Carl Allen

I fully expect DMs, EMs, etc to blow up in the next 48 hrs If I don't respond, don't worry. I know who has supported my work and I won't lose sight of that If you'd like to join me in developing a set of unified scientific standards for this field, my book is available in bio

"Carl, have any positions to share? How confident are you?" Vegas flights are booked and victory speeches written

It's beautiful, isn't it? A pollster bragging about how closely their poll margin matched the election margin People like @NateSilver538 & @gelliottmorris complaining about herding Oblivious to the fact they are the ones who popularized the invalid metric promoting herding 👌

Comparing poll margin to election margin to measure "accuracy" is in itself unscientific Imposing this invalid metric onto today to see "what would happen" is the gambler's fallacy for poll nerds Snapshot down by 2 but won by 3! Off by 5! Here's what that means for this race..

If Democrats win statewide in places like Texas and Ohio (very possible) And especially places like Florida, Kansas, Iowa, South Carolina (far from implausible) Dems need to be clear about what their message is: A final, resounding rejection of MAGA.

Hey so is anyone going to talk about Trump's obsession with men or are we just going to pretend simultaneous homophobia and homoeroticism is Presidential?

Trump cares more about rich nepo baby far right losers than he cares about the average American. And also, it seems, this Bolsonaro dude

Prediction Markets are growing because if someone wants to get five or six figures of sharp money down, most sportsbooks are going to scrutinize it, and in the rare case they take it, move them closer to getting cut off Prediction Markets literally do not care

"once you see it you can't unsee it" The way "poll accuracy" is calculated isn't an insignificant metric. It makes and breaks reputations, gains or destroys public trust, and contributes to a basic understanding of a statistical tool. The current methods are invalid.

Reminder: despite the 19% undecided, the prevailing belief (and stated methodology) among poll analysts is that unless this race is extremely close, the poll was wrong.

Simpson is a below average MLB hitter whose elite contact rate & speed make him serviceable The fact that statistical illiterates use him (and/or Arraez) as their shining example of what Real Baseball looks like tells me they have no idea what it takes to score runs in baseball

"He was only 17% to win, then he won! Predict markets was wrong!"

Stat nerds: hitting the ball hard and in the air is good. Casuals: noooooooooo

5 events with probability 83% happen Silence 1 event with probability 17% happen *Uncontrollable screeching* I guess statistical illiteracy keeps me employed

Behold The Prediction Market Hater Paradox