Will May
@wmay.bsky.social
Social sciences/statistics/programming, and other stuff. Work with weather researchers
That socially left flank of the Democratic Party has waged a years-long campaign of harassment and intimidation against people who disagree with them, smearing critics as transphobes and trying to get them fired or shunned.
So yeah, we're fixated on how awful they are now. 1/2
So yeah, we're fixated on how awful they are now. 1/2
And yet elite mainstream discourse seems to be hyperfixated on how the left flank of the Democratic Party alienates people
Not the point of @gelliottmorris.com piece, but look at this divide among Republicans on social issues. Half of them look like normal people and the other half are way out of step with the rest of the country. There’s a bit of a mirror on the left but much smaller.
November 21, 2025 at 5:03 AM
That socially left flank of the Democratic Party has waged a years-long campaign of harassment and intimidation against people who disagree with them, smearing critics as transphobes and trying to get them fired or shunned.
So yeah, we're fixated on how awful they are now. 1/2
So yeah, we're fixated on how awful they are now. 1/2
Bluesky has been terrible for political science's reputation, due to the way it encourages ideological slop.
I instinctively defend political science because I like a lot of it, and I've dabbled in it as a hobby for years. Now even I'm wondering if the field has a big problem with political bias
I instinctively defend political science because I like a lot of it, and I've dabbled in it as a hobby for years. Now even I'm wondering if the field has a big problem with political bias
"Actually the Democratic Party can't moderate because Gen Z is racially progressive"
Has anyone pointed out that this argument makes no sense? The anti-moderation crowd is getting weird and desperate with this stuff.
Also "low racial resentment" is not the same as "racially progressive". 1/2
Has anyone pointed out that this argument makes no sense? The anti-moderation crowd is getting weird and desperate with this stuff.
Also "low racial resentment" is not the same as "racially progressive". 1/2
Those claiming Dems should retreat on racial justice aren't hard-headed realists, they're pushing against the electoral tide rather than leaning into it. The story of Gen Z isn't about racist backlash or red-pilled young men. It's the most racially progressive generation in American history. 🧵
November 17, 2025 at 6:42 AM
Bluesky has been terrible for political science's reputation, due to the way it encourages ideological slop.
I instinctively defend political science because I like a lot of it, and I've dabbled in it as a hobby for years. Now even I'm wondering if the field has a big problem with political bias
I instinctively defend political science because I like a lot of it, and I've dabbled in it as a hobby for years. Now even I'm wondering if the field has a big problem with political bias
"Actually the Democratic Party can't moderate because Gen Z is racially progressive"
Has anyone pointed out that this argument makes no sense? The anti-moderation crowd is getting weird and desperate with this stuff.
Also "low racial resentment" is not the same as "racially progressive". 1/2
Has anyone pointed out that this argument makes no sense? The anti-moderation crowd is getting weird and desperate with this stuff.
Also "low racial resentment" is not the same as "racially progressive". 1/2
Those claiming Dems should retreat on racial justice aren't hard-headed realists, they're pushing against the electoral tide rather than leaning into it. The story of Gen Z isn't about racist backlash or red-pilled young men. It's the most racially progressive generation in American history. 🧵
November 17, 2025 at 6:04 AM
"Actually the Democratic Party can't moderate because Gen Z is racially progressive"
Has anyone pointed out that this argument makes no sense? The anti-moderation crowd is getting weird and desperate with this stuff.
Also "low racial resentment" is not the same as "racially progressive". 1/2
Has anyone pointed out that this argument makes no sense? The anti-moderation crowd is getting weird and desperate with this stuff.
Also "low racial resentment" is not the same as "racially progressive". 1/2
A weird coincidence about this "punching-left" attempt at a gotcha on moderates, is that it's happening as conservatives have also been debating "no enemies to the right" on social media.
The extremists on both sides want us to give them a free pass.
No. Bad ideas deserve criticism
The extremists on both sides want us to give them a free pass.
No. Bad ideas deserve criticism
1/2 One of the big problems for the always-punching-left Democratic consultants is that they are waging ideological battles in public that actually _reinforce_ the left-wing image of the party that they are trying to fight. That is a risk they have decided
substack.com/@lakshyajain...
substack.com/@lakshyajain...
November 14, 2025 at 4:22 AM
A weird coincidence about this "punching-left" attempt at a gotcha on moderates, is that it's happening as conservatives have also been debating "no enemies to the right" on social media.
The extremists on both sides want us to give them a free pass.
No. Bad ideas deserve criticism
The extremists on both sides want us to give them a free pass.
No. Bad ideas deserve criticism
This kind of stupidly partisan analysis from Elliot has gradually shifted me from "more political data analysis, great!" to "ABC was right to close 538 and they should have fired this guy so much earlier". 1/
1/2 One of the big problems for the always-punching-left Democratic consultants is that they are waging ideological battles in public that actually _reinforce_ the left-wing image of the party that they are trying to fight. That is a risk they have decided
substack.com/@lakshyajain...
substack.com/@lakshyajain...
November 13, 2025 at 8:49 PM
This kind of stupidly partisan analysis from Elliot has gradually shifted me from "more political data analysis, great!" to "ABC was right to close 538 and they should have fired this guy so much earlier". 1/
I'm increasingly frustrated with @gelliottmorris.com. I think he's irresponsible and bad for Democrats.
Here he is dismissing attempts to win the senate as "shifting the goal posts", as if the Senate plays some negligible role in US politics. He seems to want Dems to give up the Senate. 1/
Here he is dismissing attempts to win the senate as "shifting the goal posts", as if the Senate plays some negligible role in US politics. He seems to want Dems to give up the Senate. 1/
I have noticed how the goal posts have shifted significantly in this debate from the Matt et al. point of view. First, it was winning the presidency, then winning "the median voter," and now winning the Senate majority. The sleight of hand here is to pick an increasingly hard goal for Dems...
November 8, 2025 at 1:51 AM
I'm increasingly frustrated with @gelliottmorris.com. I think he's irresponsible and bad for Democrats.
Here he is dismissing attempts to win the senate as "shifting the goal posts", as if the Senate plays some negligible role in US politics. He seems to want Dems to give up the Senate. 1/
Here he is dismissing attempts to win the senate as "shifting the goal posts", as if the Senate plays some negligible role in US politics. He seems to want Dems to give up the Senate. 1/
Reposted by Will May
"There are many more examples of academic studies being retracted or condemned for reasons unrelated to merit, credibility, integrity, or validity. And unfortunately, these cases are just the tip of the iceberg."
Who Should You Trust? Why Appeals to Scientific Consensus Are Often Uncompelling
The public is frequently told to “trust the science,” and then ridiculed for holding any views that differ from what is reported to be the scientific consensus. Should non-experts then naively accept ...
www.skeptic.com
November 3, 2025 at 10:59 PM
"There are many more examples of academic studies being retracted or condemned for reasons unrelated to merit, credibility, integrity, or validity. And unfortunately, these cases are just the tip of the iceberg."
All the bickering aside, those of us looking at the stats of political moderation (@jakemgrumbach.bsky.social @adambonica.bsky.social @gelliottmorris.com) seem to have mostly agreed that a large dose of moderation can cause an increase of .5-1.5% of vote share (increasing the lead by 1-3%). 1/
October 31, 2025 at 6:32 AM
All the bickering aside, those of us looking at the stats of political moderation (@jakemgrumbach.bsky.social @adambonica.bsky.social @gelliottmorris.com) seem to have mostly agreed that a large dose of moderation can cause an increase of .5-1.5% of vote share (increasing the lead by 1-3%). 1/
Nope.
Your and Lakshya's approaches lead to consistent results regarding moderation, *when the analysis is done correctly*.
I showed you this two months ago. 1/ rpubs.com/wmay/war-mod...
Your and Lakshya's approaches lead to consistent results regarding moderation, *when the analysis is done correctly*.
I showed you this two months ago. 1/ rpubs.com/wmay/war-mod...
October 31, 2025 at 4:09 AM
Nope.
Your and Lakshya's approaches lead to consistent results regarding moderation, *when the analysis is done correctly*.
I showed you this two months ago. 1/ rpubs.com/wmay/war-mod...
Your and Lakshya's approaches lead to consistent results regarding moderation, *when the analysis is done correctly*.
I showed you this two months ago. 1/ rpubs.com/wmay/war-mod...
This is just the fallacy of composition, by the way. I'm used to it appearing in economics, this is the first I've seen it show up in poli sci
en.wikipedia.org/wiki/Fallacy...
en.wikipedia.org/wiki/Fallacy...
October 30, 2025 at 12:17 AM
This is just the fallacy of composition, by the way. I'm used to it appearing in economics, this is the first I've seen it show up in poli sci
en.wikipedia.org/wiki/Fallacy...
en.wikipedia.org/wiki/Fallacy...
Repeating another common false claim about moderation.
Nationalization of politics does not imply moderation doesn't matter.
The "party brand" is itself a function of Dem legislators and candidates. Democrats have agency and can change the brand. Nominating moderates likely affects the brand 1/
Nationalization of politics does not imply moderation doesn't matter.
The "party brand" is itself a function of Dem legislators and candidates. Democrats have agency and can change the brand. Nominating moderates likely affects the brand 1/
"Does moderation matter? In our current two-party system? Not really. Presidential vote explains 98% of House outcomes. Candidate quality has collapsed to statistical noise."
open.substack.com/pub/leedrutm...
open.substack.com/pub/leedrutm...
October 30, 2025 at 12:06 AM
Repeating another common false claim about moderation.
Nationalization of politics does not imply moderation doesn't matter.
The "party brand" is itself a function of Dem legislators and candidates. Democrats have agency and can change the brand. Nominating moderates likely affects the brand 1/
Nationalization of politics does not imply moderation doesn't matter.
The "party brand" is itself a function of Dem legislators and candidates. Democrats have agency and can change the brand. Nominating moderates likely affects the brand 1/
Here we go again with the moderation.
Yes, Bonica points out real flaws with the NYT analysis.
But Bonica conveniently omits that his *own data* and his *own code* shows an electoral benefit to moderation. 1/
Yes, Bonica points out real flaws with the NYT analysis.
But Bonica conveniently omits that his *own data* and his *own code* shows an electoral benefit to moderation. 1/
Look. It would be great if there was one simple trick for winning elections. But 'just be more moderate' isn't it.
In fact, you can use the NYT's exact method to 'prove' a 'Progressive Advantage' of +1.4 pts.
This piece shows what's really going on: funded candidates do better than unfunded ones.
In fact, you can use the NYT's exact method to 'prove' a 'Progressive Advantage' of +1.4 pts.
This piece shows what's really going on: funded candidates do better than unfunded ones.
The New York Times’ “Moderation Advantage” Is a Statistical Illusion
After accounting for money and incumbency the supposed electoral bonus for moderate candidates vanishes entirely.
open.substack.com
October 27, 2025 at 12:59 AM
Here we go again with the moderation.
Yes, Bonica points out real flaws with the NYT analysis.
But Bonica conveniently omits that his *own data* and his *own code* shows an electoral benefit to moderation. 1/
Yes, Bonica points out real flaws with the NYT analysis.
But Bonica conveniently omits that his *own data* and his *own code* shows an electoral benefit to moderation. 1/
This is moronic. It's so embarrassing that a portion of Democrats think males in women's powerlifting is "human rights".
Democrats need to convince voters they are normal, sane people, and supporting this makes it difficult. 1/3
www.cbsnews.com/minnesota/ne...
Democrats need to convince voters they are normal, sane people, and supporting this makes it difficult. 1/3
www.cbsnews.com/minnesota/ne...
Minnesota Supreme Court rules USA Powerlifting discriminated against trans athlete
Advocates on Wednesday celebrated a Minnesota Supreme Court ruling that found USA Powerlifting violated the state's Human Rights Act in barring a transgender woman from competition.
www.cbsnews.com
October 23, 2025 at 4:21 AM
This is moronic. It's so embarrassing that a portion of Democrats think males in women's powerlifting is "human rights".
Democrats need to convince voters they are normal, sane people, and supporting this makes it difficult. 1/3
www.cbsnews.com/minnesota/ne...
Democrats need to convince voters they are normal, sane people, and supporting this makes it difficult. 1/3
www.cbsnews.com/minnesota/ne...
Reality:
- for high performance computing, you need to know slurm
- for systems engineering, well, no one knows what that is anyway
- for compilers, why are you writing a compiler? Are you crazy? Use any existing language
1/
- for high performance computing, you need to know slurm
- for systems engineering, well, no one knows what that is anyway
- for compilers, why are you writing a compiler? Are you crazy? Use any existing language
1/
October 3, 2025 at 12:34 AM
Reality:
- for high performance computing, you need to know slurm
- for systems engineering, well, no one knows what that is anyway
- for compilers, why are you writing a compiler? Are you crazy? Use any existing language
1/
- for high performance computing, you need to know slurm
- for systems engineering, well, no one knows what that is anyway
- for compilers, why are you writing a compiler? Are you crazy? Use any existing language
1/
It's going to be a total disaster next time I have to build my own logic gates from scratch at work. Ticking time bomb tbh
this comment just has me wondering what else you don't know
October 3, 2025 at 12:05 AM
It's going to be a total disaster next time I have to build my own logic gates from scratch at work. Ticking time bomb tbh
Reposted by Will May
The "there's more to election outcomes than issue-positioning" vs "issue-positioning doesn't matter to election outcomes" motte and bailey is not very helpful at addressing any concrete choices.
September 29, 2025 at 4:37 PM
The "there's more to election outcomes than issue-positioning" vs "issue-positioning doesn't matter to election outcomes" motte and bailey is not very helpful at addressing any concrete choices.
I'm absolutely baffled by the description of this issue as a "moral panic". I've seen zero panic at all, from anyone.
Moral panic is when you think the Satanist cults are sexually abusing children. It's not just someone interpreting evidence differently than you 1/
Moral panic is when you think the Satanist cults are sexually abusing children. It's not just someone interpreting evidence differently than you 1/
Recently the Economist Magazine claimed a study in India provided a “ringing endorsement” of cellphone bans in schools. But did the study in question do any such thing? Or was this yet another example of the kind of sloppy journalism that typifies moral panics?
open.substack.com/pub/grimoire...
open.substack.com/pub/grimoire...
Did a Large Study Really Provide a "Ringing Endorsement" of School Cellphone Bans?
How The Economist completely screwed this up.
open.substack.com
September 27, 2025 at 11:12 PM
I'm absolutely baffled by the description of this issue as a "moral panic". I've seen zero panic at all, from anyone.
Moral panic is when you think the Satanist cults are sexually abusing children. It's not just someone interpreting evidence differently than you 1/
Moral panic is when you think the Satanist cults are sexually abusing children. It's not just someone interpreting evidence differently than you 1/
I still have tons of doubt about this WAR model. I don't see how this model can be identifiable.
It basically isn't, which is why the uncertainty intervals are absurdly huge. Does anyone in their right mind actually believe Susan Collins' WAR could be *negative* 6?
1/
It basically isn't, which is why the uncertainty intervals are absurdly huge. Does anyone in their right mind actually believe Susan Collins' WAR could be *negative* 6?
1/
It's very gratifying to write with a focus on statistical rigor and earnestly attempting to get the methodology right while being able to open-source the codebase. If you dig this sort of thing, you should subscribe to Elliott's substack --- multiple publications per week!
www.gelliottmorris.com
www.gelliottmorris.com
September 18, 2025 at 4:44 AM
I still have tons of doubt about this WAR model. I don't see how this model can be identifiable.
It basically isn't, which is why the uncertainty intervals are absurdly huge. Does anyone in their right mind actually believe Susan Collins' WAR could be *negative* 6?
1/
It basically isn't, which is why the uncertainty intervals are absurdly huge. Does anyone in their right mind actually believe Susan Collins' WAR could be *negative* 6?
1/
I'm staging an intervention into the WAR debate because everyone is wrong. Seriously!
Wrong about WAR, wrong about moderation, the whole thing.
Many results are driven purely by omitted variable bias:
rpubs.com/wmay/war-mod... 1/
Wrong about WAR, wrong about moderation, the whole thing.
Many results are driven purely by omitted variable bias:
rpubs.com/wmay/war-mod... 1/
RPubs - Everyone's wrong about WAR and moderation
rpubs.com
August 24, 2025 at 12:04 AM
I'm staging an intervention into the WAR debate because everyone is wrong. Seriously!
Wrong about WAR, wrong about moderation, the whole thing.
Many results are driven purely by omitted variable bias:
rpubs.com/wmay/war-mod... 1/
Wrong about WAR, wrong about moderation, the whole thing.
Many results are driven purely by omitted variable bias:
rpubs.com/wmay/war-mod... 1/
It does seem very awkward to have a debate from behind a paywall. I'm not sure how people are expecting this to work
www.gelliottmorris.com/p/data-over-...
www.gelliottmorris.com/p/data-over-...
August 20, 2025 at 2:30 AM
It does seem very awkward to have a debate from behind a paywall. I'm not sure how people are expecting this to work
www.gelliottmorris.com/p/data-over-...
www.gelliottmorris.com/p/data-over-...
@gelliottmorris.com: "I'm not using dw nominate"
*looks at the analysis code*
*it uses a score mostly derived from DW-NOMINATE*
Are you kidding me guys? You don't even know what's in your own analysis?? 1/
*looks at the analysis code*
*it uses a score mostly derived from DW-NOMINATE*
Are you kidding me guys? You don't even know what's in your own analysis?? 1/
i'm not using dw nominate for this analysis, will
August 16, 2025 at 9:06 PM
@gelliottmorris.com: "I'm not using dw nominate"
*looks at the analysis code*
*it uses a score mostly derived from DW-NOMINATE*
Are you kidding me guys? You don't even know what's in your own analysis?? 1/
*looks at the analysis code*
*it uses a score mostly derived from DW-NOMINATE*
Are you kidding me guys? You don't even know what's in your own analysis?? 1/
Catching up here on parts of the moderation debate I haven't been following.
I have this funny reaction where I'm inclined to agree with the methodology criticisms from all sides. Yes, you're all wrong
(except the Blue Rose people, they do great work) 1/
data4democracy.substack.com/p/do-moderat...
I have this funny reaction where I'm inclined to agree with the methodology criticisms from all sides. Yes, you're all wrong
(except the Blue Rose people, they do great work) 1/
data4democracy.substack.com/p/do-moderat...
Do Moderates Do Better?
Uncovering Bias in Split Ticket’s WAR Scores
data4democracy.substack.com
August 15, 2025 at 4:10 AM
Catching up here on parts of the moderation debate I haven't been following.
I have this funny reaction where I'm inclined to agree with the methodology criticisms from all sides. Yes, you're all wrong
(except the Blue Rose people, they do great work) 1/
data4democracy.substack.com/p/do-moderat...
I have this funny reaction where I'm inclined to agree with the methodology criticisms from all sides. Yes, you're all wrong
(except the Blue Rose people, they do great work) 1/
data4democracy.substack.com/p/do-moderat...
And I didn't even address the most complicated part, the ideology scores themselves!
It took me years to learn how DW-NOMINATE works. The scores alone are already very complicated, statistically, with a bunch of difficult nuances. This is a very difficult subject to analyze accurately
It took me years to learn how DW-NOMINATE works. The scores alone are already very complicated, statistically, with a bunch of difficult nuances. This is a very difficult subject to analyze accurately
This is a complicated issue. If you approach it as if there's a straightforward answer you're going to get from running a regression, you're just doing everyone a disservice.
Embrace the complexity, politics is complicated sometimes 13/13
Embrace the complexity, politics is complicated sometimes 13/13
August 14, 2025 at 2:26 AM
And I didn't even address the most complicated part, the ideology scores themselves!
It took me years to learn how DW-NOMINATE works. The scores alone are already very complicated, statistically, with a bunch of difficult nuances. This is a very difficult subject to analyze accurately
It took me years to learn how DW-NOMINATE works. The scores alone are already very complicated, statistically, with a bunch of difficult nuances. This is a very difficult subject to analyze accurately
Haven't read the paywalled portion of this, but I already have some issues to be frustrated about (that don't depend on the unread parts).
Let's start with the claim that looking at caucuses is "cherry-picking" 1/
Let's start with the claim that looking at caucuses is "cherry-picking" 1/
I put a lot of work into this. I would be so happy if you would read and share:
Why moving to the center won't save democrats:
www.gelliottmorris.com/p/moderation...
Why moving to the center won't save democrats:
www.gelliottmorris.com/p/moderation...
Moderation is overrated
In 2024, moderate candidates for the U.S. House barely outperformed the average partisan
www.gelliottmorris.com
August 14, 2025 at 2:10 AM
Haven't read the paywalled portion of this, but I already have some issues to be frustrated about (that don't depend on the unread parts).
Let's start with the claim that looking at caucuses is "cherry-picking" 1/
Let's start with the claim that looking at caucuses is "cherry-picking" 1/
Very cool to see Carol Tavris appear here. I bought her book a few years back.
She sees some parallels between the 1980s Satanic panic and current gender medicine, and suggests we shouldn't necessarily trust doctors or therapists to read the research correctly
www.broadview.news/p/when-sleep...
She sees some parallels between the 1980s Satanic panic and current gender medicine, and suggests we shouldn't necessarily trust doctors or therapists to read the research correctly
www.broadview.news/p/when-sleep...
When Sleepless Nights Are Called For: What Will, and Should, Happen to the Most Ardent Gender-Affirming Clinicians?
Carol Tavris on self-justification, cognitive dissonance, and who will and won't come around.
www.broadview.news
July 3, 2025 at 2:36 AM
Very cool to see Carol Tavris appear here. I bought her book a few years back.
She sees some parallels between the 1980s Satanic panic and current gender medicine, and suggests we shouldn't necessarily trust doctors or therapists to read the research correctly
www.broadview.news/p/when-sleep...
She sees some parallels between the 1980s Satanic panic and current gender medicine, and suggests we shouldn't necessarily trust doctors or therapists to read the research correctly
www.broadview.news/p/when-sleep...