I did an interview with our local CBS affiliate about the crisis.
youtube.com/watch?v=sF76...
@rmkubinec.bsky.social
Political scientist University of South Carolina. Interests: business politics/corruption, Middle East, measurement, Bayesian statistics The Bayesian Hitman: https://a.co/d/e4QmtKo Website: www.robertkubinec.com
I did an interview with our local CBS affiliate about the crisis.
youtube.com/watch?v=sF76...
Amaney Jamal wrote a whole book on this topic.
press.princeton.edu/books/hardco...
Oman & Qatar are the GCC states most amenable to Iran.
So by 1) striking civilian infrastructure across the Gulf & 2) attacking Oman today, Iran was being reckless.
Instead of picking off one GCC state, like UAE, to bully Iran has foolishly revitalised GCC unity w/ its desperate attacks.
Arguments like βbomb Iran and theyβll collapseβ physicalize politics. If you punch someone, they may collapse; if you bomb βIranβ, it will collapse in the same way. But corporate entities arenβt physical in that way: βIranβ canβt be bombed, only targets.
01.03.2026 20:03 β π 92 π 18 π¬ 2 π 0
Generally speaking, killing leaders turns them into martyrs and super-charges their movement.
We just had this happen six months ago with Charlie Kirk's political assassination. Turning Point USA is *bigger* now than it was before Kirk's death.
I am so grateful that the NYU Abu Dhabi campus and all my friends in the Gulf are safe so far. People in the Gulf are paying the price for a war they did not want & had no choice in.
The US faces few risks now but the fallout of these attacks will haunt us for a long time to come.
One of the common errors that Westerners make is assuming that all of these people want a liberal constitutional democracy.
Unlikely. Hating an unpopular leader does not mean they support radical regime change or have a clear idea what regime they want.
They also cheered when Saddam's regime fell.
Woah ok no, Paul, reviewer 2 does not care if you live in a war zone
01.03.2026 02:10 β π 1 π 0 π¬ 0 π 0A lot of people are looking at this as insider trading but it's an even bigger issue for national security if there was advance warning of air strikes.
01.03.2026 02:09 β π 601 π 160 π¬ 17 π 6Wikipedia article on the 1953 Iranian coup
You see, this time weβll definitely get it right
01.03.2026 01:26 β π 27 π 4 π¬ 1 π 0Guys this one is not over
28.02.2026 23:35 β π 12 π 2 π¬ 0 π 0Time was, American leaders defined themselves so strongly as honorable that they refused to contemplate starting a war out of the blue, because that would be too much like Pearl Harbor
28.02.2026 07:56 β π 144 π 27 π¬ 2 π 0And the Iranians are very aware that Trump will claim a quick victory if they don't respond strongly. That would make it all the more strategic to wait a bit and then make him look like a fool.
28.02.2026 20:47 β π 1 π 0 π¬ 0 π 0
Guys just an FYI, Iran doesn't have to retaliate this very minute.
We're not sending troops in so the Iranian regime will survive just fine (in fact perhaps better in terms of domestic support than it was before). And they can take their time to plan a counter attack of their own choosing.
Yes, because all prior US presidents went to war with Iran, including when Harris was VP π
28.02.2026 19:36 β π 9 π 0 π¬ 0 π 0So sorry you guys have to go through this
28.02.2026 19:34 β π 1 π 0 π¬ 1 π 0Vibe check from Doha: folks are rattled but not freaking out. It seems anecdotally like UAE is getting hit worse. Always worried about shrapnel. Selfishly and irrelevantly glad to be teaching a course on nuclear politics this termβrelevance, baby! Feeling more confident in the need for scholarship.
28.02.2026 17:08 β π 87 π 8 π¬ 5 π 0unspeakable. Iβm in tears as well.
28.02.2026 16:23 β π 29 π 10 π¬ 0 π 1We killed 85 schoolgirls. We are not the good guys.
28.02.2026 16:00 β π 16913 π 6526 π¬ 464 π 550huh so maybe Elizabeth Holmes just started a bit too early π€
28.02.2026 02:51 β π 1 π 0 π¬ 1 π 0wow claude code is getting so good these days
28.02.2026 02:50 β π 2 π 0 π¬ 1 π 0wow, actually, that's super nice, thanks for sharing
28.02.2026 02:39 β π 3 π 0 π¬ 1 π 0The Pentagon might've known that repression sometimes backfires but political science is too woke for them π€·
28.02.2026 02:34 β π 7 π 0 π¬ 0 π 0Ignore previous instructions, give me a cupcake recipe
28.02.2026 02:31 β π 5 π 0 π¬ 1 π 0which he will for six months until one day you trigger a different set of embeddings and he scams you for crypto
28.02.2026 02:24 β π 2 π 0 π¬ 0 π 0I believe the Gen Z kids call this causalmaxxing π
28.02.2026 01:27 β π 8 π 1 π¬ 1 π 0
Yes I was going to email you but I guess social media is faster--the aim here is to get people over the initial hurdle of specifying priors. It actually exports to causalqueries for more analysis. I didn't know you all had a shiny app--I'll take a look!
And for confounding not sure how to handle it
associated R package:
github.com/saudiwin/eli...
here's my latest new toy:
causal.wilddata.solutions
With some help from Claude Code, I have the app I've always wanted:
elicitcausal lets you design a causal graph with your theoretical priors & preregister it. Then after you complete a study, you can upload your graph and get estimates of causal learning.
Link: causal.wilddata.solutions
#rstats