How election forecasting has similarities to sports analysis

0:10Welcome to the sports media podcast. I'm your host Richard Deitsch. My producer is James Jackson. One guest on this episode, we though are taking a detour from the usual sports media conversations. I don't know how much like we can do on, you know, JJ Watt, Tom Brady, Greg Olson, Troy Aikman. But apparently according to the downloads, we can't do enough. But I want to take a detour on this because I am fascinated by the US midterms. And particularly, obviously, as an American living in Canada, it is very, very impactful to me. And the gentleman who I'm bringing on today, in my opinion, is absolutely one of the best if you want to follow someone who's an election forecaster, because he's incredibly thoughtful, works for an incredible organization, and helps educate me and I think those who read him on what is obviously a very, very important topic.

0:59Kyle Kondik is the managing editor of Sabato's Crystal Ball. That's the University of Virginia Center for Politics. It is a nonpartisan election forecasting newsletter. So he analyzes American campaigns and elections. He's a guy who's frequently a guest on places like C-SPAN. If you want to check out this site, and if you're into politics, you probably already know it, but center for politics.org slash crystal ball. That's where you can get like what the updated, their updated sort of, um, places right now in terms of the house and the Senate, giving Kyle a little bit more promotion here on election night center for politics is doing a live stream with Katie Kirk, Jim Acosta and Tara Setmeyer from the center in Charlottesville.

1:40So you can go on YouTube center of politics is YouTube page. Check out that live stream. Kyle will also be in New York City with the CBS Decision Desk, where they have a new results-focused nerd show stream, which CBS is doing for the first time. Obviously, you've seen NBC done a ton of this with Kornacki, etc. So this is one of the foremost, at least in my opinion, election analysts in the United States. He's also a big Browns fan. He can't be perfect. Kyle, welcome to the podcast. Richard, I've listened to your podcast many times, and I am very flattered that you asked me to come on. Yeah, absolutely.

2:15Okay, so here's where I want to start. And I think both of us are in agreement on this. There are, in my opinion, a ton of similarities between people who are interested in elections, election forecasting, figuring out what the early vote may or may not mean. And those who work in sports, whether it's sports analytics, whether you're working for a team, and you're trying to figure out what free agent might fit for you. in some ways you could distill it to the fact that it's a way to analyze data, right? To present it to an audience that makes the audience either informed, or if you're working for a sports team, that makes your sports organization better.

2:49Do you agree with me? And if you see the parallels as a sports fan too, how do you see these two, uh, kind of intersecting with each other? Uh, I absolutely agree with that. You know, both sports and politics are data-driven, although I think there are some key differences, you know, I guess, I guess what probably baseball is the most analytically driven of the, you know, the major, the major American sports. And like, you can, you know, you're not gonna be perfect on all your calls. But but I mean, and I feel like the amount of data that particularly in baseball that there is, is like, it's just kind of incredible.

3:19And so you can know that, you know, okay, this guy hits lefties at 400, and he hits righties at 150. So you can make a pretty informed choice about it. I'd say in politics, sometimes the data we have is actually like, you don't actually know how accurate it is. I mean, And that's not to say that if you make the right choice in baseball, you still might make a mistake. But we're so reliant on polling, for instance, and polling is very difficult. And sometimes you're not getting exactly what you think you're getting. And so all the polls can point to one thing and then maybe something else happens.

3:53So I feel like some of the data from the sports is sort of harder in that you at least know it's like, okay, we've done this. We've got a thousand at bats with this guy and we know what the numbers are. But sometimes in politics, it's like, oh, you might have a hundred polls and they all might have something wonky with them, I guess. But it is all data driven. And there's lots of things that you try to take into account, particularly close to the election. And some of them, some might be more valuable than others. There's all this early vote analysis that goes on. And yet the baseline you compare it to is constantly changing.

4:28So like, I think maybe 10 years from now, when early voting is kind of more of a standard thing in the US, and we have more set patterns, we might it might have more predictive value. But it's hard to know now how good some of it is. I say broadly speaking, the early vote in the US election is kind of reflects the polls in that, you know, Democrats seem to be doing pretty well, you know, Democratic motivation is up, Republican motivation, not quite there. But, you know, it seems like there were more Democrats who wanted to vote by mail in 2022 because of COVID. And so you have to take that into account.

5:00So it's like, oh, well, the mail voting is down. And so therefore, Democrats are doing worse. And it's like, well, maybe they're voting in person. So like, there's all this stuff that's confounding. I guess that's also true in sports as well. That, you know, it's like, you know, you think about the numbers and then you kind of, you drill down on them. And then you could find all these reasons as to why that they're not right or they're weak or whatever. And this is also, I think, a good way to sort of detour into some of the election modeling because, you know, there are the quote unquote qualitative forecasters like myself and like in the crystal ball and like Cook Political Report and Inside Elections, the former Rothenberg political report, Nathan Gonzalez does Inside Elections.

5:42And we, you know, we take into account all sorts of things. But ultimately, the ratings that we do are our decision. It's like, I want to change the South Carolina Senate race from likely Republican to Leeds Republican, which we did the other day. Okay, we do that. Nate Silver and Vote Hub and Decision Desk HQ and other things, they're all, you know, they're quantitative models. So there's like poll based and also lots of historical factors. And they build a model and then it sort of spits out the projection. So there's a little bit of a difference there. But we're all we're all kind of doing the same thing, trying to look at who's favorite and who's not.

6:15That's really interesting. And we'll get to we're taping the six days before Election Day. So on Wednesday, October 28th, and we will get to what the center has right now in terms of the projections that they have as of this moment, they will change. Obviously, I think Kyle will confirm that they do make decisions right before the election. We push all of our toss ups one way or the other at the end of the election, which is bold and foolhardy because, you know, a lot of these races end up right on the butter knife. But we feel like it's like, you know, we follow these things all year. will like kind of go on the record, at least with a lean, a lean everywhere.

6:51So the question I wanted to ask you is like, obviously, I am someone who's very knee deep in a lot of the people you mentioned. I have my own list on Twitter, which basically is like election Twitter. And there's like 75 people, including people who are really like into certainly just one state, like their their their expertise might just be Iowa or Kansas. And it's really interesting to me. But I am curious, just given where you work, and it's a very, very well known place. you're obviously looking, I imagine, as much polling as possible. But do you also talk to campaigns? Or do you talk to insiders or operatives?

7:25Because I would think that to really be informed, there is value in going beyond, I would think, the numbers in some ways to at least maybe talk to people who are on the ground. Even if they're trying to spin you a little bit, there's value in hearing perhaps from both sides what their internal compass is. Do you guys see that at all? Yeah, we talk to people here and there. We have, you know, folks, lots of places in the country that we've talked to in the past and also some sort of steady hands. You know, we basically all of the stuff is kind of off record. I mean, it's off record in the sense that we generally don't quote anyone.

7:58We're generally not sharing, you know, internal polling and whatnot. And but, you know, a lot of what we do is based on what's out in the public public space, too. I'd say particularly this time of year. I mean, one of the big signals that you sometimes get is the party committees will be moving money around. And there's always reporting about like, oh, well, the Republicans are not spending in this district anymore. That looks like they're kind of giving up on this. That's the perspective of this year in 2026. It's a good looking Democratic year to at least some extent, maybe to a lot of extent.

8:30You know, the Republicans are retrenching here. The Democrats are moving in. And in some ways, the money talks more so than maybe whatever whatever comments you might you might get. And, you know, we do have there are people who will reach out sometimes from campaigns because they feel like they want they want a different rating, basically. And, you know, we hear that we hear those folks out. We don't we don't always do what people want. There are some people who are who want to be in like toss up races when they're not. Because I think it could actually help with like fundraising and stuff. So, you know, yeah, there's there's there's some of that.

9:05But, you know, we end up being quoted, you know, just a lot of news stories and, you know, we're on TV and whatnot. But we talked to a lot of reporters, too. And, you know, reporters can give you good information based on what they're hearing on the ground as well. So there's some of that. But, again, you know, there's not like, you know, there's not some like big secret that is like guiding the work that we're doing. I mean, again, we're hearing from people, but a lot of it is trying to dissect and sort of make sure we're taking in all the information and that is publicly available. I want to ask you about the predictions that you make, because what's interesting in sports, and myself, my producer, James Jackson, obviously can both confirm this.

9:48Everybody makes predictions when it comes to sports, whether it's an ESPN analyst making a prediction on a weekend or a Super Bowl. Obviously, I've had to do that in the sports that I covered at Sports Illustrated and The Athletic. And the reality is like three weeks later, five weeks later, nobody really remembers any of this. Maybe if you have some big, bold prediction that you made, people might remember or like reverse big, bold prediction that you missed and people might get on you for that. I'm wondering for you, though, like how important is it for you to try to get this pretty close? And do you feel the people at the Center for Politics should should be in the end judged on how good their predictions are for a given cycle?

10:26Um, so it is very important to us, particularly because we push everything. And look, you know, we've had our hits and misses over the years, just like anybody else. And so, yeah, like, like I said, it, it, it is, it is important. But, you know, also, I think, I think sports are a little bit different in that there's always something happening next. And, you know, elections, it's like, two years of commentary leading up to one election. So there's like a lot of talk about who's up and who's down and what the campaign's going on. But ultimately, you just – I mean, obviously there are elections that are happening in between.

11:01You have a state Supreme Court election in Wisconsin or something. That's a big event sometimes in the first week in April, for instance. Or there are the primaries or there are special elections and whatnot. In Virginia, Virginia has elections all the time. There's some sort of big election every November in Virginia, which is rare for most states because in the odd numbered years, there's either the governor's race one year or there's the whole state legislature is up in another year that's coming up next year. But but, you know, so it it might it might stick a little bit and be a little bit harder because, again, it's like, OK, you you know, you pick the you pick the Browns to beat the Steelers and it doesn't it doesn't happen.

11:41you know, there's a game next week, you know? And so there's, there's, there's maybe a little bit more there in that there's so much talk that then leads up to this single event and it's not spaced out. I mean, you know, all the house is up, you know, third of the Senate, all these governor's races and whatnot. And, you know, I will also say that I think the elections are, I think they're hard. I mean, like, like the, I think the Senate is hard. The house, I don't like the actual margin I think is maybe hard, but like, it'd be just a shock if the Republicans held the house at this point. But, you know, and then you have to, like, you have all these priors about, you know, at, for much of the cycle, I thought, well, I think the Republicans are favored in the Senate just because the Democrats have to win at least two states in, or two seats in states that Trump won by double digits.

12:33And that recently has been hard just because it feels like the floor is higher for the parties in a lot of places, but the ceiling's also lower. And then just as time has gone on and on and on, it just looks like the Senate was really in play. And you may remember back in 2024, there was this constant focus on seven key swing states, seven key swing states. And as we're talking, there's actually seven Senate races, I think, that are at least somewhat it up in the air. I think, you know, Michigan on the democratic side and then Maine, Alaska, Iowa, Kansas, Texas, and Ohio. I think those are all like, I have a sense as to where I think they might go, but you know, there's a lot of, there's a lot of drama with this too.

13:12And then you also, you don't have a lot of trust in, I don't, I, I love the polls and I love that they're conducted and that there, and there's stuff out there, but I also don't, you can't necessarily trust it perfectly okay so let's there was a i think there was a like i think i think nate silvers talked about this that you know when he i think when he first started doing his like poll based like forecasting model um you know there was a there was this peak in like 08 and 12 where like the poll forecast were like really really pretty you know pretty spot on and the drop came along and the polls weren't quite quite as good and um so that's also that's also part of the challenge So let's get to where, as of now, again, understanding that, as Kyle said, they're going to make formal decisions on the toss ups.

13:57I'll let the listeners know where we are right now. Now, Senate for Politics has for the House 225 at least leaning D, Democratic, 193 at least leaning Republican, 17 toss ups. Their Senate at the moment, 49 at least leaning D, 46 at least leaning R, five toss ups. Governor race is 26 right now leaning D, 22 at least leaning R, two toss ups. Obviously, it amplifies what Kyle just said. The House looks at 225. They're already in majority territory. It's the Senate that's really, really interesting. Okay. So a couple of things off those numbers, Kyle. First, when do you finalize? Do you make, does the, on election day, is that the day where you pick the toss-ups or before?

14:43We usually do it on the day before. Sometime on Monday, we will make the moves. And there may be things we do between now and then, but that's going to be the big day for pushing the toss-ups. Okay. So I want to ask you, what has informed you on these decisions? And obviously, it seems like the House, I think, you're obviously not the only ones on this. Pretty much most people in the business forecast would be, like you said, stunned if the House doesn't go Democratic. It's like 90 percent, 85 percent. If you believe in prediction markets, I'm not sure how much I do. That's obviously where they're sitting.

15:17The Senate, obviously, is kind of the more fascinating setup because we have some of these states where it's like a true toss up. Maine is a perfect example. Susan Collins has always sort of figured out a way to win, regardless of what the wins are at the moment. Troy Jackson obviously leads that in the polling by a little bit, but that's very, very close. Michigan obviously is very tight. And then this is what is very fascinating to me as someone who's been watching this from Canada. You have some races this year, including like Kansas, where the Democrats have never been competitive in terms of the Senate.

15:52And there are some polls that have Hamilton up over Marshall. Like that has never happened in my lifetime. And you can correct me if I'm wrong. I don't think Kansas has elected a Senate, a Republican, a Democratic senator since 1932. So for you as someone who's analyzing this, this is pretty fascinating because like you have to analyze some races that go against historical traditions, right? And it goes against at least everything you might have learned as an election forecaster in terms of past results. Yeah, that's right. And, you know, Kansas, though, there was actually a little bit of doubt in both.

16:29Well, actually, a lot of doubt in 2014, because, you know, there's been this trend of independence kind of running sort of in place of Democrats. I mean, they wouldn't say that. Dan Osborne in Nebraska is a good example. He's running again as an independent after he did in 2024. There's a guy named Greg Orman in Kansas who was running against pat roberts and you know a lot of people who thought or orman was actually going to win um i remember we were like sweating it out last weekend we were like yeah we just have to stick with the republicans here but and we ended up getting it um and we're one of the few who actually did but i mean it was and it ended up being like a 10 or 11 point race though now 2014 was also like a really terrible year for democrats across the country that was uh in some ways it kind of feels like a mirror of this year um except that was a pro-republican year and this is a pro of democratic year to some extent.

17:20So, you know, but you've got this Kansas situation now where it's like Adam Hamilton, the Democrat is either probably tied or slightly leading. But the thing you also ask yourself is like, so if there's a poll that comes out and it's like 47, 46 Hamilton or whatever, it's like, well, yeah, there's a libertarian in the race too. So maybe he'll take a little bit from Roger Marshall, the Republican. But like, who's got the better a chance to actually get the plurality. I think ultimately it's a Republican. And so that, you know, forced to choose, I think that's where I would go in that race. I mean, we, again, we'll, we'll, we'll do our final ratings on that, but it also then gets into a conversation where we talk about poll error and like, you know, as opposed to underestimating the Republicans and whatnot.

18:03And there is some historical research that Patrick Raffini, a Republican pollster has done. Also, The Economist had something about this recently about how it's pretty common for, for Democrats to be understated in blue states and Republicans to be understated in red states. And what would be a reason for that? Well, like if you're showing, you know, if you're, if you don't have, if you've got some undecideds, it sort of stands to reason that those people would sort of break toward the sort of majority party in the state. And so if, you know, if, if the polls show it tied in Kansas and then Marshall wins by four, they say, oh, the polls are off by four points.

18:36The Republicans underestimated again. It's like, well, that might just be like sort of the natural progression of things that, that they were legitimately undecided, but then they're like, well, I got to stick with the, stick with the team, you know? And then there's also the, the, the, the challenge with the undecideds. It's like, what if the decision is that they just don't vote? And that's, I, that to me is the, is the scary thing for Republicans is that, you know, if you've got, you know, Hamilton up one and undecided, a bunch of undecideds, and then the turnout just isn't there on election day, all of a sudden, you know, 47, 46 actually is like 50 to 49 because those undecided bleed away or they go to the third party or whatever.

19:12So that's the thing. But, you know, you sort of, particularly in this sort of partisanized, polarized era, you probably give the benefit of the doubt to sort of the home team party. I mean, Republicans are the home team party in Kansas. And that's what I kind of, as an analyst, because I'm leaning a little bit toward the Democrats in the Senate. But if you were to tell me that, oh, well, actually, Ken Paxton's going to win by one in Texas and John Husted is going to win by one in Ohio and Dan Sullivan's going to win by two in Alaska, because those states have just become so Republican. We shouldn't be shocked if that happens.

19:45And by the way, let me add something else. You asked about the sports predictions or politics predictions before. Yeah, please. This business has taught me a lot of humility, I will say, because I've missed stuff in the past that, hey, we've done well in the past too. But I do think that in a lot of sports media, and I think politics can be like this too, there can be this sort of like chest thumping about this is going to happen and this is for sure the thing that – And I'm you know, I personally am not like that. I mean, you know, we I've actually I think I've probably gotten better at it over the years, just being conveying more uncertainty about the about these things.

20:23the the the two race i mean every one of these senate races fascinates me including the kansas race might be the most fascinating race that i've been following um but the the the two i want to ask you about at the moment are texas and maine because the the the reason why they are so interesting is that i think in both places there's a disbelief on one side that something can happen So in the Texas case, James Tallarico has essentially been leading the polls even slightly for a long time now. You'd be able to tell me maybe there's one or two outlying Ken Paxton polls where he leads. But generally speaking, Tallarico has led throughout the early vote numbers have been pretty good for the Democrats just in terms of enthusiasm, etc.

21:06So like if you were going to make a prediction, I think just on like polls and sort of what your what your eyes tell you, you might take Tallarico. The problem is it's still Texas, you know, with this long history of of of Republican senators. And, you know, we've all been alive for for Ted Cruz supposedly going down, et cetera. You know, he couldn't be Beto, couldn't put him away. And, you know, all these other sort of races where eventually the Republican pulls out in Maine, a similar thing. I remember the polling. It was so off. Sarah Gideon was expected to win. There were some polls that had her beating Susan Collins.

21:44It never happened. In this case, again, Troy Jackson has been leading a lot of polls. But I think there's like this belief that once again, Susan Collins is sort of going to overcome the the wins that exist. So when you think about these races, I'm curious, like how much, how do I sort of ask this question? Like, how much do you trust your sort of eyes? You know what I'm saying? And the data that's in front of you versus all of this historical data that I think a logical and smart person would be like, I have to trust this historical data because X has never happened before. Yeah. Yeah. I mean, the, and the history can be informative and, and, you know, one, one call that I am kind of proud of that we made over the years is that, um, back in 2020, we only missed one state, North Carolina, you know, for Biden, but one that we picked, we, we had Trump in Florida and very few.

22:36I remember that. That's correct. Very few people had that. Um, and that history was, was part of it too, because, you know, just in 2018, you know, the, that was actually the site of what at the time I thought was a pretty big upset, which was that, you know, Rick Scott beat Bill Nelson in the Senate race down there by just, you know, inches, basically, you know, a tenth of a percentage, tenth or two tenths of a percentage point. And, you know, it turns out that it sort of makes more sense now because Florida has moved considerably more Republican in that time. And Trump did markedly better in Florida in 2020 than he'd done in 2016.

23:13And then That's an instance where the polls, the history was a better guide. And I constantly, I think about how in certain races, sometimes you can overanalyze yourself to death. Like, let me give you an example of this in this year's election. So I think if you really look at Michigan, okay, with Abdul El-Sayed, who I think is probably going to run behind the usual baseline for a Democrat in that state versus Republican and Mike Rogers. I think the more you look at that, the more you might think, oh, wow, Rogers, like Rogers may actually win here. There's these little weaknesses for Abdul Sayed and this and that.

23:52But I think if you sort of take a step back and think about it for like two seconds, you think, oh, actually, Elsa probably should win. And actually, I'm on that race. I'm in the look at it for five seconds. I think the first blush analysis there in that state is right. But the thing is, is that that sort of thinking doesn't apply everywhere because sometimes the deeper dive actually will tell you more. So, you know, again, it's not like there's like a consistent thing you can analyze here. And I would say Kansas is probably the other way. Kansas is where I think the sort of quick and dirty analysis would say, oh, yeah, Republicans win Kansas.

24:29And that I feel like that's probably a safer place to be at. I mean, you know, now you mentioned Maine and Texas. I've been citing those as I think the two most important, or at least I think they're important because I feel the most conflicted about them. And so therefore, in my head, those are the ones that are the deciders for me in terms of who I think is going to win the Senate. I don't think Democrats need to win them both, but I think they need to win at least one of the two. Because the other ones, I'm not giving anything away here necessarily because we sort of signaled this in a piece we wrote earlier this week.

25:03But like, you know, we have Michigan as lean down in our ratings. We have Alaska as lean down in our ratings. And, you know, those are both tricky races. Others look at those as toss-ups. You know, Ohio, I want to see all the data there, but it looks like Sherrod Brown is probably ahead there. You know, Kansas, I said before, probably toward the Republicans ultimately. And I think Iowa is that way as well. Agreed. But that's, so I'd mentioned that, you know, see seven key races. Well, the two that are left are Maine and Texas. I don't feel like I have as good of a grasp on those. Again, maybe I'm wrong on some of these other ones, too.

25:36I mean, it's tricky. But yeah, they're vastly different states. And you also have, you know, there's a lot of talk about candidate quality this year. You know, Collins obviously has great candidate quality. She's proven she's won in the past before. She is the only Republican senator in the state that Kamala Harris won. Ken Paxton, the Republican in Texas, he's been a pretty poor candidate. And, you know, so there's that there's that contrast there. But, yeah, there's also the to your point, the sort of unbelievability. Now, you know, I kind of I kind of wonder if because polling was off on Collins in Maine in 2020 and like that wasn't it wasn't just Maine.

26:17I mean, you know, Trump did better than polls suggested, even though he lost. And one of the consequences of that is that some of the down ballot races look bluer than they actually were. But I almost wonder if we maybe are overcorrecting a little too much on Collins' strength in that race. And then Texas, I've kept waiting and waiting and waiting for Paxton to take the lead for these attacks on Tallarico to stick. Maybe that's where it'll end up, but here we are just days from the election and there's not much polling evidence that that's actually happening. Yeah, I'll just get it on the record for myself.

26:50If I was going to analyze this, I would absolutely look for the upset in Kansas before Iowa. I'm with you. I think Hinson generally is a quality candidate for Iowa. I think Marshall is a problematic candidate for Kansas. But I would never be confident enough to throw my actual money on the Kansas race because, as you have talked about here, I just would be a little too worried about this historical trend. But that is one that I've tried to read everything on. I'm following people who are really deep into the Kansas data. And that data, at least to my knucklehead eyes, feels like they're in a better place right now early for a potential upset than Iowa.

27:36But who knows? The reason I want to bring up early voting is because so many people now do it. There's so many people, particularly online, who follow it. And one of the things that I learned in terms of just reading this stuff, obviously now for years, is the one place that I really actually trust early data is Nevada and guys like John Ralston and John Samuelson, because that seems to be the one state where you can actually use the early data to give a real analysis of where things are on election day. I wonder if you can explain to my listener. One, is that correct in your mind? And two, if you can explain to my listeners why Nevada is like one of the few states where the early data can actually inform you as to the probabilities of what would happen on election day.

28:20Yeah, because you get a sense that, you know, that state does have registration by party, and so you could kind of look at that mix, and Ralston's had his way of doing it over the years, but, you know, you can kind of look at the pace by day and compare it to the previous. So even there, you know, there's been sort of an explosion in mail-in voting, which skews things a little bit. And also, you know, there are a lot of voters who aren't affiliated, and that those folks can be kind of a black box. I mean, I think that, you know, in a year like 2026, I think you might expect on affiliates to be, to, to lean more toward Democrats.

28:54I mean, that's what polling suggests. But you don't have that hard party ID label. And there are certain states, I mean, people sometimes will talk about like the Michigan is like a hard state to poll. Well, part of it is they don't have registration by party. Yeah. And, and, you know, particularly like in the primary, like, you know, in any primary, you don't, you know, you don't, you don't have, obviously, you know, they're all, you know, if you're in a Democratic primary, you're probably a Democrat or, you know, but you may even be, maybe Republicans crossing over just following where the, where the action is.

29:23And there are a number of states that don't have, you know, their party registration and Wisconsin is that way. Ohio has like a soft form of party registration. Basically whatever ballot you take is what you are. So if I take the Republican ballot, I'm Republican, but then next time I can take the democratic ballot and I'll be a Democrat. Um, but you know, then like, like a Pennsylvania is a state where you've got really kind of hard party registration and some unaffiliated, but not, not a ton. Um, but so yeah, the Nevada has, has been, it's been helpful over the years to look at the early vote, but even that is not, uh, is not foolproof.

29:57There are certain people like I, I always read Ralston about Nevada. There's a, uh, a group called vote vote hub, um, uh, A young analyst named Zach Danini, who I think is brilliant. Yeah, does a great job. They've been doing a lot of early vote analysis and comparing it to previous years. And again, I think it's telling a pretty consistent story about, as I mentioned earlier, that Democrats are doing better. Republicans are lagging a little bit. Maybe they're catching up in certain places. I remember back in 2024, and I was more bullish on Harris than that ended up happening. So I'm not blaming others for this, but I do remember just tracking it.

30:36It seemed like Democrats might have been making a little bit of a comeback near the end, but even that, it just ended up being kind of fool's gold. And, you know, whatever the early voting, you know, whatever you can glean from early voting, it doesn't actually tell you about persuasion. Kind of need the polling piece of it and then obviously the actual results to let you know what the persuasion aspect of it is. Because you can have electorates in which the party mix is the same as it's been in the past, but the result's actually a lot different. because maybe the R's are not as cohesive as the D's are and the unaffiliated may be breaking one way or the, uh, or the other.

31:15So again, it's, uh, um, you know, again, it's, it's, it's just challenging. I think over time, the early voting will probably be more and more helpful again, because it's like the longer the track record is that more people settle into established routines, um, the, the more useful, I think, I think it'll, it'll be. All right. A couple more here and then I'll let you go. One of the, I'm fascinated by how you approach this, when there's a poll from a very respected pollster that just feels like something that's just so different or an outlier versus everything you've else have seen. And the reason I bring that up is because today, as we're taping this on Wednesday, the New York Times Siena poll, which I think a lot of people believe is a very, very quality poll historically, had in Minnesota, Peggy Flanagan up 5440 over Michelle Tafoya.

32:03So again, Minnesota has generally speaking, been a left-leaning Democratic state, but that is a gigantic spread, which at least for my purposes, from all the other polling I've seen would be like, I'm not sure I believe that. That seems way outlier. The other one, perhaps even more fascinating, was Annie Andrews and Darlene Graham in that poll are tied at 48, which again would go against everything historically in South Carolina. The reason I bring those two polls up, Kyle, is I want to ask you, when you see something like that from a very reputable pollster? Like, do you, are you in one of those people?

32:36Hey, I'm just throwing it into the, to the average and we'll figure it out. Or do you step back when someone with the credibility of that poll says something that is very different than anyone else has said? Well, and you know, there's a, there's a famous example to the, you know, the Ann Selzer poll in 2024, which ended up being way off, but Selzer, she'd really been good at producing outlier polls that actually were signals. And so that's why I really, I did take the 2024 poll seriously. Because not that I thought that Harris was going to win Iowa, but that I thought, oh, well, maybe there's something going on here.

33:12And it turns out there wasn't anything going on. But again, her final Iowa poll in both 2016 and 2020 was really indicative of what was going on, not just in Iowa, but also in the broader Midwest. And so I do take the New York Times poll seriously. And I will say that they're probably, this is, I don't want to dump on other pollsters. I love pollsters. I love looking at all these polls. I couldn't do my job if people weren't paying for these things. But I do think in terms of like the sort of more highly rated pollsters, the Times-Siena is probably the one that we've seen in Minnesota that that's gone in there and they've seen a higher spread.

33:54And, you know, frankly, I think it probably corresponds a little bit more to reality than Flanagan just being up a little bit given just given what the year is. And I know I know there's some specific problems that Democrats have in Minnesota. The governor, echoing governor, Tim Walz is not popular. There's this social services scandal that is per the administration. But, you know, we are at a kind of a nationalized political era. And I think it'd be a pretty big surprise if Peggy flanagan even though you know she's also kind of the more left-wing candidate there was a she defeated angie craig who uh is more kind of the make your traditional mainstream moderate democrat she had been in a battleground district and i think probably would be would have been a stronger general election candidate i agree um you know it's funny nate cohen at the new york times in his write-up he said yeah this might be our version of the selzer iowa poll because he talked about some of the things i talked about earlier about how you know minnesota is also not a party registration stayed.

34:43And so maybe the samples skewed. And I think we know that there are a lot of Democrats, particularly kind of like older, older white women who just want to take the polls, basically. And so maybe they, that sort of skews things. And it may even be the case that even controlling for demographics, certain Republican leaning demographics, the people who are likely to likelier to respond to polls may also be a little bit more democratic. And pollsters have ways of, of, of, of accounting for that. But, you know, we have Minnesota's like liens democratic. I, you know, that's probably selling them a little short there.

35:17And, and the, you know, the thing about polling is like, sometimes like, oh, I like the polls that confirm my priors, you know, it's like, oh, I have an idea about this race. And oh, this poll is here that, that think that goes the same way I am. And like that Minnesota poll is actually kind of confirms my priors a little bit in that. I think some of the other ones are maybe a little light, on the Democratic lead, even if that one might be a little too heavy. And South Carolina, you know, same thing. We talk about Kansas, and it's like, you know, 47-47 or 48-48. In that state, what are those, that last bit of undecideds, you'd probably expect it to be Republican leaning.

35:53It is interesting, you know, the late Senator Lindsey Graham, he always had weakness in his primaries in, you know, upstate, you know, northwestern Colorado, South Carolina, like Greenville and Spartanburg, which are very Republican places, but Graham was never the right fit for them. And so he always had weakness. And that poll indicated some weakness there. Now, what does that also tell me? It's like, well, those folks will probably come home and Graham will probably still, you know, Darlene Graham will probably still be okay. But that's, you know, that's, I mean, one of the stories, the election is you've got the Carolinas and it's like North Carolina.

36:29It looks like Roy Cooper is just on the glide path right now. Yep. And Darlene Graham is in real trouble. You know, start of this, you know, six months ago, you said, oh, North Carolina probably ended up being, you know, more of a dogfight and South Carolina should be no problem. And that's, that's not where we're at, you know, right here before the election. Yeah, that would be an earthquake if that happened. And if the turnout is low on election day, that would be your possibility of that happening. My final two for you, I want to ask you where we should look in terms of early in the election results coming in.

37:02And I'm not sure, I don't, You might have written this, but I know there's a lot of smart people who usually write this. And tell me if I'm incorrect on this. They say that the fourth congressional district of Kentucky, which I think the vote – The sixth. The sixth, I'm sorry. The sixth congressional district of Kentucky where the vote – where I think the close is pretty early for the polls. Yeah, that's one of the – I think they're Indiana and Kentucky, and I'm not even sure it's the whole state in both instances. But you're going to get some results in the 6 p.m. hour. Okay. So like, so, so it would you, is it, is it enough of a bellwether where if I'm paying attention to this and it's whatever, six 30 Eastern time, should I go look to see what's going on in the sixth congressional district of Kentucky, which may give me an indication of what's going on in the rest of the country?

37:48Yeah, because that, you know, it's a Trump plus 15 state, but it's kind of historically been a little more competitive down the ballot. And, uh, you know, the parties are spending there. It seems like a real race this time. Um, in 2018, Andy Barr, who's going to be, he's now running for Mitch McConnell, Senate seat, he's going to be the next Senate in Kentucky. But Andy Barr held on against Amy McGrath, who was a veteran and kind of a prominent candidate that year. She later ran unsuccessfully statewide for U.S. Senate herself. But that's a seat that Republicans probably should hold. But if they don't, then that's a real early signal that the House is going to flip, I think, and that it might be pretty big.

38:25So that is one to watch. Although, um, I was talking to, I mentioned, I was talking to Anthony Salvato. He's, you know, running, he runs the election unit at CBS and he's going to be very involved in the, uh, the stream that we're doing the so-called nerd show stream. Um, and, uh, you know, I mentioned Kentucky six to him and he's like, well, Kentucky is not always fast about counting votes. So it might be, we're just kind of sitting around at the six o 'clock hour. We're getting a little trickle in, um, Indiana five, uh, Victoria sparts. That's like, uh, Indianapolis. But isn't that another Trump plus something crazy?

38:56Yeah, it's like Trump plus 17, but it's also, it's got Hamilton County, which is, you know, kind of a historically really Republican, like highly educated, very affluent, suburban, exurban place. It's been getting more competitive over the years and there's been some spending there and polling indicating it's closer. So that's another one that would be, you know, that's one we have as likely Republican. We have Kentucky six as a toss up. And I actually, we just moved it there. We had his liens Republican, but we decided to move it and about a half dozen other districts to toss up just to park them there for the last week of the election, just in case we wanted to flag them as possible Democratic upsets.

39:33That's a big move for you guys to put that as a toss up. I mean, that's. Yeah. And again, it would be indicative probably of a Republican turnout problem because you've got Lexington in that district, the University of Kentucky. And if you get big blue turnout in, I think it's Fayette County, and then the surrounding red areas, maybe it's not quite as good. That's how you can maybe add up to an upset. So those are two that's early in the night. You know, I would, you know, Florida counts votes very fast, early poll close, a lot of competitive house races in Florida, including some places where, you know, Republicans tried to try to gerrymander and they may not win.

40:14So like Jared Moskowitz in South Florida, Kathy Castor, another Democrat in the Tampa area. There's an open seat, Florida 22. too. So those are ones that Republicans made better for them, and yet they may not win them. So those are some to watch early on as well. All right, last one. By the way, I'm going to use my, for Jared Moskowitz, I'm going to use my formula of viral social media and getting publicity, which he is excellent at. And so I will take him to win based on my sort of curating philosophy there, independent of whatever politics exists for him. I know we're already pushing it, talking politics on your show, but, um, you know, uh, Moscow, it's him, Moscow, which is Jewish and is, and is pro Israel, very pro Israel.

40:58Yeah. The district, the district religious statistic, it's kind of harder to get the exact demographics, but, um, you know, I think that this, some people say that it's about 25 % Jewish, which is, which is really quite high. Yeah. And it seems like even if Jewish voters may be moving away from the Democratic Party in certain ways because of. But it's but it's that that's a that's a local guy who, you know, is representing Israel. And so if there's if there's a guy who could be a better fit for that district, then maybe your average Democrat Moskowitz would probably be the person. Yeah, that's that's well said.

41:31All right. Last one is obviously when the midterms are over, you'll dissect it and you'll certainly be writing for the site. And, you know, you'd be a guest on a lot of these places. But when do you really then start to think about 2028 in terms of both the primaries for the Republicans and the Democrats? I mean, I think it's going to get going, you know, right after the election. I mean, we're going to see announcements and that sort of thing. I mean, I don't know if we're going to I don't know how we're going to handle looking at the primary season, but we'll start looking ahead to the calendar, which is, you know, is, you know, the Democrats changed their early state lineup a little bit.

42:01We haven't really written about that yet, but I'm sure we will. And you'll start to see announcements from candidates or kind of soft launches from certain people. So, yeah, I mean, we'll be looking at that. And we usually do our first ratings for like Senate and House for 2028, like probably early next year. House is probably going to be more like gerrymandering attempts. So the House map is probably going to change in certain places. So that's something we'll have to have to account for. But, you know, but there's always there's always something to look forward to. But we do also want to look back at whatever the results are and look for look for the trends and see what's interesting about it.

42:43I mean, that is the great thing. I mean, I mean, American politics. I mean, a lot of it never ends. There's no there's no season. It never ends. And look, we we we have free and fair elections in this country. I'm confident and hopeful that we will have one on Tuesday. And the results themselves, that's what it's it's always really interesting because you would be as an American in Canada, you probably know this. You would be stunned at how the laws exist so that there is such little campaigning. You know, it's all done in a condensed time. Right. By federal statute. It is mind blowing to me as an American sometimes to be up here and not be inundated 24 seven with politics.

43:21Yeah, I mean, it's a permanent campaign here, which is great for people like me and great for the vast majority of other people. But I mean, it is for you. For you, it's great. Last one. This will be fun. I want to get you out of here in this. The political junkie I am, I have watched so many C-SPAN pranks. I apologize to C-SPAN anchors who are awesome. Have you ever been a guest on C-SPAN where someone has called and tried to prank the network slash anchor? I don't think so. I mean, I've had, I do have some, I've had some kind of off the wall callers. I did have someone basically threaten me one time.

43:54Well, that's, that's, that's not, that's not good. It was a, it was a, it was, I was kind of dark. I mean, I don't think it's a real thing. I mean, that's one, one good thing about actually not being that famous is that. Yeah, right. I mean, it's sort of the penalty for. That's still disturbing. I mean, if you're threatening an election forecast, that's pretty disturbing, but anyway. Yeah, I mean, it was a very kind of dark thing, but no, I, you know, and actually a lot of the callers, I mean, sometimes you have to sift through it, but there's something you could glom onto. Yeah, I was just on Washington Journal, I don't know, the last couple of weeks.

44:27I've stopped. They always give you those. They have a mug there. Do you get to keep the mug? Take the mug. And I'm like, I don't need any more mugs. At least they do that. Those guys, C-SPAN does a great job. Yeah, I know they're doing an election special too. I think there are a lot of folks who are doing kind of election night streams and whatnot. All right. That's a good segue for me to give you some more pub here. Kyle Connick is, of course, the managing editor for Sabato's Crystal Ball, which is the University of Virginia Center for Politics's election forecasting newsletter. You can check that out, centerforpolitics.org slash crystalball.

45:03On election night, the Center for Politics will be doing a live stream with Katie Couric, Jim Acosta, and Tara Setmeyer live from Charlotte. So Kyle will be popping in there, but it'll also be in New York City with the CBS Decision Desk as part of their new results-focused nerd show stream. CBS is doing that for the first time. I would imagine if you go on CBS News, their site, no matter where you are in the world, you'll be able to download that. Kyle, you're in demand. This makes me very happy. Well, I think for the first time on election night, I'll be able to play with one of those, you know, the board.

45:37Oh, nice. I was up in New York a couple of weeks ago, kind of practicing with it a little bit. If you tune in, you might see me just kind of fumble over, you know, trying to hit the button. Now you know it's not working. But I will do my best to keep my cool. And, you know, we'll be talking just about like, you know, what are the trends? What are the results? What do the candidates need to do? Where you should be watching? all that stuff, which is, uh, has always been what I've enjoyed and what's always in our kind of wheelhouse at the, uh, crystal ball. I love it. All right. Okay. This Kyle's going to get, he's going to get his Troy Aikman, Tony Romo, Chris Collinsworth, uh, telestrator on, which is going to be awesome.

46:17All right. Follow Kyle's work, obviously. Um, Kyle Kondik from the Sabato's crystal ball, the managing editor there, Kyle, thank you very much. Always, uh, always appreciate your work. All right. Stay here, Kyle. As a, as this mediocre host takes us out here. um again this conversation with kyle and i thank him for his time will obviously be very very different than what you're used to hearing on this podcast head to the archives there are a ton of sports media conversations we just did why cbs locked up jj watt and whether uh wmba broadcaster should be more critical of the product prior to that uh michael grady speaking of the timberwolves uh nba play-by-play broadcaster for them but now i'm much more known nationally on nba on nbc and NBA on Prime.

47:01Who else? Peter King was on this podcast not too long ago with his new book that was out. Obviously, if you want to go back to the archives, Troy Aikman was on this podcast in September. That started a bit of an avalanche when Troy said that ESPN had indeed not been negotiating with him on a new contract. You see where that story has gone from. Thanks for all the reviews and notes. Always appreciated very, very much. I want to thank James Jackson for all his hard work. Thanks to everybody at Odyssey seek their support. And thank you for listening. We'll see you soon on the Sports Media Podcast.