Econ prof @WakeForest. Research interests = prediction mkts, digitization, public econ, media mkts, illicit mkts, ... co-organizer VIDE seminar @digitaleconorg
Joined April 2022
- Tweets801
- Following272
- Followers427
- Likes2.4K
New paper on political prediction markets with @HarryDCrane .
arxiv.org/abs/2610.28277
papers.ssrn.com/sol3/papers.…
🧵below (1/n)
New study on DEI in political prediction markets with @KolemanStrumpf
Diversity, Efficiency, and Incentives in Political Prediction Markets posted at
papers.ssrn.com/sol3/papers.…
We studied every election trade made on Kalshi in 2024 and matched the traders to their politics.
Top Line: political prediction market traders are not representative of voters, and most traders are politically biased, but the market as a whole is not biased.
How? Most traders are politically biased, but they bet small. The big traders who are motivated by profit drive most of the volume and they are not politically biased.
More details:
- Traders who placed a single trade put 73% of their money on Harris.
- Kalshi's volume came from zip codes that voted just 28% for Trump, yet 60% of it went to Republican positions.
Which traders are politically biased:
- Traders who placed a single trade put 73% of their money on Harris and had the lowest returns of any group.
- The tie between a trader's home zip code and the side the trader took weakens as trading activity rises and is not distinguishable from zero among traders with more than 100 trades.
- Active traders (>100 trades) supplied 93% of the volume, and took the same positions in red and blue zip codes.
Link to full paper below
These markets differ in structure (historical markets non-anonymous: newspaper report names), scope, and type of skew (Kalshi traders older) but generally forecast well. Incentives not diversity is they key (6/n)
Please read the full paper (linked at the top) and send along notes and especially things for us to work on! (/end)
Prediction markets did not do too well with econ Nobel prize: Hall and Woodford not even options despite often mentioned as contenders. Similarly many misses with other prizes last week. As I tweeted last month, these markets are not great with closed room decisions.
Econ Nobel prize:
Still a bit early but figure should put this out here. Kalshi market is thin as last year. PMs do a bad job forecasting questions where info is sequstered from the public, eg papal selection. Wondering how well this will work this year.
kalshi.com/markets/kxnobelec…
For those of us in the south monitoring weather: wafflehousetracker.com/
I guess the NBA reads my tweets 😂.
xitter.cf/KolemanStrumpf/status/…
NBA seems to be diverging from NFL:
“Frankly, from at least the NBA standpoint, we have roughly 40 different jurisdictions regulating us right now. So to have a consistent policy nationally, I think would be in our best interest,” [Commissioner Adam Silver] said. @AnanyaBChetia
NFL is not happy with prediction markets and are asking for state regulation. I am not sure I follow the NFL's reasoning: what if some states permit the kinds of bets they oppose and others allow them, then what? (1/n)
nytimes.com/athletic/7671964…
Here in NC the state recognizes CFTC authority to regulate PMs (and also taxes them). Presumably a uniform set of rules would make more sense to address specific concerns. (2/n)
decrypt.co/373209/north-caro…
@HarryDCrane and @a_kane47 have many additional insights in their posts on this. (/end)
xitter.cf/HarryDCrane/status/210…
xitter.cf/HarryDCrane/status/210…
xitter.cf/a_kane47/status/210826…
Such a move would see the entire sports prediction market ecosystem; brokers, exchanges, clearinghouses, not to mention thousands of innovators, entrepreneurs, startups building ancillary services to this new industry be forced to move off shore or quit completely.
Instead, I’m confident further protections that the League is seeking can be addressed without annihilating the American entrepreneur.
The next VIDE seminar (10/7, 11 AM ET) will have Elliott Ash (elliottash.com/) presenting "COURTS OF TOMORROW: Evidence from a Nationwide Rollout of Generative AI" (elliottash.com/papers/Mehmoo…).
Signup=digitalecon.org
AI and job losses.
Familiar to NYC straphangers is subway ad for "AI law firm" Crosby which features billable hours on a gravestone. Goto their webpage... Lots of human lawyers.
crosby.ai/
Econ Nobel prize:
Still a bit early but figure should put this out here. Kalshi market is thin as last year. PMs do a bad job forecasting questions where info is sequstered from the public, eg papal selection. Wondering how well this will work this year.
kalshi.com/markets/kxnobelec…
OK I think I found it. I was about 50% correct which is not bad.
econjobrumors.com/topic/when…
EJMR had some skilled writers. My favorite post: combines incredible song (Daft Punk's Giorgio by Moroder) with high econ nerd humor. I believe it is called "Estimators of the Future" and I imagined it was read by Heckman. Thanks anonymous poaster! (text below).
It's back (and better than ever?). So many experiments to assess from its one month or so absence. @florianederer ?
reddit.com/r/academiceconomi…
econjobrumors.com/
In the beginning, I wanted to do a paper with the estimators of the '50s, the estimators of the '60s, of the '70s...
And then have an estimator of the future. [4/n]