well, go evening, everyone.

hi, joe. good eveninggood thanks for bei. and thyo politics and prose for making this possible, as well as he are with us tonight. and welcome to everybody who was tcas well. i'm really, really delighted han about the updated of lost in the gulf with my friend and a colleague joe campbell. and i'm sure that you'll have a lot of questions afterwardwe'reu know, start by listening to joe. questions you and we'll see how this is going.

but please know, we are going to be looking forward to hearing from you as well later on. thank you, filippo. thanks very much. okay. thanks many thanks to the fine staff and professional staff here at politics prose. it's been a real pleasure working them and and their efforts and making tonight possible so thank you thank very much all right let's get started i've got my copy it's a copy as well. i encourage you to do take the opportunity and do that later on as well towards the end you know, this is your seventh solo author book. i hope i'm not messing up the numbers here. that's a lot. and as■f you those who have written a book and even those who haven'tome probably imagine every book has own story. and it's a little by in and of itself about why. what's the know, how that all ce together. so i wusup.

you could tell us a little bit more about the story of e origin of this and why you decided to write and how you went abo great question then. thank you for that. yes, it it is the seventh solo authored book and as i've written and when i this book, i was sort of in between books, the number six book had been■á■ nd i w looking for the next book project9th, 2016. and that was the after the 2016 election at about 3 a.m., i started working on a blog post that, that posteder morninf the 2016 election. donald trump's defeating hillary clinton up in the popular vote, this was a shock that was felt around country and around the world and nobody really had anticipated no pollster, no pundit. the press certainly had not

anticipated hillary clinton defeat. and so so i said to at about 3 a.m. writing a blogt that said essentially that for the news media today lot like 1948. and that was a reference to etrn lican governor of new york, was the hands down favorite win election that year. all the polls, all the pundits, the press all said that dewey was going to defeattruman, the t who became president franklin roosevelt's death■ in 1945. and so the t eio close and at ln shock value. d th's when i began thinking this book, what became this? an w iwas over the course of, i guess three and a half years between that idea that night and the time when the book was published by university of california press that so sort

of relatively fast timeline is or a book which is based on academictalks to a variety of audiences as weec do like about it. it's it's it' detail an interesting sort of historical facts and comparisons, but it's also really, really accese and you know, kudos to you for being up at 3 a.m. after the election. s sn't writing a book i think many of someone's me that she was in tears that that early that morning and and was pleased to know that somebody went to work right at you know they say three hour wow i that's what you do you know you're good i get to do something do something about it so as i said you know the book really of charts the history of what you call pollg surprises

or polling failures. i mean, using both of these roughout the book, it en onward.ologically i36 but i think, you know, for a feg too much of a spoiler this new updated does havelú a chapter on the 2020 election and then it goes on to talk a little bit 's happening this year and 20 for this newn cycle, which i think, you know, presen p journalists, analysts with, again, new difrent maybe presend of landscape. and i just want toe bit of indulge me in this from the final part of the book. so towards the end you say in 2e interplay of polls pollsters a'o

expect surpre. av the recent past is revealing in meaningful ways. but whatever happens, whatever polling controversy arises, it may not be of 2020 or 2016. voters 2024 are well-advised to regard election polls and poll based predictions with skepticismnd y know what you seem to suggest here is that what matter't, jus5;t polling per se, but how we interpret it, how we understand hoit framed. so i was just wondering if, you co role do you see polling playing in this cycle with what' happening around all of us this year? the polls thiar have in all presidential elections, since perhaps 36, even■v before, do set the narrative. they set the narrative for the news media. they set the narrative for pundits and they set the narrative for the public at large. this is hownamics of of a presil

election campaign. who's ahead, who's behind? what's the gap? what's the w facing, so forth? so polls reallyta narrative and they already have they already have this year in wh the polls are showing so far is that this is shaping up theaas tight race between trump, the former president, bidthe this an unusl election in that regard. but the polls consistently since september since, mid-september have shown trump this is the collective polling average by real clear politics on a daily basis and they show they showed trump with a very. narrow lead. so tteus i thi that consistently through the past and a half months that we're looking at a likely looking at a very tight election in 2024. so the polls are setting the narrative as they there is pollg misfire, there's a polling

going to have probably the same sort of polling failure that we had in 2016 when a few states, the midwest, were to be pretty secure. and hillary clinton's camp. buto tnarrowly. i'm talking wisconsin. michigan andns states tip the election, the electoral college to trump andt pundits anticipated and those states that the polls in those states being off and that's what happened that was the key to polling misfire in 2016 and 2020 it was different many the polls thought that joe biden was going to win by a healthy margin. won popular vote by foud a half percentage points. him up by 12. quinnipiac university him up by 11. the wall street journal nbc news poll had are anticipate hitting a double trump. it didn't turn out that was a mu in fact, it was so that the

of 43,000 votes in arizona, georgia and wisconsin have tie. and you can imagine what that would have done to the c covid d urban rioting and just general upheaval to an election that ended in ap tie. that would have been an awful outcome. i think so. i think we can anticipate if there is another polling misfire, it's notsame order of 0 or of 1948, for that any websites, polls, services that you're watching closely that you think you know and is a lot of i must say as well,e challenges, there is a lot of well and how difficult their j really has become over this last few years. so you

think, you know, there might be somebody doing something yeah. i'm glad you mention that because it is a very difficult pursuit. pollingone analyst said. it's li a 15 a scale of 1 to 10 in terms of difficult, the polling difficulty to it right is probably even higher than that. but pollsters don't do this. they don't take the time, energy and spend the money to get it wrong. they're hoping that they're right and to to that end they have been and they have been experimenting with methodologies over the the surprise of 2016. and then a the 2020 election to their their moving around with different kinds of ways of■b tapping into public opinion the goldrmer golf random digit dial calling has b. i mean, it's still used by some

also other different attempts to erch to to tap public opinion in a presidential. now, whether those are going to work and be successful in terms of, you know, getting close to the to the outcome remains to be seen so pollsters wouldn't be going through all this experimentation if. they didn't have a problem tha d one one way that they're trying to do this through what are called panels. they recruans, people whom theyn go back to on a number of to ask, you km complete the polling questionnaire. so whether panels really the future of polling it is a an el certainly an aspect of the future of polling. but to answer your question about what am i looking at one of the sites that i do on a rea. and they have a polling average da daily. and the polling average today show trump washepoints. this is this is a collection, an

averaging a number of of polls are done in the country. and on a regular basis. so 1.6 points is not a vhealth't anythingld sounds and plus, we'e you know, we're election day. so yeah, a lot canappen. glad they're still getting people to pick the phone. i mean that's you know, that's like the first bar a more aftern i'm people, you know, who are here with us tonhtmore questio's happening year and are going to in take on that. but i do want to take a step back a little bit now and think, you know, the other the kind of )uthat the book explores when it relatess

about polling methodology orok'g quite a bit about it and and t a bit in the book. well, albeit in a very excessive way, as i was i was saying earlier. but really, this is a history of the between polling and the citizenry as intended as voters or readers and and, you know, one of the aspects that i think is most enjoyable, the book is that while talk about numbers, you alsoalk people and some those people are colorfulracters. and with, you know, distinct thinking they were, you know, doing their somt of drastic decisions in how they were doing their work, whether would be continuing to do polling after a certain in the e if they thought that one of the candidates was ahead.

and so so i'm just wri you out of all of the different characters, there are many in thentioned talk about whether there are any racters that are some of your favorites, maybe some anecdotes you'd ke to share with us. yeah, that's a great question. thanks. you know, in doing this book, i was really struck by the fact nobody had tried to do this before that a book about polling failure in presidential w, had never really been attempted. and onof t you suggesting, phillipos, that there are a number of colorful character fes who turned up in the book. and some of emy of their own biographies. in he was he was the sort of a founding father of contemporary or modern opinion resech and he has this reputation of being a real and he died in 1984. reputation of being a real

avuncular character, open and warm, you know, and, yound ofte. and it's sort of like the founding father of opinion research. but you look a little closer to george gallup was a real prickly guy and was he was inclinedo atta his critics openly, and. this really had never come been able to research about about theisryele. so i was digging around in the archives at the gallup organization archives athex2 .muniversity iowa. and i got i was this guy was re going the handle of his critics not once or twice. but, you know, man times so he was he was certainly one of the more colorfulanother one is a cf gauss's elmo roper, who failed jeweler in iowa who

became gotuw into market researh in new york and public of that work that he did. and he was he was an election pollster, but he had a lot of reservations about■ polls and he wasn't really a big fan it as gallup was and he was he was a real tepid kind of supporter pollin and one reason is it cost him a lot of money to do electiond butter from commercial and consumer consumer opinion. that's where he was really most at home. but elmo roper is one of those founding fathers as well, a founding figures in polling, ■ñ■úpc another and perhaps the book is a guy named warren metcalf. he was news and was a great innovator in in polling and opinion research. he he developed or helped

developephone calling, which possible polling at accessible way that, you know, people wouldn't have to go do to go to opinion research could reach peopleph and that was a real breakthrough in polling because it it accessible and less expensive. he also warren metascore also wait founder father, if you will, of exit polling, which is when pollsters and their representatives asked preferences as they're the polling place on election was that quite an innovation too. and kotoki in the history of polling, and unlike many pollsters and figures in opinion research who are all always, it inclined to look back from time to time. and he kept on his wall an image of the famousof harry truman holding up the chicago tribune.

that'defeats truman the 1948 election, which is most memorabe photograph ilitical history. andús and and said at the 50th anniversary of that of that election failure, of the polling failure, there's a lot of room, he said, for humility in polling every time, you get coc, yo lose. and sure enough he's he's right. so i think my is aery interesting character but none of them gallup kotowski has biography rather interesting and i'm not going to do it. i'm not going to do the biography. this that would take six years. youguys. i don't think so. yeah. i mean, some of that advice you know, would be true in politics as well. an, that kind of reality check, um, when you were talking =ñou is interesting. so we have caller for collectors

side some other colorful characters though on the journalistic side of things as well. and i think that's the other, you know, quite amusing at times element in the book as well. you talk quite a bit about what you call pahing journalists some of you know some of these columnists ju repr having strong on polling and i just again you know reading a small quo of the figures you talk about in mike royko newspaper columnists in chicago who in one of his columns invited to lie in opinion polls, none of us have ever done that■z right right, because he called pollsters a hired brain picker, trying to figure out what your personal fears hopes and prejudices are so that he can advise a re skillfully lie to you.

that may be radical advice. i'm not going to, you know, vouch for that, but today would be your advice to, based on the research you've done, what wentj d voters when it comes to ■avigate all of this, you know, the poll bashingmo commonplace up until 4 and then it stopped the last real poll bashing journalist was jimmy columnist. he was well-known figure, a pulitzer prize that election in, johnrys orge wush, bush was the. jimmy bresliwe the pollsters almost the same with the same vigor thik chicago thaa moment ago. and he said that this is essentially just lies because they''re not reaching people had cell phones in 2000, four, cell phone use wasg■ growing, but it

wasn't quite the way it had expanded today. so most pollsters were polling people via cell phones in 2004. and jimmeswas just insane was l. and and said so. he also predicted that john was going to win the election by an turn that way in 2004,004, there do't seem to be much in the way of of high profile poll bashing by journalists and one of the reasons that w emergence of data journalism. and that was best represented by nate silver, who founded ■0fivethirtyeight.com and then e 2008 election could predict of the 50 states in thattwn obama and mccain. and that was quite a quite a feat. and nate silver topped that in 2012 when he predicted outcome of all 50 states in the obama mitt romney race.

so that's one reason i think that that this poll bashing phenomenon has disappeared because there's this the rise of datarominent way. silver is the data journalist out there. it's it's i fully expected to find more and poll bashingut t's not been case so i think i don't knl) to your question about how how should people respond to polls. i don't think take them as prophecy. know, they deserve to be checked out and looked at and the and have today. i mean, just dozens in the course course an election season,eople should take a look at them. but treat them as okay, this is how it'sng be. i think there should be certain a certain wearing this should attach to polling because. the recent past tells us that are not going to be't expect th.

and of course, you know, a orgas nowadays or so partner with polling opinion to run their own polls andof builds a certain tyf relationship it certainly does on a certain of coverage as well and that's where another innovation award, the skew whom i mentioned a moment ago, he was s poll operations andw york times in td ■ó association between news organizations and polling. so news organizations moved polling big time starting in the seventies, and it continued to shifted the years■nnd of them do their own own polling. others are you know contracted outssentially but they news their wall streetbig time into journal abcpost, nbc ns,ou knowd

down the they're major media orgazed so yeah it's that relationship is is really really strong these days ys who are vey suspicious about polls and about ■c this. even now, there's some some to do.tionuthe are we making n news. and it's still it's an interesting debate. ttled that we're not news we are you reporting the news in a different nd o way because this is important information to have and in a democratic republic. who's ahead and who's behind? i mean, this is this isaoot jus. i thinkhat's you know, is certainly a kind ofere when trying to make the distinction

making news. and, you know, what's the role of that. and one thing that i think is resting as well as that you've seen over the years this model, you know, that the originated in the united■z stats being exported to other organizations and other news organizations in, you know, havs in european countries between polling institutes, newspapers, for example, that would publish. but in sort of in different systems in, different democratic systems as one thing that's interesting to consider is not everybody welcome, the publicizing of polling right untilction day. and i it's interesting to think about it becauseof takes a longt our relationship with the press has a relationship with polling as and= so you know if we look outside the united states, there are a number of around the world, many of them

european countries that publican polls, either a few days■7 you have extremes like my own native country of italyth polls two weeks out of election da the argument there being this is justeople's choices is shaping people's choices. so, you know what's your view around that, as you know, i think it's it's difficult for us think about elections without polling. but we did have elections without although that would have been maybe a different politics, almost 100 , you know, is there something is there something to be learned from kind of models? mean, they might not necessarily apply directly to the american case, but what's your take on that? yeah, a number of a number of elements there. for one thing, in the united

states, a ban on publication of hlling data would be running up against the first amendment. real quickly. so that would sort of be unconstitutional if some agency ororganizations and others from reporting polling, it just wouldn't work. i mean, it would be a first amendment violation pretty ■óclearly. plus, if you stop polling two weeks out, 's almost like, why bother? i mean, because the that's the window.often,ot alw■'ays, but on when decisions are made, voters maake heir minds as to which way they're going to go, especially if there's a large number of undecided voters, whh we some elections 2016, for example, there was a sizable number of undecided voters. and in the end, in the last couple of weeks orn days, manyy states, wisconsin pennsylvania, michigan, tipped to trump rather than hillary clinton and was a major outcome. rs don't always remember this to continug

as final before the election. now, some of this is early and o very important to understand that the dynamics of the electoratesh late the game late in the late in the campaign and peomgpl minds do get shifted and do do change. so you're ban p>d from published or released in the last two weeks befor like, well, why bother? why bother? i mean, you're missing the key elements here, or at least you risk missing those key elements. and i think, as i said earlier, it's important information as to i mean, elections are about who wins and who loses essentially. at is and who's ahead and who's behind. that's an important information data point for for people in a democrati't see any any reason t all we should consider ignoring banning polls, you know, at any

point in the election, in an tion, there is a misunderstanding. i believe,■t th polls shape. than reality. i mean, you yeah, i decided to jump on the bandwagon again at the last minute because i wanted to be with the winning that be a lot of good persuasive evidence that that that is indeed what happens. and these days in an era of os elections we haven't had a landslide in this country sin id or h.w. bush in 1988. i mean, it's been a time since we've had a landslide election. so this is this is an era of tight elections, and it's also a larized electorate. there are people on either side and. they're not inclined to be° movd one way or the other. so, you know, i think we're going to be lookin a tight election and it'to know who's 'n

if that's the case. and so far, you know, as of late march, trump seems to a very narrow advantage, but he's sustained that advantage mid r ■■q tells something to that's important information to have for an electorate that is on you know i think potential implications and effects of polls i' also about, you know, how difficult it is to go vote and what effect theng information might have on on that decision. and even more so than maybe, you know, am i goi than another one? and then, of course, all of detail that goes into, you know, what national level polling saying versus what swing states polls are saying. d or having a high number of mber of people who haven't

decided they will vote or not. ■dd, know, i think we could sit here and talk about it for much longer. but i do want to hear from everybody ee o's here with us tonight. so. as we've been instructed. we'd love to hear from you and there is a microphone right here behind you. you are sitting over there. ght not be able to see it. it is right behind this column if you want to ask a question and you want to line up and you know where the could you ask the questioners to identify themselves? sure. please tell us your name. yes, my name is for the first time since ross perot. we have a credible party candidate. what do you think are his chances winning? and given how close the race is, if he acts as a spoiler, who will he benefit? great question.

thank you.■9ank u, david. you're talking robert f kennedy jr. of course. yes. i think the chances of winning the election is next to none. i don't think he has much of a chance at all, but he as third party candidate could could tip the election or tip the outcome one way or the other. and i think that as of now, most ofhe reading by pundits and taking more support from biden. he is from trump, of course, could change one thing that is probably going to change is lev. robert f kennedy jr which i think as today wasike 14 percentage points. and in the real clear politics overall polling average. i think that that number likely to decline to shrink■■g'. most third party candidates in the united states o a ross peroo indeed did have a pretty sizable

the 1992 election. d i's still debated today whether he siphoned more support from george clinton. but still, he was example. well, most third party candidates lose sut as election draws near because people just don't want to feel like they've lost their vote and that they'gxomebody who has no f winning. and i think that i think that robert f kennedy probably faces the same dilemma that he's he's not going to win. and his and his support is going to shrehcloser. of course, we can say that six months out ogh know. until after the election. what do you think about the racy of gambling odds compared to polling? it's an interesting question. you. thank you. i haven't seen that that that sort of the gambling marketshe g

failure? i don't think that that'she th, you know, we're going to. oh, okay. we'll see what the gambling markets are saying. an■ud way. it's an intriguing it's intrig oioprobably keeping in md watching. i don't think it's going to be it's not the oracle for for. well, he's the david right. he's not that froquestion. okay. please convince me that polling is worth anything right. you'r's like, why do i want to someone to tell me today the polls you're telling me way way way you're telling me that they do not predict and you're telling me worse than that you're telling me. it doesn't affect people's opinions. so what is it? it's not. and this isn that we are justcommercialization of nea

to get more eyeballs, to make into a play, you know, like in all sports, like all the betting, who's going to win in this tnis? i don't care. i want to watch the match right. game. they turn it into a game. you called it a game and then you drop them. t a game. it's a campaign, you know. yeah, all these to mf you go deeper into telling me why someone is saying they a supporting and then get a little in country with the bad education that we have you going to ask them why are you voting for trump. oh i think you so strongly. it's very strong. is very strong. okay, good. that gives a good indication of what your thoughts are into wh our government should be. what should we, etc. okay, let me let me to unpack some of that. it's, it's important to

underscore, i'm not a pollster. i don'ling i'm not. i know. okay. but pollsters i mean, there's there's multiple parts here. pollsters don't do work with t wrong. they devote a lot of time and scientific analysis they get it wrong often. butintend to do. they at it with the expectation that they're going to get it right. also, it's like, what else do we what else is there out there? if we didn't do polling and media crit jack shafer pointed this out a yor awe suddenly bang and rid of, what would we be left with? we'd be left with ns. well, what do need impressionistic observations like?

okay, how many yard signs are there out there? well, count the yard signs for for one candidate or another or. maybe we should co number of people who show up at a rally and see if that's ■ñ way to to judge support for a■■7 candidat. or maybe we just send people out, journalists into the street andsl journalism, which is which is a fine way to many years, but it's no substitute for signs, nor is counting audience sizes. so polling is is is an attempt to get a representative sample of what's going on in the country at a particular moment in9fime. now, it's often called a snapshot in time. am r impatient with that characterization because given number of polls, we have those snapshots arrayed together are going to give us a something. they tell us important things as to ahead, who's behind and

why and i think that's very important to to understand that. sookay. so fine. if you want to wait until election day choice but i think it's also important for for the electorate to understand that there is a lot that's going on andhat is is ao reach an estim that's not my line but it's a great line and i like that line a lot. so i thiha if we can keep that in mind and that it's not a prophecy tha s about polle them altogether and realize what u know, should journalism isn't the answer. justlism isn't the answer. it's not counting yard signs. it's not counting people counting noses at a assess. you know how this election election is likely to turn out? it's certainly helpful to think about it as an approximate

nation, which it is. i mean, no pollster will come along and tell you, you know, this is l this is 90% of the story and this is hote can . and but it is an indication of where the country is going and that way, you know, not thinking about not waiting, not having to wait and, having that sort■ scientifically informed impression. i mean, again, it'sin to know what up. it's not predictive in the sense of out and who's going to turn out and still do. and pollsters areer impatient with the notion that their work ishat and say, yeah,s prediction,ecially late in the late in the campaign, the last two weeks before an pr territory. starts to get into and i think that pollsters should embrace that. but they tend to be really disincli to that disinclined to do that they want to because feel that this is you know more estimate than■a■+en

prognosis prognostication. thank you for making that point. i mean i think we will want to listen from our our other friend well. i'll take that compliment. thank you i was bring my name is james. i have never not because i've never vote because i'm not american. i just listening in the bookshops. saidethi earlier that some there has been a landslide. sl and when we looked things like when we looked at like 1988 or we looked at like 1980 like 1980, reagan a landslide, obviously, because jimmy carter seen as weak and things like and all that. but when we at 2008 where george bush or republicans were like heavily blamed for the like the economy crashing obama did win by a lot but he never wasn't even by like even his w margins which was slimmer it's like where are your opinions on trump? i think it's argued that he he did make a lot of mistakes in the pandemic.

and obviously that probably clinched electfoairly close by, like 10,000 votes in several swing statesike whdo you think these landslides just don'tm to be possible? is the polarization or is it something else entirely? great que polarization. largely, the country seems■thaty narrow,ms banyway. but of course, you know, pollsters in in 2020 thought that biden's mef them cnn and quinnipiac, among otherllstt biden was headed for a landslide victory. ow■bq tenage ints or more is essentially shorthand for a landslide. and you're right. weavent se that since 1988. bill clinton won eight and a half percentage pthat's close te dimensions. we saw a ten point victory or more was was back in the early late eighties. so i do think it's part of this period that we're in that that

people's views are really solidified on one side or anothe it's always going to be like that, that there's no chance of a i mean, they do. and sometimes carter and ronald reagan, that was ahat was a polling failure in which the all thought that that was going to be very close jimmy carter had a real chance of defeating reagan right up until last. and it turned out that reagan won by a landslide that no pollster sawin and it was it was quite a quite a surprise it could happen. but in this era in the last, what, 30, 40 years, it hasn't happened. and not likely this year. i just have one more thing and then that's it. polarization has happened? like i remember lreading a rick perlstein book, like there would always be like democrats for eisenhower. mo like that. like you don't really see you never you rarely. i suppose oject that's probablya twitter thing. you never really■l.ír

primaries or you don't ever see like democrats or trump or like but like, just don't see this much bipartisan or kind sw right you're right. although there is some effort, i think by theiden camp to to woo nikki haley voters. now whether at's ine very successful or not, i have my doubts. i really don't see that that nikki haley supporters if they're republicans and some there was some crossover primaries and some democrats voted for her in in some 't see republican who supported nikki haley going i really don't not in any great numbers. and not to not to to find you know, republicans for biden there might be some, but it's going to muffled moment, that's a very interesting,ens at really interesting point on this year mean trwhich again does isg the book gets a other consequend

implicns the it's also about wht campaigns do, what they decide to do, where they think there might be an opportunity for them versus, you know, where they don'i's useful to, invest on where they place their money where they placez tin the2 midterm. job of reporting this. the patty murray campaign. senator p in shinon state, look as if she was in some to wi year. and the and were not very looked for a time like she was, you know, in the low double the low single digits,9r know, two or three percentage points over her republican rival. and so there was shift of funds away from, you know, that back tore that she was going to win this and and she not support other or did not extend the funds to otherwhere as she had in the pat

or as it turned out, she won easily by a landslide like six percentage points or so. ese polls were terribly off base, but it does demonstrate just how polling numbers can shift. strategy and tactics similar to 4myou know susan collins of maie and and what up and more recently as well asthink this ys relieved with enough so maryland worry about. but you know that's another story. yes. and i see friend has been up for some time, but let me just say, maryland is a great story because know larry hogan came sort of out of nowhere mean he was a two term governor, but like, oh, wow, he's going to run for senate. my goodness. and that that has really shaken up this race. you bet he has were terrifi nam? caroline poplin and i'm a liberal democrat. alwaysand we're looking for thoe trump, the white house, and now say that would be a catoped thoe people out there for biden among

among republicans, you know, you're there is a you know, working for trump not been a real good career move for many for many people that's that's indeed the case. well, they've been horrifie john bolton is one of them. mark milley. yeah. and you there is a it's a not an insignia accent list, but i don't see folks going out there and vigorously campaigning for joe biden. mae will. i hope so. and i don't know if they have much a constituency to bring along, too. that's that's a great point, actually, just interject. but if of you watched anthony christiane amanpour last night, that's maybe you know, we watched the same program. yes. and you know, he was very vocal about how campaigning against donald trump. and butr following and what voe th g to ■move. i'm not sure. so it's a good point but i

right.did have a ques. i thought you were going to share those. fine. going. but and i was a little bit late, so you mayt this. you talked about data journalism and fivethirtyeight. i at fivethirtyeight, occasionally, seems me it's just a list of weighted polls. is there something else involved? nate silver, who was the founder of fivethirtyeight, has has moved on from from 538. okay. now is that abc news that owns fivethirtyeighso the so the ownership has changed and the management has ced. so it's a defensive team. okay. and it's a different entity than it was from what iwa when. what's data journalism as opposed to polling? it's when you use data and collect data from public to infr your reporting's it's not far 's

it's the way fivethirtyeight. it was polling plus it was more than polling. and so he would he would take all these polls andors model anr elem well. and, you know, run these simulation mtimes, come up withe projections. were t failsafe. but he he succeeded often gh, enougho be something of a rock. and in data. and it does it it doesn't work that way anymore. well, in 2016 he he missed on five states and thought that s going to win those states. and donald trump won them. i think he a■hink he said there was like a 10% chance of a split deci. is, a candidate wins the popular vote. and that's what happened. so that or

possibility came true 2016 andw. you know, as i said earlier, tr was within a few thousand votes. he hasn't won a popular election some. the republicans haven't won since since bush in■á00 popular. 50, 50% or more. thank you. okay, guys have won. thank you. yeah, i see a couple more questions. let's see if we can get them all in before the top joe, one thing you'd mentionfq more would be te impact of artificial intelligence. since so much ofto■ this is now data analysis and through 100 oo simulation wins or 200. you know, a.i. can do that what, artificial intelligence and in the future in 2020. 's a great of fortunately or

unfortunately my■ book research ended before i off in a big time waan don't have a good answer for you on that. it. thank you. thank you. is that a possibility because i do not after the election or tonight. i'm bruce so so i don't support what trumplection integrity but can youvy how pole in which your take on that bee a nsn wi you know at the end hillary die some of the stuff and to investigate what's going on in the rural communities in those three states and with the russians involved taken on that there isn't much in the

book about that it'sthose close. michigan, pennsylvania, enough o to wonder, to raise some questions a a a movement in in congress when meets to certify the there was some those returns. and, you know, joe biden, then the vice president, president, the senate, you know, cuitff d's a done deal. we're not going to we're not going to ger credit he, you know, kept that from from really exploding in a big time way. i don't know. i don't think that there was a lot of persuasive evidence that that that that that would have swung the the outcome in those states. but that is a topic i really didn't get into in the in the book. and if i can just adds in do see anybody looking at that now for for this election the what's

going on in the rural communities polling or the voter integrity or in the rural communities. i don't know. merural communities largely tend to be trump territory and and so and and they've long been republican too in many places in many states. but itection might come down to maybe seven states, seven swing states, including the three that i've mentioned a number of times. well, as georgia and and maybe north carolina and, maybe arizona,a.maybe seven states whs election is going to swing. and that's it kind of rules out the rest of in some respects. 's where the electoral college outcome may may well be determined. an's that's judging from the last two elections, that's where it's where outcome was determined in 2016 and 2020. yeah. elections like we've been having over the last couplereal kind of compound those problems

for polling. ihink of the stories that's maybe sort of unreported. if you want to go down f a 3% to a 1% margin of error in your poll, your costs go up exponentially. so who's going to pay for that? that's also, you know, part of the■2 package and why and why wy spend the money to go

committed michigan a hundred thousand people are uncommitted. so how do you see that playing out? worried. michigan in particular less so about ohio, which seems to have become a pretty red state in the stthree presidential elections. it in what, 2000, four? it was it was these decisiveut y ty red. so i think i think ohio is pretty much in the in the trump camp, pennsylvania is i don't know i don't know if if if the dynamics are quite as as dramatic as they seeto b unfolding in michigan. and the biden camp is very concerned about that. and as and as of now i think trump michigan of course it's and we don't have many polls in these in these

swg states we have far national polling that's going polls. but but as the election draws closer have mond that tends to a problem that we have fewer and fe swing polls that tell us, you know, how these dise states are going to are going to turn out and it can be a real problem f for predictions or forecast or estimates of the thank you. thanks, everybody for your questions. i think, you know,onvene after this election and talk about maybe three days afterwards we can get together and thanks, everybody, for being here. and i know our friends of politics and process thank you. thank you, joe. 2q■■