English
Koleman Strumpf
Report

Koleman Strumpf

@KolemanStrumpf

Econ prof @WakeForest. Research interests = prediction mkts, digitization, public econ, media mkts, illicit mkts, ... co-organizer VIDE seminar @digitaleconorg

280 Following444 Followers819 Posts
  1. Koleman Strumpf@KolemanStrumpf
    Op-ed in @WSJ with @HarryDCrane . Polls, not prediction markets, are the bigger threat to election integrity. Extends and emphasizes earlier thread,
    Wall Street Journal Opinion @WSJopinion
    Regulate polls, not prediction markets. There’s no evidence that betting harms election integrity, but fake surveys have done so this year, write Harry Crane @HarryDCrane and @KolemanStrumpf
    35233.3K
  4. Koleman Strumpf@KolemanStrumpf
    Election integrity is important. That is why is time that Congress (and state congresses) turns it attention to regulating polls. Not prediction markets, polls. (1/n)
    4101595.9K
  5. Koleman Strumpf@KolemanStrumpf
    Posts @KolemanStrumpf
    If so-called experts use phony numbers and can be so easily tricked, this sounds like a tool that can be used maliciously to distort election. I say we need to start looking into this industry. Sen. Warren will you sponsor a bill? (/end)
    20175
  6. Koleman Strumpf@KolemanStrumpf
    Posts @KolemanStrumpf
    Sheesh the gift that never stops giving. I think perhaps there is a problem in poll-land. How many grifters can there be?
    Cooper Burton @cooperrburton
    NEW: @50plus1news has uncovered even more pollster shenanigans, this time from SoCal Strategies. After we requested comment from them, they too took down their website and X account. You can read our full investigation here:
    00018
  7. Cooper Burton✓@cooperrburton
    NEW: @50plus1news has uncovered even more pollster shenanigans, this time from SoCal Strategies. After we requested comment from them, they too took down their website and X account. You can read our full investigation here:
    1210932975.6K
  8. Koleman Strumpf@KolemanStrumpf
    Posts @KolemanStrumpf
    Forgot to add this great post from @MickBransfield. This is nothing new: polls have had rotten apples for decades.
    Mick Bransfield @MickBransfield
    This is great work, but the problem has existed for a long time. There are a lot of sketchy pollsters who do not show up in polling averages. The industry needs meaningful self-policing. And this doesn't address 'real' pollsters who don't know their job but can sell clients.
    00138
  9. Mick Bransfield✓@MickBransfield
    This is great work, but the problem has existed for a long time. There are a lot of sketchy pollsters who do not show up in polling averages. The industry needs meaningful self-policing. And this doesn't address 'real' pollsters who don't know their job but can sell clients.
    Download
    101477
  10. Koleman Strumpf@KolemanStrumpf
    New paper on political prediction markets with @HarryDCrane . 🧵below (1/n)
    Harry Crane @HarryDCrane
    New study on DEI in political prediction markets with @KolemanStrumpf Diversity, Efficiency, and Incentives in Political Prediction Markets posted at 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
    14131.3K
  11. Harry Crane✓@HarryDCrane
    New study on DEI in political prediction markets with @KolemanStrumpf Diversity, Efficiency, and Incentives in Political Prediction Markets posted at 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
    25243.4K
  12. Koleman Strumpf@KolemanStrumpf
    Posts @KolemanStrumpf
    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)
    20036
  13. Koleman Strumpf@KolemanStrumpf
    Posts @KolemanStrumpf
    Typo: should have been historical PM traders in 1900 were older!
    00010
  14. Koleman Strumpf@KolemanStrumpf
    Posts @KolemanStrumpf
    Please read the full paper (linked at the top) and send along notes and especially things for us to work on! (/end)
    00031
  15. Koleman Strumpf@KolemanStrumpf
    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.
    Koleman Strumpf @KolemanStrumpf
    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.
    44233.3K
  16. Koleman Strumpf@KolemanStrumpf
    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.
    2154.1K
  17. Koleman Strumpf@KolemanStrumpf
    For those of us in the south monitoring weather:
    002203
  18. Koleman Strumpf@KolemanStrumpf
    I guess the NBA reads my tweets 😂.
    Mick Bransfield @MickBransfield
    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
    004293
  19. Mick Bransfield✓@MickBransfield
    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
    21143.8K
  20. Show more posts