early 20's and had volunteered as a writer and full-time campaign staffperson for a mayoral candidate who was challenging a long-time small city mayor. Although the early returns looked good, one of the more-experienced campaign volunteers explained that they weren't good enough to balance the expected later results from the outlying precincts. Because I had taken a political risk, and had no permanent job to fall back on, several people came to my table during the election night part and told me I should leave town. The band played "New York New York" and "Charlie on the MTA" and eventually I chose Boston.
JulieB (NYC)
@Jennifer L. Election night 2016 was the WORST night of my life. I had not felt despair like that before
SAH (New York)
How much do you have to push “instant gratification!” How infantile in patience are you pushing the electorate to be with this RIDICULOUS needle! When I was a kid “election night” was a big deal! We cooked up a huge amount of popcorn and settled in for hours of suspense until the votes were counted. What fun!! That all ended when they started calling elections 5 minutes after the polls closed. Ruined part of the great American experience in democracy. People DO NOT need to know what your toy needle is saying every second. America hasn’t dummed itself down that much (yet!) I certainly can wait until all the votes are actually counted in a race to hear the results! We all can because it worked fine years ago when we all had to!
Sophia Berg (Tucson)
Why this night, and not any other night?
Dave L. (Boston MA)
Why are there 4 needles tonight? Sounds like a passover hagadah question...
M. (California)
My goodness, there's an awful lot of shooting the messenger going on in the comments here. The 2016 election wasn't traumatic because of the needle, it was traumatic because of the outcome. If I told you someone has a 5/6 chance of winning, but then then that candidate loses, was I wrong? No, that's just how uncertainty and probability work. It's like rolling a die: I can accurately predict a 5/6 chance it won't come up 1 (for example), but about 1/6 of the time it'll come up 1 anyway.
Blair (Los Angeles)
@M. Valiant effort. You might as well try explaining to most Americans why antibiotics don't work on a virus.
Leading Edge Boomer (Ever More Arid and Warmer Southwest)
NYT, you will be reporting on preliminary things before definitive results occur. Many people will see your "needles" on their phones in precincts that are still in flux. You will, inadvertently or otherwise, be affecting the outcomes. I think this is malpractice.
DJY (San Francisco, CA)
Suffering PTSD symptoms here. Where's my service animal to cuddle?
Reader (San Francisco)
@DJY Oh, bless your heart!! I'm totally with you!
Pass the MORE Act: 202-224-3121 (Tex Mex)
No needle forecast means anything when we suffer the loss of our Republic when the Koch network steals the election for Trump AGAIN... if we don’t get all the Democrats that were purged reregistered; http://www.palastinvestigativefund.org/?StopThePurge2020
Is (Albany)
It was accurate a few hours ahead of the 2016 result, preparing me for the "IT'S TRUMP" headline before midnight. Let's be nicer to each other so we can see "YOU'RE FIRED" in November
klm (Atlanta)
It was boring just to read about "the needle".
Eric (Missouri)
The authors of this piece should report how their various probabilities are calculated, and show their mathematical formulae. In the 2016 election when the "stats" were showing 85% and higher for a Clinton win, I wondered whether they were showing their calculated confidence intervals, which could be skewed badly. Please explain how you are arriving at your results.
Statistical Atheist (Greenwich, CT)
Does anyone expect this to be taken seriously after 2016? I need statistician and analysts to do a little public confession before I will ever trust them again.
PT (Melbourne, FL)
Indeed, I share the trauma of many commentators here, about the 2016 debacle. Clinton was showing as having an 85% of winning... till it all went south. You can have the needle, and it may help some of your readers. But we thought an 85% change was pretty solid... it turned out to have been wrong.
Liz (USA)
No! No! Make it go away!
C.S. (NYC)
I suggest the Upshot preview a "Needle for THE NEEDLE" that gauges the likelihood that THE NEEDLE will be accurate and ultimately helpful to readers tonight. Thank you.
George (Europe)
If the needle comes back for the general election it will be the stuff of nightmares and the end of my subscription. My wife and I are still traumatized by the 2016 one.
M. (California)
I clicked hoping to see a bit more information on how the needle handles multi-way races like this. Can you discuss? Shouldn't the needle have to rotate in an 8-dimensional space or something?
Trent (NYC)
Thank you for bringing it back. I am obsessed with it.
Fred (ca)
watch out DNC, a revolutionary democracy is coming your way.
W in the Middle (NY State)
You don't plan to re-use this needle for the NH vote, do you???
Aaron (Orange County, CA)
Better realize now Trump will be re-elected, because we fielded a bunch of losers..
JT FLORIDA (Venice, FL)
Every Democrat tonight should be the new reporting that Joe Manchin is going to introduce a motion to censure Donald Trump. This can put several GOP Senators on the spot who already have said that Trump did it on the withholding of weapons to Ukraine and that actions were inappropriate. This is a brilliant move by Senator Manchin to recover some dignity for the effort by House Managers and for him to take a strong stand by a Democrat from West Virginia where Trump won by 42% in 2016. There is precedence for censure in the Senate: President Andrew Jackson in 1834 and 9 times of Senators.
woofer (Seattle)
Never have so many labored so hard over so little.
PM (Los Angeles)
Satellite Iowa Caucus results are trickling in, are they included in this method?
Mkm (Nyc)
Do us all a favor and point your needle to the winning lottery numbers. Your odds of being correct are about the same as your needle showed in 2016.
Is (Albany)
@Mkm They got it right in 2016, just too late.
Mike (Harrison, New York)
Maybe I'll just sleep through. I'll stick the needles in my eyes tomorrow.
dawn (princeton, nj)
The NY Times, like most media outlets, learned nothing from their 2016 election coverage.
Pat (Somewhere)
@dawn But they did. They learned that nonstop coverage of Trump's every outrage was good for ratings, as was the nonstop horse-race coverage for 2 years. The NYT, like all other commercial media, is in business to make money and this is what sells.
T. Warren (San Francisco, CA)
The Needle of Doom returneth! As if I didn't have to worry about my blood pressure enough already.
Rob (SF)
Just wait the extra few hours.
Owen (Queens)
Put this needle in my arm and shoot me up with that ol' drug we call ""democracy"".
Jim Lynn (Columbus, Ga.)
"A 65 percent chance of winning means losing quite often. Likely to happen does not mean certainly will happen. We don’t quite think of this as “wrong.” (We know that some people do. This year, the needle will also describe probabilities with a simple phrase; it would describe the odds of someone with a 65 percent chance of winning as “likelier than not.”)" Whew! This is what's wrong with this approach. How is this usable to anyone except perhaps campaign managers?
Melissa (USA)
I'll refrain from making the more pertinent comments that many others have already made. From the perspectives of data visualization and human perception, humans are very bad at assessing angles, on which these needle gauges rely. Multiply that by four and you have a data visualization nightmare.
Andrew (Des Moines)
Is one of the reasons you do it "for clicks"? Did we not learn a lesson from over-touting a major prediction (and possibly affecting turnout) in 2016?
Michael (Santa Fe)
Please just don’t. This sounds like some sentient entity lurking behind the curtain. Can’t we have something a bit more nuanced and, errr, intelligent? I’d suggest that if Yang or Steyer (not my choices) came in third at 20%, they would be the winner. I can wait until tomorrow for the bloviating to start.
Heath McCasland (Dallas)
Will the needle wobble between the upper and lower bounds of the margin of error?
D.R.F. (Ithaca, NY)
I've seen the needle and the damage done.
ExPatMX (Ajijic, Jalisco Mexico)
@D.R.F. A little part of it in everyone
Drew (Bay Area)
@D.R.F. Brilliant! (Hello to Ithaca.)
Tim Clark (Los Angeles)
@D.R.F. You should post early more often.
Len Maniace (Jackson Heights, Queens, N.Y.)
We probably need less horse-race coverage of our elections - and The Needle seems to be the latest innovation in this - than a better understanding of why we are so divided and the long-term impact of that. Ironic that while we are closing our borders to foreigners, Americans just may be the most hated people in America.
Bob (Pennsylvania)
The needle is most certainly fine for stimulating the brain's visual cortex, but for little else. Do not use it.
Ben Linn (Iowa City)
I'm excited about the needle. Guys do you understand that the needle is a visual tool, that helps consolidate huge amounts of data from different sources in real-time? And it's nothing more than that - it's not God, it's just a way to view data. So don't be scared of the needle. Don't worship the needle. Just look at the needle, and incorporate it, along with other sources including your own personal experience, to see the world from a more data-centric approach.
Frank (South Orange)
What an absolutely ridiculous and convoluted process. If you work evenings as a police officer, fire fighter, nurse, a restaurant worker, or a store clerk, your voice remains silent. The further we get away from "one person, one vote", the less democratic we are as a nation.
Adam Shore (Brooklyn)
Please please do not do this. Your needle so clearly said that Hillary would will the election, that it was such an inevitability, that people stayed home and did not vote. The nuance you write in this article pales in comparison to a simple graphic. This tells people their vote does not matter.
John (Rochester, NY)
@Adam Shore That's why while I have no malice towards the Upshot, I still remember them saying the chance of Secretary Clinton winning was 89%. Likening it to the chances of making a field goal from a short distance. The kick was blocked.
C Kim (Evanston, IL)
NYT, PLEASE retire the needle for 2020. Your needle’s prediction in 2016 (up until right before the final results were announced) could not have been any less accurate. We can handle reading and listening to actual facts on the ground — no needle.
August West (Midwest)
@C Kim Agreed. With apologies to D.R.F. and Neil Young.. I've seen the needle and the damage done Last time it said Clinton had all but won Then Trump proved the needle was dumb. Electoral college smack, gimme some more With elections you don't know what's in store But the needle keeps telling the score Years back, it woulda gone with Gore Milk silly ideas to keep from running out...
SJG (NY, NY)
A lot of explanation intended to convince us that the needle is good for us while we all know that it isn't. More than anything, the needle is a tool created by the NY Times with the intention of keeping NY Times readers glued to the NY Times for as much time was possible. Sorry, but I already watched your awful TV show in anticipation of learning who you were going to endorse, only to be left feeling like you had just opened Al Capon's vault. There's no way I'm going to spend the evening watching your needle.
Daniel Hagerty (Phoenix, AZ)
This is a terrible idea to go back to using these needles. They mean absolutely nothing. On election night 2016, I sat and watched with the rest of NYT app users as the needle moved progressively and quickly downward from a 90-plus percent likelihood that Clinton would win (it had been that way for weeks leading I to the election). This dubious number crunching you are doing does everyone a disservice. In 2016 it provided false hope in advance of the election - and I personally believed contributed at some level to disincentivizing voter turnout. At the least, it encouraged complacently. Clearly this needle is a tool to use to appeal to the visual, “only digestible information, please” culture we are living in. I just thought the New York Times was better than that.
mike (Massachusetts)
@Daniel Hagerty It isn't NYTimes' fault that some people don't understand math. 90% chance means you will lose 1 in 10 times: far from a certainty.
paperfan (westcentral Ohio)
@Daniel Hagerty Sadly, I think it's been 30 years since the NYTimes was better than this. They're in great company though, as all other "significant" cultural/societal totems have been tumbling from on high for the past couple of decades.
Eliot C (NC)
Still have PTSD from your 2016 needle.
Sandy (CA)
@Eliot C, you are not alone. #CancelTheNeedle
CP (NYC)
Watching the needle on election night 2016 was legitimately one of the most traumatic experiences of my life. Tonight should be interesting, if a bit lower in stakes.
Blair (Los Angeles)
Les jeux sont faits.
Bill M (Montreal, Quebec)
As the prediction season begins, politically anyhow, always a good time to remind people to read Nate Silver’s “The Signal and the Noise”.
John Smithson (California)
Bill M, I never have understood why some people seem so enamored of Nate Silver and his ilk. Political models are abstract simplifications of a complex reality, yet they are presented as being able to separate signal from noise. Problem is, in predicting the future there is no signal hidden in the noise. The future is, inherently, unpredictable in any way that really matters. Models add little if anything to our understanding, yet they are presented as being prophetic tea leaves. Instead of rereading Nate Silver's book The Signal and the Noise (which is not a bad book but not great either), I always remember the old joke about a spherical cow in a vacuum. That helps when I see pundits and prophets bloviate as we wait for real, rather than imagined, election results.
Bill M (Montreal, Quebec)
Fair comment.
Mark H (Boston)
2 endorsements, 4 needles. Can we have 8 nominees?
Bran (Long Island)
Please don’t do this.
Cecelia (CA)
@Bran Agreed. Makes no sense. Read a book. Go hear some live music. Wait until Bloomberg is on the ballet in a month. Then the question of what that will do to the balance of the remaining candidates is worth pondering.
Lawren (San Diego)
@Cecelia, Bloomberg might be able to buy his way onto the ballot but he won't be able to buy enough votes to carry this election.
SR (Bronx, NY)
"but he won't be able to buy enough votes" Here's hoping!
John Smithson (California)
So if you follow the movement of the needles here tonight you may, with an emphasis on "may", know the winner in the four categories a few (or maybe even several!) hours before people who just wait for the official results. I'm not exactly thrilled by that prospect. I'll wait. Tomorrow's soon enough. Or maybe even Wednesday. Better to know than to guess.
mutabilis (Hayward)
The best came at the end when you note: "For this reason and others, the needle is not used to make definitive statements about who has won." Hey, mainline your sharp implements and return to less graphic reportage!
Michael Livingston’s (Cheltenham PA)
This is the needle that pointed to Trump from about 7:30 and on last time correct?
Christos Ioannou (Bexley, OH)
At last, the needle has returned!
Barbara Lutes (Petaluma, CA)
I will never forget watching the election needle during the 2016 results. I could not believe my eyes when it swung to Trump. The horror!!! A very effective visual tool.
Doro252 (New York, NY)
@Barbara Lutes It's still a trigger for you. Don't think I can watch this, even though it's only Iowa.
Bjh (Berkeley)
@Barbara Lutes - it didn’t swing to trump until the actual election. Until then it was wrong and by a very wide margin.
John Smithson (California)
Barbara Lutes, I went to bed thinking Hillary Clinton had been elected president of the United States. Imagine my surprise when I woke up the next morning and my wife told me that was a mistake. It was like waking from a nightmare.
Bill M (Montreal, Quebec)
“Four Needles and a Caucus”. If Hugh Grant’s in, I’ll watch it.
Brandon (Boston, MA)
Pleaseeee noooo not four of them!!! The torture!