If your first election was 2016, you experienced politics in a way no other generation has. The first woman nominee winning the popular vote and losing the election.
I vividly remember Donald Trump winning the election and waking up shocked the next day because I was told by everyone who I thought was way smarter than I was and in tune with everything that this was impossible. And I looked in articles left, right, and center telling me that Donald Trump could not win the election, that Hillary Clinton had a 91% chance of winning, that she had a 98% chance of winning, that she was a 90% bet to win Florida. And I started looking at the polls, and I was startled by how close it actually was. And I just remember thinking, on what basis were we so confident?
And so, it makes sense that if you are of this generation, one that trusts the media less, one that grew up with disinformation and uses the word facts as a compliment, one that has access to AI and more data in their pockets than any generation before, then you might take matters into your own hands.
But the problem is, when you're 23 coming out of college, you think you know everything.
And you don't. And you really don't.
On today's episode, we meet up with Lakshya Jain, the young CEO of Split Ticket, a small band of politicos bound by an obsession with elections and distrust of punditry. I'm Audie Cornish, this is The Assignment.
'We chose to take this episode to an arcade in Northern Virginia, Doyle's Outpost. It was empty, which, for anyone who's ever stood in a line with a sweaty fist full of tickets waiting for a chance at an unwinnable claw machine or Dance Dance Revolution, you know, is awesome. Lakshya Jain, who at 27-years-old still looks boyish in his blue polo and navy blazer, was down to play.
As a kid. My dad used to bring me to the arcade to have fun on Fridays, and I remember looking to see which place would yield the most tickets because that would allow me to get the best prizes over at the front desk. And I found that I was really good at skee ball.
Yeah. And I found that if I just kept playing stickball, I could get a lot of tickets really quickly. And so there were a whole bunch of prizes that I wanted at the front. I wanted the candy, I wanted the basketball hoop. And obviously, I didn't want to ask my parents to buy it. So instead, when I went to the arcade, I was like, Cool, I'm going to save money and just win a whole bunch of tickets and that'll get me the basketball hoop. It took me until I grew up to realize that going to the arcade probably cost more money than just buying the hoop myself.
But I like that you you you gathered a data set, right? You looked around the room and you decided, what am I most likely...
'That the return on investment in your 11-year-old mind.
That's basically what it was. And, you know, looking back on it, maybe it wasn't exactly the best conclusion, but the process was there.
Was a little bit like politics then, basically.
Almost exactly like politics, just with significantly lower consequences.
I wanted to talk to Lakshya because his team is behind an online game developed with the news site Politico called You Be the Campaign Manager. Basically, people can pretend they're helping their candidate go after the right mix of states or demographics and electoral College votes to win the presidential election. The game of it is what struck me like, what's the point of this thing that feels very high stakes reduced to something so low stakes? And why did it feel so much like the kind of work we as political journalists are asked to do all the time.
In common political terms, most of what you hear is like, let's talk about the Rust Belt states or let's talk about the Sun Belt states. But even in those regions, there are many different parts that behave very differently. I'll give you just one example. In Georgia, Metro Atlanta behaves very differently from the rest of the state, which is significantly more Republican. And Metro Atlanta swung very far to the left in 2020, which is what powered Joe Biden to victory, even as the rest of the state didn't have the same level of friendliness towards him. So our goal here was to show people that it's not states. That's the dividing line. It's really people and the more granular you get. So we did it at a county level, the better you can show them how these groups behave. What's one of the best ways to characterize the divide in America politically? It's urban and rural. People in cities broadly vote for Democrats. People in the rural areas much more frequently, and they vote for Republicans.
And so many times we actually slice and dice it through the demographics. Women, Men. Black, white. Hispanic. Education levels.
Exactly. But we don't pay as much attention to urbanization, I think, in the way that we should. We don't talk about the fact that if you're a Republican candidate and you're gaining even 5% with urban voters. Well, that's such a big group that you can actually offset losses in a whole bunch of other places simply because you're appealing to more people. The discourse in political talk, I think, has really shifted to the groups that had the largest swings in 2016. It's a lagging indicator of sorts.
So you're saying we're still focused on the same communities that helped win 2016?
I think so, personally, and I believe that you got to skate to where the puck is going, not where it's been. And the best way to understand how different trade offs can impact the election is to explore how different groups and different areas shifting different clusters, if I may say, would impact the election. And I here's one. So people talk about the Republicans being primarily a rural party. This isn't actually true. There's not enough rural voters in America to power Republicans to victory by themselves. What is true is that, yes, they get great margins in the rurals, but they actually do very well in the exurbs of the country, the places that border the suburbs and the rurales, the places that straddle the line. That's the power of the Republican Party. That's where they really get a lot of their votes. That's why they're so good in places like North Carolina. But you never hear that. You never hear the fact that that is the group that donates the most to Republicans also. Right? Compared to what you would expect compared to how much they're talked about. This is interesting to me because politics is ultimately at the state level. It's who gets more votes. So what that concretely means is a Republican getting 15% in San Francisco and gaining 5% from the previous election actually matters more than a Republican gaining 20 points in a rural county that has 3000 people.
When I hear you talk like this, I think there's probably someone out there who also says, well, what's the point of voting then? Like, basically, if the numbers tell this story, then what's the point?
Well, I think voting has two important things. Firstly...
But are they right? First of all, like when you basically say, look, if you live in this county, maybe it doesn't matter if you vote one way or another or if you don't live in any of these battleground states, in these battleground counties. The slimmer, the slimmer, the slimmer, a number of them, then like, why bother?
Well, I think it depends. I think there is a lot of granularities at which elections happen. There's a school board level...
Would you say that to such a person?
Okay. So tell me the truth. What...Because there's people your age who are very much checked out of this election. Right? In some ways, maybe not in your community. But I don't think having more information, more data, more maps that are red and blue at a smaller and smaller point is more encouraging. It feels more discouraging.
You're you're right that there is a very big sense of discouragement. A lot of people in Gen Z and all across the nation have because if you're a Republican voting in Illinois or in California, your vote, frankly, does not matter nearly as much as it would in Wisconsin or Pennsylvania when it comes to the president. And that is an undeniable fact of the reality that we live in a system determined by the Electoral College, where it's really seven core battleground states, maybe ten, if you squint. And it is true that living in California, my vote, frankly, rarely and barely matters for president.
Or we should say, does it have the impact? Still matters, but the impact is not the same.
Exactly. Which is an unfortunate reality. But elections happen at all levels of the ballot. There is House of Representatives, there is the state legislature. There is a school board.
It strikes me your generation would be way more acutely aware of these things than I was. So, like my the most impactful election I remember I was an intern in like 2000, right? That was an election that had its big contested moment. And that's when I learned about red and blue maps, right? That's when I got this wakeup call about what the Electoral College was versus the popular vote. And it really does shape you. Like meaning you have different expectations.
It absolutely does. And my generation, I think, is really tuned into this fact because in 2016 Hillary Clinton won the popular vote by two points and lost the election. And for a lot of people in my generation, my social group, there was a sense of what the heck? How did that happen? And the political engagement that the Trump era created among Democrats, among Republicans, even among independents. The spike in news coverage, the spike in attention that people paid was unlike anything that I had seen growing up. As a result, the most socially liberal group of the country, Gen Z, sees a president who is diametrically opposed to most of the values that they care about trans rights, gay rights, civil rights, police brutality, gun rights and gun control. Really. They see a president who stands in opposition to all of that.
Yeah. So you're saying there is a kind of the way I'm talking about 2000 as kind of a catalyzing effect. You think that had a catalyzing effect for people your age?
I think so, because they said we didn't want this and they don't think the country wanted it either, because Hillary Clinton won the popular vote. Now, that doesn't matter. We live in an electoral college system.
Now you all know what that is.
You're just you're so into it that you as an adult would end up creating a data set and model to make a game out of it.
Exactly. I think and the reason for that is. I realized that. It's not super fun to be lectured by people or be told by people how things are. It's much more fun to discover it for yourself.
Is it also about control? Like, it's sort of hard to feel like you can have impact. It's hard to feel like, you know, especially with the amount of news that's happening. It feels like it might give you guys a measure of control to be able to look at these numbers.
I think to the extent that there's any control associated, are we doing this because it makes us feel better, because we're more important, because we feel we have more influence. And the truth is I don't think we have nearly as much influence as people would believe. The vast majority of the country does not vote based on what political analysts say. But what is true is that. We also at the same time have a responsibility to correct things that we see are either wrong or are missing in coverage. When we see someone putting out data analysis to journalists and politicians, politicians who make policy. Journalists who cover news, we feel like if we have the skills, then we also have a responsibility to put out information that we think is unbiased and informative to help people better understand how to frame things and how to cover them.
More with Lakshya Jain in a moment.
So, I want you to tell me a little bit about who you are so I can understand how you got to this point, right? Where you would be doing this work as like some kind of fun side hustle. What was your, sort of, introduction to politics as a kid?
So, I come from a family where my father's an engineer and my mother did her master's degree as well. So, really, education was always just heavily, heavily emphasized. And the biggest principle taught to me was never assume you're smarter than you are. Never assume that you know all the answers and never assume that you can fill in the holes of missing data.
My household was very Democratic, and I lean Democratic as well, myself. But I try to make sure that this doesn't impact my work. This is actually partly why Split Ticket has registered Republicans and registered Democrats as well. But for me, what really influenced me was that because I was told to never really assume I knew all the answers, I started reading articles about the elections in 2018, 2019, 2020, and I was very surprised by how many people just projected their desires and their assumptions onto the data without responsibly characterizing it.
And how old were you then?
So you are just getting politically active, yourself, in a lot of ways. You're coming of age politically.
And you're doing it in this, in the time of Donald Trump, right? And after that very, like, then, later on, contested election in so many ways. And I wonder how that affected how you see the way we talk about politics?
Yes. I vividly remember Donald Trump winning the election and waking up shocked the next day because I was told by everyone who I thought was way smarter than I was and in tune with everything that this was impossible. And I looked at the polls...
Meaning you were reading data, or...
I was reading articles left, right and center telling me that Donald Trump could not win the election, that Hillary Clinton had a 91% chance of winning, that she had a 98% chance of winning, that she was a 90% bet to win Florida. And I started looking at the polls and I was startled by how close it actually was. And I just remember thinking, on what basis for being so confident? Fast forward to 2020, and I just remember because of that experience, I had a very hard time trusting the expert punditry. As a result, I decided maybe I'd try my hand at data analysis. But the problem is, when you're 23 coming out of college, you think you know everything.
And you don't. And you really don't. And I fell into the same trap that I had been warned about for so many years by my parents and by all my teachers. Never assume you're smarter than you are. And I thought Biden is going to win 413 electoral votes. This is not close. Every bit of data that I saw became something that would confirm my priors. And I didn't realize this until the election happened. And Biden squeaked out a victory by the skin of his teeth. And I remember thinking, well, you know, now I understand. Because if it was that easy to control your biases and look at everything in an objective manner, more people would be doing it. And that, to me, was like a pie in the face. It was the best possible thing that could have happened to me. And that's how I think I really started to think about, okay, make sure that whatever you do, you check your biases. You lead with data. Never assume, again, that you can fill in the gaps better than everyone else.
I want to ask you about the red map and blue map that we've all kind of grown up with, but some of us for longer than others. Can you remember the first time you really saw that map?
I think the first time I really saw the starkness of a red and blue map was 2016. And that's the first time it really stuck with me. 2012? Yeah. You see the election map but doesn't really register when you're 15. You're just like, cool. Obama won the election. In 2016, I remember thinking, how did I not see that so much of the country supports Donald Trump? And I saw the map and I was just struck by how much red there was and not just in a geographic sense, but in a in a sense of where people lived and how red certain areas were with respect to others and how blue my part of the country was. And I just remember thinking, well, we're all kind of in our own echo chambers and it's really easy to hear whatever you want. And that goes for the data as well.
Did that shape your, kind of, understanding of politics in that moment?
Yes, because of that, I never trust it when people say, well, in my community, I don't see a single person supporting Donald Trump. And I'm like, well, cool, But you live in Manhattan.
Just the fact that it's called Split Ticket kind of gives me a sense of what your sort of hopes and orientation are. What have you learned about political behavior that was unexpected?
I think voters are substantially stranger than most political analysts like to believe. And I think that analysts have a much poorer understanding of voters than what we like to believe. And stranger is not an insult, to be clear. I think that when I use the word stranger, it means that, to us, we have a harder time understanding what drives them, what drives their motivations and what drives their choices than what we would believe because it's not always a linear process. It's hard for people to understand why a conservative may vote for a liberal senator. It's hard for people to understand why someone who shows up in 20204 may not have voted in 2020. It's hard for people to understand why someone who votes in a midterm election for Democratic candidates up and down the ballot could potentially just set out 2024 or choose to back the Republican at the top of the ticket and Democrats down ballot. To me, the biggest thing that we've learned in all of this is that all of this is substantially harder with a lot more uncertainty than anyone covering it would have you believe.
That's pretty intense considering you kind of got into the business for a little more certainty.
I got into it thinking I knew everything and I walked out of it realizing I know nothing. But I think what I'm happy about there is that I'm okay not knowing a lot of things as long as I know that I don't know a lot of things, which prevents me from making a complete fool of myself.
Fair, Fair, Fair. I like that. This lesson that you've learned is about human nature, basically, that we're still human beings, and the data can't tell you everything, right, about how we behave, especially when we get in the ballot box. Are there some myths that you've busted or questions you still get from people about how we all vote?
Yeah, I think the biggest one is that candidate quality really matters. People are not so polarized that you can put up a skeleton and expect to earn their vote.
Okay. But we're in this moment, which I call the end of shame, where all kinds of political candidates go through major scandals and they don't resign, they don't step back and their campaigns struggle and they continue. That feels like the opposite of the lesson that you're learning.
But I would point out that a lot of those candidates lose or underperform. One of the things that we do at Split Ticket is this thing called wins above replacement. It's based off of the baseball concept, again, because people find that more engaging so they're willing to participate in the exercise and engage with the content more. But in wins above replacement, the key thing that we try to measure is how well does this candidate do against the other candidate? How well does this Republican do against that Democrat and how does that compare to how a generic pair of candidates would have done?
The skeleton you referenced earlier.
'Exactly. Well, the best example I can give you is Mark Kelly versus Blake Masters in Arizona. We found that Mark Kelly did about six points better than generic Democrat would have against a generic Republican. A lot of this is because voters who backed Republicans at the House level refused to vote for Blake Masters. They drew the line there and they said, absolutely not. I'm not voting for that guy. You know, his campaign went through many scandals and a lot of people in 2022 said it doesn't matter because they looked at Donald Trump and they said, well, look at Donald Trump, he's had so many scandals and they didn't move the needle. Why would they move the needle for Blake Masters? And what that fundamentally misunderstands, I think, is the image of candidates and how variable that is to people and how it's not always a 1-to-1 mapping. It's not always the case that what doesn't hurt Donald Trump will also have no impact on Blake Masters or J.D. Vance or Mark Robinson. The reason for that is because it comes down to, again, perception, trust, all of these things that are so hard to quantify. So you're right, a lot of politicians have scandals. You're also right. A lot of them carry on. And in some cases, when when in the past they wouldn't have. But there's still a critical sliver of swing voters who exist, who exercise their power and say, no, that person does not stand for me. I don't care if he's from the party that I typically vote for, I will not be voting for that candidate and I will vote for the other party's candidate instead.
What's the biggest lesson you've learned from this process? Because you still have a day job. You haven't quit it.
Yeah. I would say the biggest lesson that I've learned from this process is other than the insane amount that we don't know, I think the biggest lesson that I've learned is that it doesn't really matter what your data says if you're unable to explain it. If you cannot convey the message to readers, it doesn't matter what you're saying, because people only have a limited amount of time and they don't have the domain expertise, understandably, so that someone deep in the weeds may.
So I shouldn't be making fun of your charts, graphs and games.
You can make fun of that all you want. I make fun of it all the time.
But it sounds like it's something very real you have in mind in conveying it.
Well, I think the reason that we do it is because we want people to understand the message that we are driving. We want people to understand the uncertainty inherent in this. We want people to understand the tradeoffs here. And if they don't get that, then it doesn't matter if your models are right or wrong, it means we failed. And to me, what I try to tell people as a result is, guys, you're doing this so that you can inform the public. If they don't come away with the takeaway you wanted them to, then you can't really count it as a success. I say this because certain pollsters will say we had the best polls on record in 2022. Sure. But those same pollsters gave the wrong advice to their readers, to their clients, because they started filling in gaps where they didn't have the data and they didn't have the ability to fill them in and they gave the wrong message. This is the problem that I struggle with. It doesn't. It's hard to count your work as a success if people come away with it having inferred something completely different from what you intended.
Lakshya Jain, who is CEO of Split Ticket. Thank you for talking with us.
Thank you so very much for having me.
And The Assignment is a production of CNN Audio. This episode was produced by Lori Galarreta and Dan Bloom. Our senior producer is Matt Martinez. Dan Dzula is our technical director and the executive producer of CNN Audio is Steve Lickteig. Helping us on the road, our video team, Harlan Schmidt, Steve Williams, Chris Turner, Oliver Janney, and our video editor was Isaac Ewart. Special thanks, as always to Katie Hinman. And thank you for listening.
