not particular to our situation. this is some kind of invented -- to justify an ability to understand our data. people do not take into account the elements of the left overs -- this is exactly like in the united states in the first days , people werection shy to admit that they were voting for trump. liberal journalist ideas spread to other categories of the the element of fear has been raised. to which extent it has an impact , not much. we have checked this many times. it depends on the technique of the question. the telephone questions give you loyalist answers regarding putin than a face-to-face interview. context motheran is more of an element of trust. -- in a normal human context, there is more of an element of .rust a lot of things, adjusting the answer to present as loyal -- attitude towards the , because there is civic duty -- everything is asked. are they afraid or not? 60% say they are afraid. 18% say i would rather stay silent. there is a personal attitude which is quite different. , this is the problem and notnterpretation the way to receive data. there's not a great deal of diffe

today, where not so much concerned about the loss of data. after the sony case, an issue of corruption of data or lack of access to our own information. why does it matter to everybody here in the room? more than attack on your infrastructure, these are caps on employees and customers. attacks on your reputation. and attacks on our autonomy and security. tol quickly, i would like talk about the impact of the automotive industry from the fbi perspective most of the folks a bunch of heard panels talking about that this morning. from our perspective, the vulnerabilities include network and autonomous systems. because new cars and infrastructure are increasingly connected to networks, an attack could prevent vehicles from communicating with each other and infrastructure. autonomous vehicles are especially vulnerable. in my previous job as a special agent in charge of the san francisco division, i worked with california highway patrol commissioner who was very interested in these particular issues. a comes and are not from negative way, but he was

product platforms, it may not know exactly what it is that they are going to do with your data. and they may not know to give you notice at that point. and at what point do you build in multiple notification points. i've recently had occasion to talk to a number of founders of new silicon valley startups, and not just silicon valley but europe where i spent time in berlin. today. a is the new value property. people are building companies based purely on the idea that they're collecting lots and lots of data. what they will do with that data, how they will monetize it, how they will pool that resource with other resources, how they will be acquired, integrated into a larger enterprise, big question mark, but undeniably not a deterrent for venture capital flowing into those companies. once again, draws into the question this basic notion of notice and consent. if we come into a world where data is pooled intentionally in a fashion to maximize the utility of personalization, it might not even be reasonable to ask a company to predict in advance all the ways in which that data may be

product platforms, it may not know exactly what it is that they are going to do with your data. and they may not know to give you notice at that point. and at what point do you build in multiple notification points. i've recently had occasion to talk to a number of founders of new silicon valley startups, and not just silicon valley but europe where i spent time in berlin. today. a is the new value property. people are building companies based purely on the idea that they're collecting lots and lots of data. what they will do with that data, how they will monetize it, how they will pool that resource with other resources, how they will be acquired, integrated into a larger enterprise, big question mark, but undeniably not a deterrent for venture capital flowing into those companies. once again, draws into the question this basic notion of notice and consent. if we come into a world where data is pooled intentionally in a fashion to maximize the utility of personalization, it might not even be reasonable to ask a company to predict in advance all the ways in which that data may be

so the creation of data is how he got his start her not taking that data for granted of that complex mathematical connections but that's how one does help them of that productivity data and the crucially from the late 1990's but there was an incident because those modelers because the unemployment was so low and then say idol see this. so if profits we're going up in to make more money and so he wanted to look into that. and that showed apparently mason added can be that. and then they are producing less so they are looking under their hurt that leaks to the post mortem of the 2008 crisis because one of those common obsessions was that was all tied up in his model. and with sap particular model they were willing to question it to stress test the data. and they wanted to choose the fed chairman who was intellectually prepared not to be trapped you may have chosen alan greenspan. so they pulled out a stack of printouts. [laughter] and you talk about inconsistencies sister with the nice anecdote so can you say edward about that in consistent use of the gold standard quick. >> like ayn r

there's not data center technology and cloud technology. it all fits together in a seamless fashion. so by putting tech data together with technology solutions from avnet, we build that end-to-end source of products for the marketplace. we like to say anything from the living room to the data center or from a printer cartridge to a in tech data is going to be capable of delivering all of those solutions to our customers in a seamless fashion, and that really differentiates us in the marketplace. >> the everyday person who watches the show will not know that tech data is involved in the transaction, right? >> yeah, because we work in the background. we're a distributor. we source products from the vendors, companies like hp and cisco and ibm and microsoft and apple. we sell them to outlets that user. so whether it's a retailer or a value-added distributor or a corporate re-marketer, we're the source of those solutions for the end users. the end user can't go to techdata.com and buy a product from us. but the technology they acquire from any

cyber security best practices is not just to provide safety for the customer, but also to keep their data private. it's the same what keeps you safe and those are all encompassing the best practices. not easy to attain, but absolutely covered in the best practices so that we can learn from each or on how to do better. >> i think a private part is unsafe, we might get it right. >> sir, previous speaker said many times she's seen. we'll see more disruption in the next five areas than we've seen in the last 50 years. i've seen more disruchgs that you all have created in the last two years. thank you for your leadership. thank you for a great conversation and being part of it and continue to push the industry forward. i look forward to coming back and seeing how far we've gotten. thank you very much. welcome everyone. we look to continue that kind of informative session and to really make this into a learning experience for everybody. we have an esteemed panel here. just to set the stage on cyber security and as we wrote about say 18 months ago, cyber security really hit our radar when a driv

there are times when some of the desire to not capture data, do not share data, not the intrusion detectionmature and the have hardcover safely got the right balance. it's entirely possible the privacy advocates and cyber safety advocates may delight in this is a architecture choices that we have. i don't think it has to be a fight if we are not really aggressive in pushing through the discomfort historically, we have to talk about what we want as an industry. >> i have a slogan for you. you can't embarrass a corpse. >> not a blessed discounts the importance of privacy. in fact, if you want a piece of congress on monday or tuesday, germany, probably the most privacy conscious country on earth is now putting out mandates for a black box for vehicles. it doesn't have to be identical to airplanes but it's the most privacy sensitive conscious country on earth realize the importance of this. we have had a conversation. i'm willing to start helping that conversation. >> i'll relay the privacy aspect back to the best practices and cybersecurity posture. josh, you talked about the privacy aspects i

this data says that 40% of you are not. the data i showed you on the death rate prematurely from heart disease and cancer says you're not. we're up to about 1.5 million new cases of cancer diagnosed each year. those people just didn't come down with cancer the night before. some of them have had it for weeks. some for months, some for years. if we can detect it earlier, we can be removed if. we are using machine learning. i can show you if you are with machine learning because it getting exciting and i will show you the potential as these databases get bigger and bigger we can make associations within it a novel findings in the genome. we started by wanting to predict things about people including eye color, hair color and faith. we have planned in fact we are better at predict people's eye color then they weren't finding that. we use the toilet imaging to get a precise color and it turns out people's eye color varies by as much as 14% from high july. most people are unaware of that. we can convert eye color pretty well. get

data are not the same as scientific evidence. we get the data but then the human mind interprets that data and then it becomes evidence for your viewpoints, evidence against it or it becomesirrelevant . evidence is really weighted that's been interpreted by the human mind. and explanations develop from acombination of what we already know and the data we collect. these are things you don't do . inferring and concluding, these are things the students looking back on what they did come to realize. this is knowledge about inquiry, it's knowledge about science. and historically, this nature of science and inquiry has always been mixed up and it's always been there but they've always been mixed up in our minds. in project 2061 which was called #4 science literacy, that came about in 1993 nature of science is the overarching theme within which inquiry was included. the national science education standards of 1996 put them together in a different way. inquiry and nature science were two separate things that were intimately related but th

campaign is they had this incredibly sophisticated data-driven machine to turn out the votes in very specific toys and places, but it turns out that if you do nota message and a leader people really believe in all the sophisticated data will not get you the results you want. emily: do you think some of the voters were on reddit? steve: i am sure they were on reddit and clearly there is methodology with the polls that were done if you spend any time on reddit in the last year, last night's results were not surprising. we have a lot of users that support both candidates, but the users per donald trump on reddit were much more engaged than other users. on average we saw donald trump morerters vote six times often on reddit then supporters of other candidates so that engagement i'm sure translated directly to turn out and the results last night. >> if a product is broken marketing cannot fix it. have they make sure this doesn't happen again? was shocked. >> almost every election you see this with the polls. in the reasons you give way people are engaging which each other in society they are much more fluid in their views and i wonder if the whole t

the "alabama school bus seat belt pilot project," researchers determined that safety belts were not cost data showed three times the number of children were injured or killed in the loading and unloading zone, than on the bus. and if these kinds of costs were to be incurred, concluded "more lives could be saved by investing in enhanced safety measures in loading/unloading zones."in fact, in data compiled since 1991 by the national academies, nhtsa and the national safety council - since 1991 by the national safety council, an average of make bus safety consistent with standards in personal vehicles. sot:hersman: "getting kids to buckle up every time on the bus. it has a lot of other benefits. keeping them seated, behavior, and everything else but it's really about safety. everyone grew up with seat belts and the standards on school buses date to the 1970's. times have changed and we need to update the standards for buses."as of now, it is a stat decision. six states: california, florida, louisiana, new jersey, new york and texas all require safety belts on school buses...and all buses under 1

julie: that are not data people. they're compiling polls. ghan: bias of entire election mitt romney, my mother, my father, all republicans everywhere. this is not just donald trump specifically. don't you think they need to look at themselves. julie: they did. arthur sulzberger junior, wrote a letter to the subscribers, asked the question did donald trump's sheer unconventionality lead us to other news outlets underestimate support among american voters? we reported on candidates fairly but need to rededicate ourself to fundamental mission of "times" journalism. which means we didn't attack him enough. we need to go after him enough. julie: to answer meghan's question, we know a lot of "new york times" reporters personally, i consider them to be friend i said this before many times. we sit in a new york city bubble. meghan: i do not. julie: you may not. >> i'm in washington bubble. julie: we all talk to like-minded people, we liberals, including people that work at "new york times" we all talk to each other. sit in a little bubble i said this

right now, if you look at all this data, there is not really a path for donald trump. again, he is within the margin of error in a lot of places. he is within striking distance in a lot of places. he is behind narrowing the gap where he would want to be six days out. mark: even as clinton plays defense in certain states by going herself and having surrogates visit, she still can look at the mouth and say, a mess -- unless there is a huge shock, she can win enough blue states to get to 270. barring a national shock and many states ending at different than the public data, she had got it. look, the way you have to look at this on some level is six days out, whose hand would you rather be playing? her hand or his hand? no doubt, you would rather be playing her hand. mark: i am here in waterloo. paul ryan is doing an event, not for donald trump but for the local congressman here. iowa, like a lot of battleground states, it is hard to move a few inches around without seeing a major player. in iowa, and elsewhere, is the answer to one question. is trump cutting into clinton'

federal >> but what i call it is it is basic is new york that is focused to create that aggregate data which had not existed . so that creation of data is the best quality data but not modeling are taking that for granted with those logical connections >> afterwards, airs on book tv saturday and sunday at 9 p.m. eastern, you can watch all previous afterward programs on our website, book tv.org. am and book tv is live at th >>host: we are live that the miami book fair for the 33rd year the festival has then held at miami dade college north of downtown miami we're now joined on the set whisky author of the book they cannot control assault . -- us all that is a long title. >> it comes from a video after the fatal shooting of martin one of the first fatal shootings after the grand jury decision not to charge the officers in ferguson . so we stuck with that because we capture the feeling of those who hadan taken the streets last few years is seems like it is happening every day that we will all rushed into the streets because there is power in the numbers. >> one of of demonstrators that night there was

specific ways and specific places but it turns out if you do not have a message and a leader that people really believe an, all of the sophisticated dataget you the result that you want. emily: do you think that some of those voters were on reddit? >> i am sure. while clearly there is a methodology issue with the polls done, if you spend any time on reddit in the last year, last night's results were not surprising. we have a lot of users that support both candidates but the users for donald trump with the donald trump supporters were much more engaged than other users. we saw on average donald trump supporters voting six more times often been supporters of other candidates. that engagement translated directly to turn out in the result less night. >> if the product is broken, marketing cannot fix it. the product was not as good for the democrats. --ly: how did he make their how do they make sure this does not happen again? >> i honestly do not know. emily: exactly. >> almost every election you see this. there are this way, wrong threasons you give, th people engaging with each other, much more fluid, i wonder if the whole thing is actua

data. the second is taking other sets not just from one company but several. a typical insurer will collect motor vehicle records from the dmv. incident data. so crash data. or look at data over time so the person who has more time with one policy is actually much less risky than someone who keeps changing policies. and then using new sources of data. so what's happening in this industry is telematics, or user driver data behavior, real-time data installed in the car of your choice, the rate at which you accelerate, what time of day do you drive. what tame time of day to you drive. all of these data points helped solve a problem of pricing risk. and back to the societal mission. issuing fair pricing to consumers. so let's talk about another industry before health care. u.s. law enforcement. the mission of the 18,000 law enforcement jurisdictions in conjunction with federal agencies like the fbi is protect u.s. citizens, find criminals, find missing people and solve crimes. fifteen years ago, twenty years ago, they would rely on feet on the street bus static dat

not because i have that hard data the only when i got my hands on any real data i talk to one teacher 96700 did 2011. and then there is us secret system. so they go back to washington d.c. one teacher got fired because it was a little too low. but the other what were 50% where the information was terrible value added score. so what more thing of lot of fourth graders had gotten really good scores at the end of fourth grade. but they could not read. they could not write. so she was suspicious of their goods course and in fact, has every reason to believe that the teachers cheated to get better value added model scores because their kids were better than expected but they set her up to get worse than expected. so there is all sorts of erasures atom of your kravis scandal but it never systematically got investigated but in any case she was fired. so it is widespread and they're using more than half of the state's or the school districts she did not understand her scores were but when she appealed but then it is destructive so not only did did destroy her life and she was rehire the follo

project i worked on was tracking police killings and how many people were killed by the police, that data is not kept. donald trump said we don't believe that the federal government should be reporting this data and don't want to harass them. he stated many times he wants to restore more law and order and said in a questionnaire to the international association of chiefs of police that he wants to rebuild the relationship between the federal government and local police departments, which sounds to most observers essentially he wants to be less critical of local police than perhaps the obama administration has been and so, what we know about the potential new attorney general, jeff sessions, or senator jeff sessions is that he also has had certainly a law and order streak. so, it seems very unlikely that many of the steps taken during the obama years to an empty to reform local policing will be continued. and i think what that most likely means is that for people who want to see changing in local police, there's going to be a need, certainly, to work at a local level because i don't think the fed

we also know that the federal government does not make data based decisions very often about the programs if funds. we should apply data to that. we also have to deal with our budget, we have to strengthen social security and medicare. this is a basic difference between me and the current majority in washington. i would not do that on the backs of our seniors. if you pay in, you have paid into it and earned benefits. we could strengthen medicare by allowing negotiations for prescription drug prices. we could strengthen social security by asking millionaires and billionaires to pay a little bit more into the system. lastly, we have to keep our economy growing. that is why critical investments that help us develop a 21st century workforce, a clean energy economy, a transportation infrastructure, and incentivize lower taxes will help us do that. the best way to deal with this is to make sure we are continuing to grow and leading economically. [applause] >> thank you for that. we have about one minute left. would you like to make a closing statement? governor hassan: thank you all for having

it's not in the it's not in the data. it's not in the number of offices any campaign has going. >> reporter: that state representative said he's witnessed something this election no one has ever seen before, scott. when the trump campaign and state party ran out of trump yard signs, supporters in this part of pennsylvania, went to local prisoners, designed their own, bought them, and then distributed them to friends, family, and fellow trump supporters. >> pelley: major garrett, thanks. major, nancy, and the rest of our campaign 2016 team will be our live coverage begins at 7:00 eastern time. the final jobs report before the election shows the economy created a healthy 161,000 jobs last month. the unemployment rate fell a tenth of a point to 4.9%. preelection jitters, however, sent stock prices lower today. the s&p 500 fell for the ninth straight session. that is the longest run since the the election is big news all over the world, especially in russia. in moscow, elizabeth palmer found who russia's president is supportin

roe reporters cannot and should not rely on antidotes to judge this race. the dataays they're not going to turn out. >> the data is really important because you need to have the data to go on along with the antidotes. >> with with the national polls had hillary clinton up to about three or four points. probably once all the full votes in from california, she will have a 1% national lead. the national polls were not all that wrong as we saw in places of florida and north carolina. donald trump won by a narrow margin in both states. what was really off was the polling in pennsylvania and wisconsin and minnesota, all of it ended up being so competitive than anyone ever thought when we did our maps. well, it is going to be hard for him because pennsylvania, hillary clinton has been ahead and wisconsin, the big supply was the blue firewall did not hold and the data did not show that. >> the last couple of months when trump started talking about a rigged system. something we hear from a lot of people that we do not buy the poll and we do not believe the polls. i want to p

not. could trump force companies to hand over more personal data?that is the interesting question. what it appears is that he is going to be much more prone law-enforcement, so he could be a lot more aggressive going forward. we need to remember there are still other branches of government that need to be engaged to enable him to be more aggressive. he would have to get a subpoena approved to gain access. i do think this is where an idea where senator warner and congressman mccaul's commission is to study encryption further. i could see that gaining some length, but i think of all the priorities he has coming up, going after and changing the way encryption is in this country, i would say that is not going to be at the top of the list. emily: what about other countries? the u.s. has an agreement with europe called the privacy shield which sort controls the transfer of data that can happen between europe and the united states. could trump undo that deal? >> of course he could. it would be an executive branch activity. but the reality is that is unlike

-watch out for "phishing" emails that n on a smart phone, use your data-- d not free w today, esident obama will spe a turkey's life... actual two turks.its part of an annual white house tradition.one turkey .. either 'tater' 'tot' will become en both will be spared from the dinner table. see you in 25 minutes for your >>> good morning, ambassador haley? nald trump pks h first female cabinet meer, ambassador this as trump acknowledges he won't be following throu on some of th proms h made during the key to the caaign. prises likehis one. >> i am going to instrt my attorn generalo get special procur to look into your tu >>> disturbing past, authorities are learni more about t-year-ol

no decision must be based on scenarios about the future that are not data dependent. they cannot be data dependent either very nature. -- by their very nature. if you act, you can expect there to be people that deplore the action you have taken. will not receive gratitude from most -- you will not receive gratitude from most of the people who use it from a worst scenario that might befall them. the fourth and final scenario that i learned that lesson i learned from writing this book -- the fourth and final lesson that i learned from writing this book was from one of kissinger's essays that was not published. simply vanished into the boxes that i was privileged to open. this is another of those exciting moments that historians do not generally get enough of. here was an unpublished kissinger book about a central figure he has most commonly identified. how many times have you read in essays that he was the bismarck of the united states? well, i have the evidence. this was to be the second volume of a trilogy. the first volume was published. work and maybe the best thing

charles: border patrol agents say, the obama administration is not sharing data with us, not fully truthful about what shaping in our country. those agents is a that flood of illegal immigrants across southern border with mexico is now quote, a crisis, situation. a new report, from breitbart said that immigration and custom enforcement has no room to house the arriving in the sunk. so many illegal immigrants are being release into communities, there no room to hold them, in the communities they are free to disappear, one said that public safety is in jeopardy, main street is complicit with obama administration, in covering up the crisis because it is an election year. charles: i find that difficult to believe, we're coming right back, stay with us. >> which clinton cartel, playing the victim after the fbi and wikileaks expose more corruption. >> hillary is not the victim, the american people are the victims. >> congressman ron desantos from house oversight committee will join us next. >> that cartel is scrambling to survive, trunk ahead in the poll, hillary struggle with her end lets lies e

data are not the same as scientific evidence. we get the data, but the human mind interprets that data and then it becomes evidence for your viewpoint, evidence evidence against it or it becomes irrelevant. evidence is really data that's been interpreted by the human mind. explorations developed, these are things you don't do like observing and inferring and concluding. these are things that students, looking back on what they did, they come to realize, this is knowledge about inquiry and knowledge about science. historically, these have always been mixed up and they've always been there but they've always been mixed up. in the benchmarks program that came about in 1993, nature science was an overarching theme within which inquiry was included. the national science education standards from 1996 put them together in different ways. they were two separate things that were related but they were different. the and gss now has a subset of inquiries and that's what i call the little red circles where they were aiming the increase sectio

not surprised. i am shocked because clearly the polling the data, not just the public polling, but also internal polling in our right? i think that's something we need a address and what happened there. i think other thing that really kind of struck us, sticks out to me is that i worked on the obama 28 campaign and the white house for obama. i realize is obama is once in a generation kind of in a lifetime politician. and he was able to create this coalition and now it no longer, clearly, no longer exists. sticks out to me tonight. >> so when you look also the republicans are riding a wave of tremendous success votes in the house and senate. they will also win the white house, was this an anti-hillary vote on the presidential side? how do you explain the success across the legislative branch, did we under understood the republicans? >> it's an extraordinary achievement. it blows every convention as y pollsters have just got to be scratching their heads because his models of polling are off. there was a brexit-like vote that they have been saying that turned out. and many, many people, including myself, w

not enough data for washington, expected to go to clinton. also, oregon, we'll wait for these. but we're paying so much attention to the battleground state. we're down to nine fro what we've seen, many many electoral paths opening up for donald trump to get to the 270 he needs. jon, let's start with the state of florida. >> looks like he running the table on the remaining states. florida, donald trump has a lead of well over 100,000 votes. just not enough votes out there, i believe, for hillary clinton trending solidly for donald trump. north carolina, another state that was a toss-up, a five-point lead with 92% of the vote in. trailing in the state of pennsylvania, with about 70% in. but look how close pennsylvania is. this is a state that hasn't gone republican since 1988. keep going up north. new hampshire, 59%. a state where trump has been the underdog for months, george. and he's 49% to clinton. and the real story here may be in the industrial midwest. the states of michigan, where he still has a lead with about half the vote in. and the state of wisconsin, where he's the

not all, but many. second, as we collect more and more data and companies become diversified across platforms, it may not know exactly what it is they are going to do with your data. and they may not know to give you notice at that point. at what point do you build in multiple points? i recently had occasion to talk to a number of founders of silicon valley start ups, not just in silicon valley, but europe where i spent the last several years in berlin. data is the new value property. people are building companies based on the idea they are collecting lots and lots of data. what they will do it with, how they will monotize it and pull that resource with other resources, acquired and integrated into another larger enterprise, big question mark. undeniably not a deterrent for venture capitalists. it draws into question the basic notion of notice and consent. if we come into a world where data is pooled intentionally to maximize the utility of personalization, it might not be reasonable to ask a company to predict in advance all the ways that data may be used. they may not be the only ones that gain access

not right now. she is talking about other things i agree about that. and also i think the datas very sparse and scarce. we will know more in the next 48 hours. but if this surveymonkey/nbc news poll is indicative, the dynamic has not been changed at all. mark: yeah. the republicans claim both kellyanne conway and other republicans claim trump was already moving ahead in battleground states before this happened. just as donald trump wanted to be dealing with a few weeks ago, the accusations of his treatment of women. the last thing in the world hillary clinton would want to deal with is a combination about wikileaks, stories about investigations of the foundation that were squashed allegedly, and any questions related to e-mails. it is testament to the clinton campaign and her standing that this has not caused the bottom to drop out, but it is also a testament to the fact that we are a divided country. as i said, a lot of the impact has been to give hillary clinton a rallying cry. john: here are two things to draw the parallel. the difference between the access hollywood tape in

there.e data is still not all of it. some of the premium data we do not put behind a pay wall. on the functionality side, crunchbase.com come other free stuff you had and how you use crunchbase, all of that is exactly the same. the original starting point, what is crunchbase pro -- what it can't be is taking crunchbase and putting pay walls in front of it. add features on top of crunchbase and charge for those. by the way, we also give some of those features away for free. anyhe left side, click of them. how many companies have female founders in crunchbase? you can do that search. i'm super lazy, but one of the things you are describing is a functionality that seems to be a little awkward. iremember at another company used to work at previously, i had access to a database. those queries were a little smoother. is there a way to refine that process? is that something that you all want to do? we are going to listen to what our users say. we are in a world where we can iterate on weekly releases now. what has crunchbase done lately? it looks like nothing. canthat it is out there,

in terms of things being different using census data from 2000-2010 the bubble and the the bust kind of fall not entirely within that interval. but the fact that i was using data ten years apart mean that the short-term dynamics were missing from my data. but nonetheless you could definitely see that a lot of the communities that were heavily impacted by the housing bust like the inland areas of california, south florida, these are areas that had a large immigrant population. so a lot of the home owners who were caught up in that were foreign born. >> i will comment on the wealth. absolutely wealth immigrant households was hit more because especially hispanic households had more of the wealth in housing. but i would like to take your question one step farther because we're doing these metrixes studies that rely on data we're not looking at what's happening right now although price has implications. what i think is interesting is that housing prices and renlt are increasing relative to the cpi and wages and income at a very rapid rate, faster than they ever had and more consistently. so in terms

not surprised. i am shocked because clearly the polling the data, not just the public polling, but also internal polling in our must have been terribly wrong. right? i think that's something we need a address and what happened there. i think other thing that really kind of struck us, sticks out to me is that i worked on the obama 28 campaign and the white house for obama. i realize is obama is once in a generation kind of in a lifetime politician. and he was able to create this coalition and now it no longer, sticks out to me tonight. >> so when you look also the republicans are riding a wave of tremendous success votes in the house and senate. they will also win the white house, was this an anti-hillary vote on the presidential side? how do you explain the success across the legislative branch, did we under understood the republicans? >> it's an extraordinary achievement. it blows every c pollsters have just got to be scratching their heads because his models of polling are off. there was a brexit-like vote that they have been saying that turned out. and many, many people, including myself, were wrong