This, to me, is Nate Silver at his best: explaining the math and probabilities behind the information in a cogent way, in plain English. Personally, while I dislike feeling anxiety all the time: I find knowing more about it soothing. Furthermore people will ask me about it and I feel informed enough to explain it, which I do not do as well as Nate but I try. I'd rather walk into the election understanding what could happen and then understand the results than be surprised. But that's just me, for others reading about things is anxiety inducing, and I get it.
Sadly Trump winning in 2016 turned our already high stakes elections into extremely tense and dangerous. If he wins again it will remain so, I keep praying we'll end up in a spot where this is no longer the case.
So here's something funny: even as a happy paying subscriber, I don't really look at the model all that often. I am obviously dimly aware of what it says (a toss-up that may lean a bit this way and that over the weeks), but I don't know what the actual numbers are right now--maybe I'll start looking at them in real depth in late October, but right now it's still late summer and the sun is still shining.
I let people like Nate--people who I think are smart and trustworthy and insightful--interpret the model and write excellent posts like this one, and I read those religiously. Honest-to-god punditry!
Yeah, I haven't looked much at the model in the last month. I was very interested when the model first came out (iirc, that was right before the Biden debate) and when the Harris model first came out, but for me personally, I know the race is probably going to be very close, and digging into the details just doesn't add that much value right now imo. Like you, I'll be more interested in late October
Agreed so much. I think it’s ironic that some people will post online “Nate is good as a modeler but bad as a pundit.” I think it’s reversed. Clearly his model is decent. But with more and more imitation clones out there it would be hard to know for sure whose was more accurate after the election. And for my purposes, does it make a difference if he forecasts 70/30 or 30/70? No. But as a commentator he excels mostly because, I think, he is better at math than most pundits. And I suspect his model’s strengths are more to do with him having good judgement about the subject matter than superiority in data science perse. Really it’s baffling to me that someone could have a good model at all if their SME was garbage. The one caveat is that his model might benefit from the most robust training set as well.
That's a little further than I'd take it, but I'll say this: I think you can't be a great election modeler without being a competent pundit, and I think you can't be even a decent pundit without understanding election modeling on at least a whiteboard level (I don't care if you actually know how to code it).
Likewise. I look at four or five aggregators, Nate included (obviously). And I mostly draw my conclusions about how I think the vote would go if it were held today based on the spread of those aggregators. I ignore the heck of out individual polls. I probably will look more often at the the prediction Nate's model yields once we get to the midway point in October. But for now, my sense is there's a lot of time to go, and it's still a ways in the future (which is always hard to predict).
Two exceptions for "ignore the heck out of individual polls" are NYT/Sienna and Anne Selzer! I sit up and read everything about those, especially after their consistent recent successes.
Trump is MASA: Make Americans Slaves Again. DEMOCRACY is on the ballot and we cannot lose. We cannot afford to fail. The world is counting on us to show the way.
That's verbatim what Ryan Routh, the latest Trump attempted assassin, wrote in his Tweets. According to a recent Rasmussen poll, 28% of Democrats surveyed believe that the nation would be better off if he was killed. TDS is the real danger to democracy.
If someone says they want to set up concentration camps, become a dictator, and patrol the street with the military to suppress dissent - I think being afraid or "deranged" about such promises/goals is reasonable.
And if we want to muddy the water even further, the "Flight 93 Emergency" comparisons really started in 2016, so it might be better if you dismounted your high horse and cooled down, too.
Assume you know this, but TDS is Trump Derangement Syndrome. MAGAs refer to Trump haters as TDS because, for some strange reason, TDS people fail to see the sense in Trump's nonsense.
Yeah, I'm one of those who finds checking the forecast daily to be soothing. If the election is a rollercoaster ride, and if you believe (like I do) that we've roughly a 50% chance to jump the track, then I'd rather keep my eyes looking forward during the ride. Feels like the least I can do.
Hang in there, Nate. The vast majority of your subscribers understand probabilities; maybe we should comment more often to that effect, to give a truer picture than the self-selected ravings of the extremists on either side.
Reviewers and commenters tend to be those with strong enough opinions to bother writing their thoughts and thus the extremes, but... I REALLY like that Nate's out here and doing this.
Maybe that's a strong enough sentiment to just dash out a first-thoughts comment on his posts moving forward?
My first thoughts are usually on the order of, "Man, Nate, good points! Thanks for clarifying that thing! Engaging writing as usual! Good luck with the [insert event you told us about]!"
It's definitely a tricky thing in online forums. I might agree with everything Nate has to say but not have much to expand on. I don't want to see a hundred top-level comments all saying "This!".
Hard to align efficient use of forum space with the psychological tendency to evaluate the proportion of pro and anti comments as the actual proportion of pro and anti readers.
Wow - I read your posts, Nate, not the comments, so I had no idea you were taking this much flak. Don't blame you for needing to blow off som steam, but... just keep a level head and keep it up! You're doing a fantastic job, and most of your readers probably agree - or why would they read you, anyway? I remain a huge fan.
Liberal Twitter has been awful to Nate. The absurd thing is, if he REALLY were in the tank for Trump, he'd surely want to skew the polls to make it look like Trump were badly losing, right? Overconfidence among Democrats is what you WANT if you hope Trump prevails.
I'm a Harris supporter, and I'm tickled pink the country's most prominent poll analyst *isn't* pulling his punches about the very real possibility that she could lose. Because she could! So Democrats need to get to the polls.
This. People like to blame the messenger - but that doesn't change reality. It's the nonsensical idea that if you believe something hard enough it'll happen that way - ultimately a superstition that people don't realize they hold to (or embrace and just believe they aren't thinking something hard enough). It's best for people to see reality as it is and do something about it, not complain that the person doing their best to quantify reality isn't showing you what you want to see.
It's unfortunate that so few people understand probability and statistics. This even shows up in material from people who should know better, like the difference between a rate of change and a change in an absolute level, like the number of people think that because the inflation rate is decreasing means that prices are decreasing—which of course it does not.
As a consultant, I’ve generally come to the conclusion that people are either idiots or willfully ignorant when asked to correctly interpret data and trends. Folks dont deal with ambiguity well and often simply are overwhelmed when trends and numbers, no matter how simple, are put in front of them.
Just look at how many people will use numbers completely out of context to justify their argument; numbers which have no baseline for comparison. It’s the old guy complaining that coffee used to be 10 cents a cup, completely ignoring that he was paid $2 an hour at the time.
There are certainly idiots and willfully ignorant people out there - but intelligence is also really varied. My wife speaks three languages, can accurately diagnose a number of physical and mental problems simply by watching someone walk, and can name every muscle in the body and how it impacts a person's movement - but doesn't understand data, trends, statistics, or much complex math. There are many varieties of intelligence, and many really valuable ones that don't overlap with math. That's a big problem in a society that's increasingly math driven - which is something we need to address both with education and better explanation when talking to people.
But we should also remember that it was only a couple hundred years ago when the vast majority of humanity didn't need to know much beyond the practical aspects of farming. Those were very complicated in a practical, hands-on sense and required an innate understanding of the uncertainties involving weather, but didn't remotely require the math we have today. People are adapting and society is still slowly adapting - we don't shift as quickly as people often assume.
Exactly. Calling people who don't understand statistics/probability "idiots" is over-simplifying and insulting. From my experience, it clearly is something that many people honestly have trouble understanding, even people who are intelligent in other ways. Maybe they weren't taught it well as a child, or maybe it just needs to be explained well as an adult, or maybe their brains will always struggle to understand it. In any case, it doesn't mean that they're idiots.
This post is why I subscribe. I'm an engineer, so I have a background in statistics and like seeing the information the model provides knowing its uses and limits. It's interesting to peer in to what's going on and see how it changes heading towards Election Day. But it is incredibly common to see people treat statistics as "fact" rather than "likelihood" and ignore error bars and forget that low probability events do happen - so it's definitely something to remind people of. Most probably will ignore it - but, like innumeracy in general, you keep trying and reminding people that numbers don't always mean what they think they mean. Not providing the data doesn't mean people won't be wrong - just that they'll be wrong with lower quality data and without the correctives you try to provide.
Yep. For me, Nate has had 2 big prediction wins, and one was in 2016 shortly before the election when he made 4 or 5 different posts like today's where he emphasized that Trump could easily win despite the model probability being way below 50%.
If you had a 1 in 5 chance of getting hit by a bus if you crossed the road, no one would cross roads - damn the probability, it's a horrible idea. But if Trump has a 1 in 5 chance of winning an election, people seem to think that he can't possibly win. Many of them despite being willing to bet (cash) on longer odds on all manner of things. Especially the lottery.
Many never forgave him for Hillary losing. Or for those happy Trump won he was also wrong for precisely the reason he’s writing about today. Trying to convince someone that a person with a 90% chance of winning leaves you wide open for criticism. That’s why I enjoy articles like today so much. I only come to read good news about Harris but stick around to read every comment.
I live in Chicago & feel it's fantastic to double are options for entertainment---> when you miss out on tickets to a Chicago concert stop, you can always try Milwaukee.
Well ... I lived in Milwaukee for many years. Contrary to Nate's prediction, I'm not going to unsubscribe. I also lived in Chicago for many years. Nate is correct, it's a bit of pain to visit Milwaukee from Chicago, and the marginal value in the visit isn't that great because there are many similarities. But when living in Milwaukee, I found the same was true about visiting Chicago. Sure, Chicago has, say, bigger and better museums, but the lines are longer and the Uber ride across town and the hotel are a lot more expensive in Chicago. It all kind of balances out -- with minimal net upside either way. And, as Nate mentions, the decay time for the memory is about a year before one starts planning the next trip.
However, some months back Nate mentioned his affinity for Kansas City. He got this right too. I grew up in the KC area and still visit often, traveling both from Chicago and Milwaukee. KC has an excellent and varied food scene (with the best Que in the world!) good sports scene, new airport, great museums. And the lines aren't that long, and Uber rides and hotels are affordable. No memory cleansing is required before planning the next trip.
At some point I would like Nate to do a write up about 538’s new, new model (the one they relaunched with Harris). This model has consistently put Harris’s chances much higher. For instance, right now it has the race outside of the toss up range - 64-36 Harris. I want to know what inputs are causing such a difference, especially after they acknowledged their first model was flawed.
I would as well. I will note that Silver’s projection is colored (according to him, written in previous posts) by secondary indicators that he calls the “fundamentals,” such as the state of the economy (unemployment, interest rates, and so forth.) Silver has said that these will fall away as the election nears, making the Nov. 5 projection entry dependent on the polls. At present, he has also written, the fundamentals favor Trump. As a result his model’s forecast is bullish on Trump. My guess is that the 538 forecast is predicated entirely on polling and not colored by “fundamentals“ or less colored by them.
Agree with this. I'd also add that 538 put out a "What if" tool. Obviously I know Nate + Eli don't have the time/resources to put a tool together, but an occasional comparison of conditional probabilities would be interesting
Glad to be part of this (generally) well-behaved group, and sorry to hear about flak on Twitter, Nate- but honestly, do you expect anything better there? Twitter is designed for those kinds of comments. Looking forward to your analysis as to why the polls feel all over the place lately. Nate Cohn’s take on why Trump and Harris are tied nationally and Harris is up 4 in PA didn’t quite clarify things for me.
There's a bunch of it on this app too. It's mostly just people not understanding how probabilistic modeling works (e.g. Nate stinks because he picked this race wrong and I got it right!!111), but it is a little sad how many people here (people who's work I generally enjoy reading!) have talked themselves into "Nate bad" since he's not selling hopium.
I love the Serenity Prayer quote at the end. This prayer has gotten me through a lot of anxious situations. In fact, one of the best personal benefits of Trump winning (yes, benefit), was forcing me to look for a philosophy of life that would help me through dark times. That philosophy was stoicism. Yeah, I’m a stoicism bro. But hear me out, it has really helped me to focus on what I control. I don’t control the polls or the outcome of the election. But I do control my actions and some of my thoughts. I can be a good father, husband, friend and worker. Politically, I can control who I donate money to, whether I phone bank or write postcards. I do control how I talk to friends and family, using my tiny platform or circle of influence to help the promote the candidate or issue I care about.
One weird benefit that happened to me after the Biden-Trump debate was more serenity. I was devastated. I felt that same sinking feeling of dread that I felt after TFG got elected in 2016. So, I disconnected from reading and watching the news. I stopped X. I didn’t read my political blogs (sorry Nate). I just hung out in the present. I focused on what I could change. I enjoyed nature, my family and friends and worked on my health. I read good books. I cooked. Though the sinking feeling remained, it shrunk in size and power. I felt a lot more at peace. This taught me to be elective about how I consumed news and information. Anyway, life is wild. A lot of times, it’s hard and cruel. But it’s also beautiful and amazing.
So, if you are anxious about democracy, the climate, reproductive rights or the death of science, do what you can. Look at your circle of influence. Worry less and do more.
And I hope the universe or God can grant you the serenity to accept the things you cannot change, courage to change the things you can, and wisdom to know the difference.
Thank you! A repeat of another four-year nightmare of Trump and the horrific collateral damage which will ensue - because the second time around will be so much worse - doesn't bear thinking about. I, too, have been worrying about how I will get through that mentally - again. Came to the conclusion, in my own way as you have, that I best concentrate on my own little world - the things I can control - and let go of the things I can't. Thank you.
Nate, you and I could not be much less alike. I'm old, you are not. I'm a Republican, you are not. I could go on, but that is probably enough to set the tone. I have followed your predictions since your FiveThirtyEight days and I was delighted to be an early subscriber to the Silver Bulletin. I will go to the polls and vote on Election Day with complete confidence that my vote doesn't matter a damn. (I live in Connecticut. At least it isn't Milwaukee.) But, I will also assume that your final "guess" on the election will be right, despite knowing full well that it is only a sophisticated analysis of the unknowable. And, if you are "wrong," I will continue to follow your predictive work as long as you keep doing it. Hopefully four years from now your scars will have healed. I for one, am a very happy subscriber and I still will be a happy subscriber on November 6th. And, finally, I loved your note.
This, to me, is Nate Silver at his best: explaining the math and probabilities behind the information in a cogent way, in plain English. Personally, while I dislike feeling anxiety all the time: I find knowing more about it soothing. Furthermore people will ask me about it and I feel informed enough to explain it, which I do not do as well as Nate but I try. I'd rather walk into the election understanding what could happen and then understand the results than be surprised. But that's just me, for others reading about things is anxiety inducing, and I get it.
Sadly Trump winning in 2016 turned our already high stakes elections into extremely tense and dangerous. If he wins again it will remain so, I keep praying we'll end up in a spot where this is no longer the case.
So here's something funny: even as a happy paying subscriber, I don't really look at the model all that often. I am obviously dimly aware of what it says (a toss-up that may lean a bit this way and that over the weeks), but I don't know what the actual numbers are right now--maybe I'll start looking at them in real depth in late October, but right now it's still late summer and the sun is still shining.
I let people like Nate--people who I think are smart and trustworthy and insightful--interpret the model and write excellent posts like this one, and I read those religiously. Honest-to-god punditry!
Yeah, I haven't looked much at the model in the last month. I was very interested when the model first came out (iirc, that was right before the Biden debate) and when the Harris model first came out, but for me personally, I know the race is probably going to be very close, and digging into the details just doesn't add that much value right now imo. Like you, I'll be more interested in late October
Agreed so much. I think it’s ironic that some people will post online “Nate is good as a modeler but bad as a pundit.” I think it’s reversed. Clearly his model is decent. But with more and more imitation clones out there it would be hard to know for sure whose was more accurate after the election. And for my purposes, does it make a difference if he forecasts 70/30 or 30/70? No. But as a commentator he excels mostly because, I think, he is better at math than most pundits. And I suspect his model’s strengths are more to do with him having good judgement about the subject matter than superiority in data science perse. Really it’s baffling to me that someone could have a good model at all if their SME was garbage. The one caveat is that his model might benefit from the most robust training set as well.
That's a little further than I'd take it, but I'll say this: I think you can't be a great election modeler without being a competent pundit, and I think you can't be even a decent pundit without understanding election modeling on at least a whiteboard level (I don't care if you actually know how to code it).
>don't really look at the model all that often.<
Likewise. I look at four or five aggregators, Nate included (obviously). And I mostly draw my conclusions about how I think the vote would go if it were held today based on the spread of those aggregators. I ignore the heck of out individual polls. I probably will look more often at the the prediction Nate's model yields once we get to the midway point in October. But for now, my sense is there's a lot of time to go, and it's still a ways in the future (which is always hard to predict).
Two exceptions for "ignore the heck out of individual polls" are NYT/Sienna and Anne Selzer! I sit up and read everything about those, especially after their consistent recent successes.
I do the same. Nate Silver, 538, decision desk, The economist, and the real clear politics average of six betting markets.
Once a day I update my spreadsheet and the average of all five I take to be the most accurate proxy for the current state of the race.
54.3% probability Harris wins.
Trump is MASA: Make Americans Slaves Again. DEMOCRACY is on the ballot and we cannot lose. We cannot afford to fail. The world is counting on us to show the way.
It's probably giving him too much credit to assume he actually read the post.
That's verbatim what Ryan Routh, the latest Trump attempted assassin, wrote in his Tweets. According to a recent Rasmussen poll, 28% of Democrats surveyed believe that the nation would be better off if he was killed. TDS is the real danger to democracy.
More than one “real danger” can exist at the same time!
If someone says they want to set up concentration camps, become a dictator, and patrol the street with the military to suppress dissent - I think being afraid or "deranged" about such promises/goals is reasonable.
And if we want to muddy the water even further, the "Flight 93 Emergency" comparisons really started in 2016, so it might be better if you dismounted your high horse and cooled down, too.
if I knew what TDS was, I'd research them and - depending on what I found - either join forces with you to push back against them, or ignore you.
Assume you know this, but TDS is Trump Derangement Syndrome. MAGAs refer to Trump haters as TDS because, for some strange reason, TDS people fail to see the sense in Trump's nonsense.
Trump Deification Syndrome.
It ties closely with psychological projection, where MAGA people accuse other people of having the thought processes common among MAGA people.
Technical Data Sheet <3
I thought it was Torpedo Defense System. Been reading too much naval architecture
Yeah, I'm one of those who finds checking the forecast daily to be soothing. If the election is a rollercoaster ride, and if you believe (like I do) that we've roughly a 50% chance to jump the track, then I'd rather keep my eyes looking forward during the ride. Feels like the least I can do.
I agree and very well said!
I will let randomness fool me if I want to. You’re not the boss of me. Vikings 2-0 baby!
Hang in there, Nate. The vast majority of your subscribers understand probabilities; maybe we should comment more often to that effect, to give a truer picture than the self-selected ravings of the extremists on either side.
I'm kinda starting to think this as well.
Reviewers and commenters tend to be those with strong enough opinions to bother writing their thoughts and thus the extremes, but... I REALLY like that Nate's out here and doing this.
Maybe that's a strong enough sentiment to just dash out a first-thoughts comment on his posts moving forward?
My first thoughts are usually on the order of, "Man, Nate, good points! Thanks for clarifying that thing! Engaging writing as usual! Good luck with the [insert event you told us about]!"
Worth a try? Maybe see ya next post?
I always assumed clicking the "Like" button on the post expressed this sentiment.
It's definitely a tricky thing in online forums. I might agree with everything Nate has to say but not have much to expand on. I don't want to see a hundred top-level comments all saying "This!".
Hard to align efficient use of forum space with the psychological tendency to evaluate the proportion of pro and anti comments as the actual proportion of pro and anti readers.
I support this and will strive to walk the walk
Let's try it out!
Oh look, it's Sisyphus again, rolling a big ball of electionwatchers up the hill of statistical/probability literacy. Good luck this time buddy!
Wow - I read your posts, Nate, not the comments, so I had no idea you were taking this much flak. Don't blame you for needing to blow off som steam, but... just keep a level head and keep it up! You're doing a fantastic job, and most of your readers probably agree - or why would they read you, anyway? I remain a huge fan.
Liberal Twitter has been awful to Nate. The absurd thing is, if he REALLY were in the tank for Trump, he'd surely want to skew the polls to make it look like Trump were badly losing, right? Overconfidence among Democrats is what you WANT if you hope Trump prevails.
I'm a Harris supporter, and I'm tickled pink the country's most prominent poll analyst *isn't* pulling his punches about the very real possibility that she could lose. Because she could! So Democrats need to get to the polls.
I remember some guy once saying "don't boo, vote".
This. People like to blame the messenger - but that doesn't change reality. It's the nonsensical idea that if you believe something hard enough it'll happen that way - ultimately a superstition that people don't realize they hold to (or embrace and just believe they aren't thinking something hard enough). It's best for people to see reality as it is and do something about it, not complain that the person doing their best to quantify reality isn't showing you what you want to see.
If wishes were horses, then beggars would ride.
Actually I subscribe to read Nate but mostly to read the comments. Fascinating insight to how & what others are thinking.
It's unfortunate that so few people understand probability and statistics. This even shows up in material from people who should know better, like the difference between a rate of change and a change in an absolute level, like the number of people think that because the inflation rate is decreasing means that prices are decreasing—which of course it does not.
As a consultant, I’ve generally come to the conclusion that people are either idiots or willfully ignorant when asked to correctly interpret data and trends. Folks dont deal with ambiguity well and often simply are overwhelmed when trends and numbers, no matter how simple, are put in front of them.
Just look at how many people will use numbers completely out of context to justify their argument; numbers which have no baseline for comparison. It’s the old guy complaining that coffee used to be 10 cents a cup, completely ignoring that he was paid $2 an hour at the time.
There are certainly idiots and willfully ignorant people out there - but intelligence is also really varied. My wife speaks three languages, can accurately diagnose a number of physical and mental problems simply by watching someone walk, and can name every muscle in the body and how it impacts a person's movement - but doesn't understand data, trends, statistics, or much complex math. There are many varieties of intelligence, and many really valuable ones that don't overlap with math. That's a big problem in a society that's increasingly math driven - which is something we need to address both with education and better explanation when talking to people.
But we should also remember that it was only a couple hundred years ago when the vast majority of humanity didn't need to know much beyond the practical aspects of farming. Those were very complicated in a practical, hands-on sense and required an innate understanding of the uncertainties involving weather, but didn't remotely require the math we have today. People are adapting and society is still slowly adapting - we don't shift as quickly as people often assume.
Exactly. Calling people who don't understand statistics/probability "idiots" is over-simplifying and insulting. From my experience, it clearly is something that many people honestly have trouble understanding, even people who are intelligent in other ways. Maybe they weren't taught it well as a child, or maybe it just needs to be explained well as an adult, or maybe their brains will always struggle to understand it. In any case, it doesn't mean that they're idiots.
This post is why I subscribe. I'm an engineer, so I have a background in statistics and like seeing the information the model provides knowing its uses and limits. It's interesting to peer in to what's going on and see how it changes heading towards Election Day. But it is incredibly common to see people treat statistics as "fact" rather than "likelihood" and ignore error bars and forget that low probability events do happen - so it's definitely something to remind people of. Most probably will ignore it - but, like innumeracy in general, you keep trying and reminding people that numbers don't always mean what they think they mean. Not providing the data doesn't mean people won't be wrong - just that they'll be wrong with lower quality data and without the correctives you try to provide.
Yep. For me, Nate has had 2 big prediction wins, and one was in 2016 shortly before the election when he made 4 or 5 different posts like today's where he emphasized that Trump could easily win despite the model probability being way below 50%.
If you had a 1 in 5 chance of getting hit by a bus if you crossed the road, no one would cross roads - damn the probability, it's a horrible idea. But if Trump has a 1 in 5 chance of winning an election, people seem to think that he can't possibly win. Many of them despite being willing to bet (cash) on longer odds on all manner of things. Especially the lottery.
Many never forgave him for Hillary losing. Or for those happy Trump won he was also wrong for precisely the reason he’s writing about today. Trying to convince someone that a person with a 90% chance of winning leaves you wide open for criticism. That’s why I enjoy articles like today so much. I only come to read good news about Harris but stick around to read every comment.
Exactly. Instead of being happy they'd hit a longshot, Trump voters in 2016 were all "I find your lack of faith disturbing." #RIPJamesEarlJones
His snake chart was right on the money in 2016.
Funny: I live in Milwaukee and feel the same way about Chicago.
I live in Chicago & feel it's fantastic to double are options for entertainment---> when you miss out on tickets to a Chicago concert stop, you can always try Milwaukee.
Well ... I lived in Milwaukee for many years. Contrary to Nate's prediction, I'm not going to unsubscribe. I also lived in Chicago for many years. Nate is correct, it's a bit of pain to visit Milwaukee from Chicago, and the marginal value in the visit isn't that great because there are many similarities. But when living in Milwaukee, I found the same was true about visiting Chicago. Sure, Chicago has, say, bigger and better museums, but the lines are longer and the Uber ride across town and the hotel are a lot more expensive in Chicago. It all kind of balances out -- with minimal net upside either way. And, as Nate mentions, the decay time for the memory is about a year before one starts planning the next trip.
However, some months back Nate mentioned his affinity for Kansas City. He got this right too. I grew up in the KC area and still visit often, traveling both from Chicago and Milwaukee. KC has an excellent and varied food scene (with the best Que in the world!) good sports scene, new airport, great museums. And the lines aren't that long, and Uber rides and hotels are affordable. No memory cleansing is required before planning the next trip.
At some point I would like Nate to do a write up about 538’s new, new model (the one they relaunched with Harris). This model has consistently put Harris’s chances much higher. For instance, right now it has the race outside of the toss up range - 64-36 Harris. I want to know what inputs are causing such a difference, especially after they acknowledged their first model was flawed.
I would as well. I will note that Silver’s projection is colored (according to him, written in previous posts) by secondary indicators that he calls the “fundamentals,” such as the state of the economy (unemployment, interest rates, and so forth.) Silver has said that these will fall away as the election nears, making the Nov. 5 projection entry dependent on the polls. At present, he has also written, the fundamentals favor Trump. As a result his model’s forecast is bullish on Trump. My guess is that the 538 forecast is predicated entirely on polling and not colored by “fundamentals“ or less colored by them.
Agree with this. I'd also add that 538 put out a "What if" tool. Obviously I know Nate + Eli don't have the time/resources to put a tool together, but an occasional comparison of conditional probabilities would be interesting
Glad to be part of this (generally) well-behaved group, and sorry to hear about flak on Twitter, Nate- but honestly, do you expect anything better there? Twitter is designed for those kinds of comments. Looking forward to your analysis as to why the polls feel all over the place lately. Nate Cohn’s take on why Trump and Harris are tied nationally and Harris is up 4 in PA didn’t quite clarify things for me.
There's a bunch of it on this app too. It's mostly just people not understanding how probabilistic modeling works (e.g. Nate stinks because he picked this race wrong and I got it right!!111), but it is a little sad how many people here (people who's work I generally enjoy reading!) have talked themselves into "Nate bad" since he's not selling hopium.
I love the Serenity Prayer quote at the end. This prayer has gotten me through a lot of anxious situations. In fact, one of the best personal benefits of Trump winning (yes, benefit), was forcing me to look for a philosophy of life that would help me through dark times. That philosophy was stoicism. Yeah, I’m a stoicism bro. But hear me out, it has really helped me to focus on what I control. I don’t control the polls or the outcome of the election. But I do control my actions and some of my thoughts. I can be a good father, husband, friend and worker. Politically, I can control who I donate money to, whether I phone bank or write postcards. I do control how I talk to friends and family, using my tiny platform or circle of influence to help the promote the candidate or issue I care about.
One weird benefit that happened to me after the Biden-Trump debate was more serenity. I was devastated. I felt that same sinking feeling of dread that I felt after TFG got elected in 2016. So, I disconnected from reading and watching the news. I stopped X. I didn’t read my political blogs (sorry Nate). I just hung out in the present. I focused on what I could change. I enjoyed nature, my family and friends and worked on my health. I read good books. I cooked. Though the sinking feeling remained, it shrunk in size and power. I felt a lot more at peace. This taught me to be elective about how I consumed news and information. Anyway, life is wild. A lot of times, it’s hard and cruel. But it’s also beautiful and amazing.
So, if you are anxious about democracy, the climate, reproductive rights or the death of science, do what you can. Look at your circle of influence. Worry less and do more.
And I hope the universe or God can grant you the serenity to accept the things you cannot change, courage to change the things you can, and wisdom to know the difference.
Thank you! A repeat of another four-year nightmare of Trump and the horrific collateral damage which will ensue - because the second time around will be so much worse - doesn't bear thinking about. I, too, have been worrying about how I will get through that mentally - again. Came to the conclusion, in my own way as you have, that I best concentrate on my own little world - the things I can control - and let go of the things I can't. Thank you.
In reality you could probably bail on X completely, lose 5% of your income but be 100% happier.
Best comment on here. Nate should estimate the actual numbers, or maybe just do it.
Nate, you and I could not be much less alike. I'm old, you are not. I'm a Republican, you are not. I could go on, but that is probably enough to set the tone. I have followed your predictions since your FiveThirtyEight days and I was delighted to be an early subscriber to the Silver Bulletin. I will go to the polls and vote on Election Day with complete confidence that my vote doesn't matter a damn. (I live in Connecticut. At least it isn't Milwaukee.) But, I will also assume that your final "guess" on the election will be right, despite knowing full well that it is only a sophisticated analysis of the unknowable. And, if you are "wrong," I will continue to follow your predictive work as long as you keep doing it. Hopefully four years from now your scars will have healed. I for one, am a very happy subscriber and I still will be a happy subscriber on November 6th. And, finally, I loved your note.
Okay, but who’s gonna win?
I prefer living in reality. The future is uncertain. Thanks for your efforts to quantify that uncertainty.
I love this. My brain is summarizing it as, "stop making me your certainty king, take me off the pedestal." Well said, Nate.