Model Talk: Democrats are slightly favored to win the Senate | FiveThirtyEight

In this edition of "Model Talk," Nate Silver and Galen Druke introduce our 2020 Senate Forecast model.
44:06 | 09/18/20

Coming up in the next {{countdown}} {{countdownlbl}}

Coming up next:



Skip to this video now

Now Playing:


Related Extras
Related Videos
Video Transcript
Transcript for Model Talk: Democrats are slightly favored to win the Senate | FiveThirtyEight
Hello and welcome to defy identity politics podcast I'm Gayle injury. I think Nate Silver. And miss. It is my hopes Iraq. We are record payments on Thursday but if you're hearing miss it means that it's at least Friday and that we have officially launched arts and it forecast model so Anchorage ever wanted to check it out on fire pretty dot com. Prosecutor and the top line numbers art. Republicans have a 42%. Chance of maintaining control of the senate and Democrats have a 58%. Chance of winning control of the chamber. That's according to our deluxe model which will get into. This single likely outcome according to the forecast is 5050 split in the chamber and of course the vice president breaks the tie in this. Nathaniel silver. Among microphone her in my computer I do you know I'm having connected to Iraq it it. It's been recording further upwards of two minutes now. RA Mary I'll keep an I am without incidents yet. You can't you can't fill me twice like that's got to be the only time this so what what what happens when it record it. So Tony records a backup. Skype audio file which is pretty poor quality. And then also requires us to go through. And clawed out everyone else's voice from your audio in the house bar so it's more work and the audio quality Sox by. We didn't have to retake the entire park in case anyone was wondering from what happened a month. Well that's yet. Now that we have cleared that neat how would you describe the steed of the race. For this Saturday since we have you've completed the forecast you can see the output. What's this data varies. The serious is that Democrats are in. A better positioned than. Then they probably thought they would been in a year from now or a year ago I should say. There are an awful lot of competitive seats. But by no means is it a done deal. There are. Basically know whose seats we would guarantee democratic pick up there is one seat in particular in Alabama where they're likely added to lose grounder news Doug Jones probably will not be reelected running against a non Roy Moore candidate. Senate's Democrats ready to pick up. Four or five seats to and if they win the presidency which is kind of it. A pretty tall order. There's also a kind a contrast between. The polls in other indicators of the race comes to look at her forecast there are actually is a return from 262018. They're actually three different versions that are called light classic in deluxe. And they layer makes me hungry we're taping this just before lunch time. Ease of burger. And they layer different versions of tuchman other different information right where his light. As much as possible is based on polls classic ads and fundamentals factors like. Like. Fundraising for example. And then deluxe also adds an expert projections so. Based on the pulls alone. Democrats have. A 69%. Chance of pretty nice chance of winning. The senate and the white version in classic like a seven to 64 and then deluxe accustomed to 58 so part of what's happening here is that. I mean Democrats pulls money senate races look pretty good right there was a poll earlier this week ahead. Lindsey Graham tying in South Carolina with Jamie Harrison. And that's too simple for Democrats right. It is the model essentially saying that yes that's a great pull for Democrats but based on other things like the partisanship of misty. Horror fund raiser or whatever else that we don't really trust that when different arts and we. There's a chance right but like the yeah I mean. I think they deluxe forecast hasn't reflected a 50% chance of losing which is not trivial it's highly a lot of people thought. But you know Democrats are competing against a lot of incumbents. In a lot of red states. Been seats at triple probably win. You know and so it's not an easy lift necessarily even though incumbents aren't nearly as an advantage as they once were. I mean look you have some. Pretty easy pickups for Democrats right and I Colorado errors are Colorado. Arizona probably the two easiest you know mark Kelly's been ahead in almost every poll although. Mix Alley has tightened the margins a little bit in some polls. Colorado's they've eerie purple state now. Mean. Is. And that relatively easy pick up a mean Susan Collins is kind of a tradition there it. She's a popular for a long time her popular is gone way down but. And coming to win by whatever it was 25 or thirty points like she won last time I think should not be written off right bits of stilts a state where. You know Gideon her polling has been strong Biden's drawings and Paula strong in Maine as well I'm certain 81 question North Carolina yet. Is that in 2018. We saw that the partisanship of the state whether or not the state voted for trop in the sixteenth election. Essentially over rode any thoughts an incumbent advantage we saw that in a lot of states right ultimately even if there as a democratic incumbent Mike in Jakarta or Indiana. Deep Republican challenger ultimately prevailed. When we look at states lights. North Carolina maybe they're more purple Iowa or Montana some of the states that Democrats could be relying on an order to win the senate. Those are fundamentally red states is there a reason that we should think things will go differently and 20/20 from the waiter they didn't when he eighteen in terms it's the partisanship really overriding everything. No I mean it probably will read it I mean there is some degree of ticket splitting but if we look at the senate map right thing okay. Democrats are favored. To win the contests. Being Colorado. Mean when the presidential race in Colorado Maine. Arizona let's places where they're. At least Biden's at least 60% to win right. Biden has almost no chance of winning Alabama so. Take. Take those three right Democrats are in net to now North Carolina. Biden is seen here but that's much closer for example but he is favor there's that would make it net three which is just be. Barely enough right but the mayor's two races in Georgia which is we have trump a little bit ahead. But very close there's a recent I we have trip a little bit ahead. But very close Texas trump a little bit ahead but very close right so you know this kind of situation here where. A really narrow Biden win let's say that Biden. Wins Pennsylvania. Probably there aren't a lot of senate races in the midwest this year Democrats are playing defense of them anyway or let's say Biden. Has a good but not great night and that consists of winning Pennsylvania. Michigan. Was currents and an Arizona. All states that that Clinton did not wind between sixteen that's. Comfortable if she Biden march and it might translate to like four or five point win in the popular vote for Biden right. But let's say he loses Florida North Carolina. And Iowa and Texas and Georgia. And every senate race files presidencies are hypocrites actually don't in the senate there they're fifty when a 49 their. One seat short. And so you know. This. Senate is a heavier lift for Democrats than the presidency. It's not Couri easy to think. That the races to haven a quirky wit a eons somehow trump wins and Democrats take this and it's not crazy crazy. But it's it's that's less likely than the other way around. Where does it look like. Racers are most likely to diverged in terms of the presidency at this sad I think for example looking at Texas right. Biden has a much better chance than the democratic challenger answers acre there. Again in mean rarity discussed by Susan Collins has a better chance of winning means. Then trump does. Yet are those the two main ticket splitting contest we see made in Montana as well are there other ones. Yeah I mean. You know Democrats would say hey look 19 and I am we have the current governor there he's very popular he's a moderate. We are other senators and Democrat Jon Tester and he won reelection last times so. You know I might not be crazy. In North Carolina Democrat cal Cunningham has generally. Pulled a bit more strongly than Biden and had clear leads the Republican committee Thai populace is as a fairly low approval rating is not raise a lot of money he barely one last time. Which going to be like a classic. Kind of weak incumbent. We get to more exotic things I think that. Lindsey Graham. I actually would probably be a buyer of or 15%. For juniors and there. Interest because I think Lindsey Graham liked. Are kind of miles of us is like a fairly strong incumbent I think like the kind of being seen as kind of a moderate squish and then being relieved Monica. You say that whichever toughly and never shirts nagger makah are hash tag. And he advocates and it regional acts. As a midwest and I think you can. Magnet. Shouldn't Lindsey Graham kind of it's made anybody in that state. Happy right but there yet we look they're. There are all these races that are Corky. And sometimes quirks to happen right Democrats they varied mid term and when he eighteen however. Bill Nelson lost in Florida. Still a very consequential loss he's an incumbent in an environment. That was Democrat plus Peter plus nine nationally and yes Florida is. Bottom been frustrating for Democrats but still that was kind of it. A surprise. Via tester and Joseph Manchin held on inclement castle in and elderly people did not write and so it's it's. You know it's. Senate races can march their own drummers. A little bit is also in Georgia there's two elections there including. A special election. With Kelly loft where among other Republicans running in several Democrats running. So there could be quirks on the map and their places where. You know in Michigan are Hooper of the dedicated John James who arm was pretty vigorous in 2018 he's running again against Gary Peters who is not. The most spectacular incumbents of partner harder quirks and either side here. You mentioned at the national environment in twenty teen was Democrat plus eater nine how do we expect. National environment to compare -- This fall and of course it's worked parting out listeners you. Maybe haven't paid. Kobe is amount of attention to the senate map that the map just looks better for Democrats this year. Bear more on the attack on on the defense as they were in twenty team. So it's not just a case. Environment nattering it's also which seats are up. But but in general in terms of dean partnered what do we expect this. Obvious the Democrats are ahead by about six points are in. The generic congressional ballot so to scan and asked which party preferred. To have control of congress or who vote for in your district that six point lead compares to an eight or nine point lead the Democrats finished with in when he. A team so Lieberman is not quite as good for them as it was. Look sell part of it it's to echo Europe 2014 when Republicans won. Everything that you know had a move that move basically. In the senate. But still remember the senate has a pretty strong. Built in. Republicans a dozen Republican bias right but when you have like a built in design where I only has as many sinners is California. That tendency to overweight. Rural states relatives their share of the population. And rural seats and to be more Republican right. So they're all these targets Democrats have been George and Texas and Arizona. You know South Carolina right. With the exception maybe Arizona history becoming more purple state. These are already leaning states. The Everett leading states in it. Environment we're a year ahead six of the generic ballot and by and had seven right. Where you can make them very competitive. But they're not easy places to win for Democrats. When it comes to you. Liberal or progressive Democrats who think about this fall as their opportunity to pass a bunch more progressive liberal legislature. Should they be looking more at the Democrats aren't in the senate almost as an indicator of whether or not they can pass that agenda then necessarily Joseph Biden's Dotson Joseph Biden odds of winning it our 75%. But our deluxe version shows that Democrats have a bar 5758%. Chance of winning the senate. Is that the actual chants are those of the actual odds that the progressive Democrats will be able to implement their agenda. No look. If your progressive Democrats in you have. Two or three problems right when it which is Joseph mansion. Maybe cures cinema in Arizona is often pretty maverick key in the Arizona and tradition. You know if Doug Jones somehow wins reelection I think you'll have to decide whether. Whether he's hoping to win reelection a ten and six shooters like screw it comes to the vote for the progressive agenda but like I think. You know that's less than 37% warriors make it actually after win. BA. As a sizable majority every member emeritus at the Myers you know I think. You know with 52. We can do more than with fifty with 54 her. You can do a lot more than with 52 right let's get to 54. Or so then you have a comfortable margin Weyers. Joseph mansion and cures is an American do their thing right. And you could have one person object for weird reasons and someone else gets coated and can't vote right and you still have. Fifty and vice president Harris breaks attack right and is also assuming that they blow up a filibuster but just to give us a sense wanted to chat says that Democrats get 54 seats or more. So in deluxe a 15% chance of having 54 or more fifty tour morgue it's up to like. You know 33%. Chance or something right so if you if you can somehow. It's a 52 in keeping or when except for mansion in cinema in line. The need to be some things. So again the odds of that happening are more or less similar to the odds of trump winning in 2000 sixteenths or. If you believe that things are at a 30% chance of happening. Can happen. Then there you have it but I wanna did answer a little more minority aspect of this for sector. How much did this out of forecast model changed since 2018. So are not really very much. We. To a couple of things though. And these are going to be fairly technical. Line is that the kind of fundamentals that we talked about before. We. Re estimated those. Equations to account for changes as a result of partisanship. You know basically now a State's partisan mean or district person knowingly release the house muddle matters even more. Factors that we cult and equality like you know experience and whether you had a scandal organize and fund raising. Those matter less than they once did as result of higher polarization and partisanship let's. One change doesn't have huge effects but some effects. Now the change is we. We are estimating house affects a bit differently we're here. Mostly we look at how a poll. Compares to others of the scene state or district. And there's not as much crossover permitted. Between districts. Because there are some pollsters that might. Have a different view of the electorate right near very friendly to Democrats and some states and that's a friendly and others. And just empirical York's better if you kind of keep those house affects mostly contained within the state. Then there are a few changes. As a result of code kind of near change in the presidential model. We assume there's a bit more uncertainty on Election Day with respected. Turnout and the voting margin. So it's to weed out. 20% more uncertainty in the default version of the model. But there aren't as many kind of complicated. Cherish sticks around cove and as we use for the presidential model also those of the main things none of them are. Hugely consequential. It does make them model teensy tiny and it more conservative so maybe you know. Maybe Democrats are 58. And they would be 61 or something but it's not really. Going to make much difference. When we look at the polling across the country and the senate contests how does that compare to past cycles are we getting a lot of senate polling more than 2008 team last. Is there let's focus on the senate because of the presidential race. I mean it's. Decently well covered. I in general we're getting liked a lot of kindness pollsters learn the lessons ABC news or parent company must've coast right. They hardly ever did state polls last cycle they're doing like. Several waves of staples this year. Our friends at the upshot. And CNN are doing we have state polls in these Kaiser Family Foundation state polls. I mean definitely written or are like the presidential seats for you get. The double dip right you know in. North Carolina. And Arizona. Pain you have you know Minnesota would Zimmer defending his or her seat excuse me. You know in those seats are getting a lot of double dipping right. The states we have a competitive senate race. But not a contender presidential race you know like South Carolina. You know at this point Colorado was barely competitive in the presidential race. Montana's. Probably neck and head of the presidential race and you montagne has been a little bit under poll but it's not. It's not terrible at me pollsters got the message that like. Pace state polls are a lot more and former utility. The national polls and one reason for that is that in some of the State's yep and have a competitive senate raced him. So I want to move on and asks. Listener questions. But before we due back. Just a general question. When is the house forecast coming out and can you preview republicans' chances of winning house. Fastballs but 90% done so they just connect komando. Relatively seen win. Win and we final are you seeing on it. It's not gonna say anything surprising in and Republicans are our. Had a credit. They're barely even. Putting up an effort to contest in half house seats really make it competitive it's a narrower playing field and into when he. I'm eighteen. And they can win it but that probably involves a case where like just everything is going to hell for Democrats some polls are way off and it's a democratic nightmare. Any scenario where Democrats. Win the senate they're very likely to you have retained the house as well. Probably by a fairly large margin in effect Democrats. My equity in the house even in the event of a trump. Presidency. That from sides are. Are better by some magnitude to Republican sides in the house. How's. Our let's get to some listener questions. We agenda got a lot of listener questions are things to ever run for sending those in our first question comes from no and me I think this might be a question that you've gotten a lot. Secure Garrett Joseph Biden's odds of waiting have gone up according to the forecast as his lead in the national polling. Hasn't really changed. Why is this and then just to give an example of the numbers at the end of August Biden pat as 67% chance of winning. Now he has a 75% chance of winning. Whereas at the end of August he had a seven point one point lead nationally and army as a six point nine point lead nationally. So not much of a difference in the national lead a noticeable difference in his odds of winning the presidency so what's going. So there is a main thing and it's a secondary thing. The main thing is that as you get closer to the election. Then there's less and certainty. On as a kind of famously talked about. We think the selection that comes with fairly high uncertainty for various reasons but a lot of that is this notion that with all the crazy news and all the crazy economic data that like. These could just the very volatile. And instead the polls have been fairly steady rights of every day that goes by. Without. Trumped. Closing the race is a big day for buying right. I think our model with the current sending the polls we get opted. You know Biden being added. And in the never exactly probably 88 issue in ninety ish percent. Favorite on Election Day with his current lead in the polls. So the past two time helps bind my toot our model has eight prior. That expects the rate race to tighten. So it varies does tighten habit that's in line with the models expectations of a dozen titan and Biden may actually be gaining relative to winner. The model thought the race would be that's that that's that main thing and how much does it expected to tighten violent and well it's confusing because it's still kind of thinks trumps. Eliminate a convention bounce which are sits in the direction that then it'll tighten after that right but it expects. Actually not that much of this clinics are to be like half a point her points were the tightening. The secondary thing is that. Is that. National polls are not used. That heavily by the model right people mistake. National polls. For our forecasts of the popular vote which is not true right national polls are. Just one way you can forecast the popular. You know we can do it. Is to kind of add up. All your state by state estimates and in fact that's closer what the model does. To national polls are influential in the model. Mainly because what we call it trend line adjustment which is let's it has been no polling. And I don't know from Alaska for a month and in that time trump has gained. Two points. More or less with the model will do is. It is taken result that we had a month to go in Alaska add two points a truck for it. And then follow given updated assessment of where that rates currently stands in Alaska. There are some more wrinkles uses similar state to tetris but. Right now we actually have quite a lot of recent state polling data so we don't have to resort to using a poll from a month ago in. Minnesota we have 37000. Poles of Minnesota. In the past week. And so therefore there. National polls don't really have much influence on the forecaster. So are you saying that bite in has improved his standing in state polls as an optional polls have remained steady and that's part of why Noah national polls have have probably tighten today. You know I mean trumps leads up to her deficits. Doubt like six point seven or something as of this morning. It was eaten half or nine at one point that's tightening right if you look at the state by state polling averages. Do you wouldn't see that much tightening U mites the might see half a point to a point where they're also states. Like Arizona and Minnesota were compass polling better than he was pre convention. Groups it's it was customers calling like at least as well as he was pre convention. So. So it's not clear if anything to lose like we have a higher quality set. Of data in the state polls this year. In lot of the expensive high budget pollsters have gravitated toward state polls which is good. A lot of weird you know. Canadian ticket. On line firms have been doing national polls and seem to be afraid to get in the states and things like that and so. So it's higher quality data in the states and that's some reason to it to be more skeptical of the notion that the race is tightening but he did the race can't tighten a bit. But that can be outweighed by the passage of time. Right and right just to clarify. What's a reason that you know we would see the race tightened nationally but not tighten in the battleground states. I am turning to trust liked it's not clear the race is tightening nationally right I believe that polling whereas we have better quality point in the status as it's about. Where as per worker perhaps strike that now. The poll average is not our estimate the popular vote it's our international polls which is when pollsters right if you think of national polls is like. Particularly as like. One really big state does that make sense right it's a polling series. Of polls and our national right but like but it's not a summation of all the individual states necessary can be a gap there. And where that gap exists in which when he twelve. National polls kind of showed. Barack Obama and Mitt Romney. Nearly tied maybe about a head by a point or two right. If you look at state polls in this it okay here's the average every state lets wait by the picked turnout and added together. About eleven more like two or 3.2 when a winning by four points right so when there's a gap between the state polls. And the national polls the state polls if you do the work. Can actually give you a better estimate of the national vote in the national polls themselves. So that's kind of the work updated map here. Next question. Does the model take into consideration and how much of the population has already voted so for example if the race were to tighten in the final week or so. But because of early votes some percentage of the population has already voted and won't be affected by that tightening. Does the model try to account for that and any my. No I mean look. In intrinsically the mile and it's more confident as you close election intrinsically polls get better as you close the election or more accurate I should say. I certainly think it's plausible. That when you have them worker early voting. That polls there for a lock in a bit earlier. I would keep in mind that in 2016. We had to share the dramatic late swing toward trump in the polls that are model picked up maybe directing it quite enough of it but like it's really picked up. Some big shifts after the company letter. And there are a lot of theories about how their I didn't mail voting numbers are good for Clinton in this feared here's food. Not to be very wise. So I don't know I especially in the year we have some people using a different voting method when they have before. Pollster to have trouble getting a handle that they might be surprised by a certain things right so why I think this is like. Like that's a change you might implement. In between between four model. If the polls are accurate this year including seats for you have a bunch of mail voting flippers and I would not. Count on that for say. To be something pulses will have an easy time with this year although the K it's it's a plausible theory. Our next pressure and has from chetry yes what with a model output currently be if every state operated like Maine and Nebraska awarding two delegates to the statewide winner. And all other delegates based on congressional district voting what this system tend to favor democratic or Republican camp. I'm. It would tend to favor. Republicans because that median congressional district is a little bit. To their right. Of the median voter. Because of kind of gerrymandering left over from. 2010. And some clustering of Democrats in urban districts. Now this has changed a bit because a lot of eggs urban suburban district room drawing to fear Republicans are now much more purple. Brittany when it fever republic cans. Overall we're would favor Republicans relative to the current Electoral College system that we. Hello I'm. In might be pretty close. It. Like in 2012. Mitt Romney would have done much better with that matter. For example but between kind of the state. Level map getting worse for Democrats. And the districts getting better than it would be. Pretty clothes but both would have kind of date. The built in Republican bias in the main issue with that as like. You know either for make gerrymandering. Super important have you toughness for the presidential race. It's a person is fine if if North Carolina. Is if Nebraska name why do it although I will say Nebraska and Maine. There's a code. That I have to write every year because of you. Georgia your bread cost Georgia have to write a lot in Louisiana got for the Louisiana I have to write a lot of extra code. Every year because of all of you states like you. MAME with their ranked choice voting. Seen stopping Yuri doing the congressional districts what you can't to rich was voting to a code. I mean come on guys. But anyway we also got a like I like Maine I'll put up with Maine if it was like like. Or give her something that was making me this I mean just forget it it. Time we gotta want to questions about her out and essentially how we consider pollsters different turnout models if today. Published various different polls based on different turnout models. And and also there's the question we got a lot which the answer is no by. Who do we'd do our own turnaround model to try to white filter the polls to what we think the current will be so yes it matters know how do we consider polls that house. Different turnout models. We do you protect turn out in the sense that we need to know how to forecast the popular vote. In a forecast a popular vote. Which can. Doesn't really matter that we give a popular vote forecast if you care about it condition Electoral College forecast. Yes kind of know at least the relative turnout is in each state. So. But we don't do those who don't say okay. We think turning that's going to be higher than the polls expect among this group therefore let's shift. The polls to Biden or whatever order trump we trust the pollsters themselves to do. So we kind of protect turnout to protect the vote based in the polls for the kind of the kind of vital to another record it really affect each other. We're doing a pulsar is multiple turnout models well we. We always use. Anything they call likely voters over anything they call registered voters. In general like the voter polls are more accurate. You have some polls that have multiple versions of the likely voter model. But we generally do is just average those together on some Monmouth for example. Today. They're recording its implicate. High turnout model and a low turnout model in Arizona when issued a tie when showed Biden plus two. We averaged just to get you on with Biden plus one right they also had a registered voter pulled had Biden plus or we don't use that Richard numbers at all. The one exception is sometimes a pulse you'll say. Here is our mean likely voter model. And here are some alternative scenarios. Right. If they pollster does that then we defer to them and say okay this is their preferred likely voter right. It's again it's a little complicated inhabitants like patchwork of rules have evolved over the years that's how we handle that particular issue. Next session is from Rick cash it's essentially. Does the margin of error that a poll coms where a fact how much the importance we place on it in our polling our partners are. Say. The margin of error of one poll of Pennsylvania it is 5% and another poll is 3% does the one with 3% get more we were. Or how to we accounts. So the pull to larger sample size gets more weight. And margin of air more or less as an indication of sample size. We don't you sample size Kirk Sadie. Beck has some polls are. Are more honest about their margin of error than others. And they are generally better poles that do that right so. There's difficulty design defect. Which means that when you Q a lot of waiting to be EI GHT in a poll. It increases the margin of air. Some pollsters will be honest about it said hey we had to do a lot of waiting here to get enough Hispanics are an African class white voters right. Therefore the march of Ayers actually 5% net 4%. We don't want to punish that Paul get a polls are doing waiting to they're probably just not. Telling you about it price who use Wii is a sample sizes short version did to figure how much weight to put on poles are in our averages. All right we are to mark Preston the last one is pretty nerdy and applies to people who have probably been following closely. All of the polls that we have been adding tear the model the question is. What on earth is going on with the USC's torrent sites tracking poll isn't considered any differently by the model. And to give some context for people who have not closely been following the USC and site tracking poll. Recently over the last couple weeks it's gone from at twelve point I didn't meet you seven point Biden meet where we haven't see in the same tightening for say. Quite nationally. What's up with apple. So there is a bit flaw. In the USC. Dorn site tracking poll. Mona the pollsters and the US CPR department. You can Simeon and asking us you want but I think you get a not every couple job with the times poll USC right but it just discovered. This morning. So the USC poll is what's called a panel survey which means they re interview the same people. I like panel survey recently EC. Movement right there some concern that. OK well let's see happening it stuck with a really Biden leading your current leading panel. When you're stuck with it the whole year which I agree is not ideal but like it's like to see whether the same voters are changing. Their minds right. On so in 2016 another blow work is they would report results last seven days. Every seven days when seven other sample was asked how do you feel today right have a chance to respond. And so the whole panel for presented. In the poll. This year. Me. Are showing you what happened last seven days but they're only healing people with every fourteen days which means it any given time. Only half the panel to present a kind of us its back and forth between one half and the other half. And if you look at their pull it kind of public a sine wave pattern we're they're seems to be a mortar. Trump leaning group just by chance alone write any more Biden leaning group. And depending and where you are. In that two week cycle. That will affect whether Biden your trump has relatively good numbers so this is like eight I just service. This morning the sink pre. Strange. Way to do the tracking poll it's kind of the worst of both worlds now right you're not really do Aaron regular seeing. Is not necessarily based on news events or people changing their minds. If the fact that the sample has completely switched. It's like you're married around right in most scientific or terrorist elements and a carousel. A little bit more bite me and went tussle that we're trump be right. And so the move it is kind of it's not fake but like. But the whole point of it panel survey as like. Yeah are serving the scene. People. Over and over right. Where some panels we're you do a mix of older and newer people but they're kind of it's kind of like. In the uncanny valley were like it's not any random sample every time it's also an early letting you show movement in effect like. The said people that are in the poll this week article the opposite of the people that were in the poll. One week ago but the same as people are in the poll two weeks ago it just doesn't make. A lot of sense my advice. USC to incite. The we doing this I mean I don't know why do this way right but go ahead and show a fourteen day interval and that's at least kind of honest where electors showing. The whole panel neck and have these kind of phantoms wings that are you know I don't know I think it's it's. It's a big problem and it doesn't change how if at the polar averages were not making like subjective adjustments based on our feelings about a poll we don't treat. Panel servicing differently maybe we should. But it does mean I think people should be very careful about. About making inferences about how the race is changing it's also a bit lagging too weird actually. But Alicia he what are we fourteen days and you have fourteen more days to respond. And so you may say oh there's a makeshift word Biden today right. But actually reflects the change from voters view from 27 days ago or something right and so it just. It's a cool concept. Like they just gave this changed its makes it really hard to interpret and makes it kind of useless is gonna probably does. Like sine wave pattern where there is. Movement at this really track with anything else zone yet. Are the last question is and it reads my question for me it is if you absolutely had to place at bat on the forecast does he lean one way or the other. If so why would you I would use owl. One candidate or the other based on the forecast model today accidents as the presidential. I will say what I've said before her in writing and on the show which is that. You know. The model came out with. A lower probability for Biden. And I expected intuitively and it's a friend since king Browner forecast right. So maybe there's not that disconnect. Anymore. Nothing else say it is. You know I he think. IA think. That all of the incentives and when he. Twenty argued the people overweight the last example over fight the last war right and same kind of all of the incentives for pollsters etc. are to be kind of quite careful certainly all the insanity people that are. Formerly in the conventional wisdom right you can't go wrong by saying oh trips look at a chance and we paid we get it to right. Our headliner kind of story in the market that last month was don't count trump out right in part because there's other models we think we're read to bulletin Biden. So we feel like we have some actual. Not hypocritical send to say that when there are you know there are some miles we think we're very. Eggs are that exorbitant exuberant on Biden's chances. But. But look. So I don't know other hand like this I think in some ways like the model makes. Assumptions that are a little bit on the conservative side other hand having Biden with a roughly three out of four chance. Feels like a pretty sensible. Position for a model to be in. Given where the races. Right now. Indiana has increased rates gone from. You know this hasn't increased that much I went from like 71 to like 67 to eight. To 75 or 76 right so there's some you see some shifts there's result of kind of getting near the election. But yet if they actually had to bet I think people are dumb about this stuff right people accounts can be some. Robot. Placing bets in my forecast amino I mean you would like you know I faced a lot of professional help you activity in some ways right and there are lots of like kind of consequences to the way people perceive our. Forecast for some reality of them and kind of how candidates win in what consequences they win by. And this consulates are probably larger than. And some stupid amount of money that most people are getting in predicted or whatever right. I mean the people like oh you know if you really its union for so I might makes and that's if I were ethnically keep in contractually allowed to do so right the bullish that but I don't have skin in the game I mean like. You I have like so much worse in the game anybody else that actually get the forecast right. You know you're better president and doesn't aux sable predict perceive are forecast to be wrong that's pretty fat for a five theory in terms and again error all types of issues about what's going to be perceived to be wrong or right I think people become more educated about that but I have all the incidents in the world to have the most accurate forecasts that we care right. So yeah. All right well that's good tips and on says thank you may. They keep it. Ultimately some now I know its data and we like to see yet. Meant it it's because you have skin in the game like you sat. Anyway of course we will have the opportunity to more listener questions in the future so if you have any questions and mart. And the -- way either on Twitter horror to podcast at five parties. Dot com. But for now my name is Damon dripped on each count is in the virtual control room you can get in touch by emailing us at pot assets are 38 dot com juveniles of course greeted us with any questions or comments. If you're a fan of the show beavers a reading reviews in the apple podcast store or tell someone about us also makes or subscribe to us on YouTube. Listening to this right now what you should know that you can also watch it you accuse. But anyway things are listening and. Yeah.

This transcript has been automatically generated and may not be 100% accurate.

{"duration":"44:06","description":"In this edition of \"Model Talk,\" Nate Silver and Galen Druke introduce our 2020 Senate Forecast model.","mediaType":"default","section":"ABCNews/fivethirtyeight","id":"73092222","title":"Model Talk: Democrats are slightly favored to win the Senate | FiveThirtyEight","url":"/fivethirtyeight/video/model-talk-democrats-slightly-favored-win-senate-fivethirtyeight-73092222"}