From: allin
In the realm of electoral politics, both traditional polls and prediction markets play significant roles in forecasting outcomes, but their accuracy and methodologies are subject to debate [01:01:42].
Inaccuracy of Traditional Polls
During the election cycle, traditional polls, such as those from the New York Times and CNN, showed significant divergence from the eventual outcomes predicted by prediction markets [01:06:07]. For instance, the New York Times initially estimated the Electoral College at 297 for Trump and 241 for Harris, indicating an 83% chance of a Trump victory [02:27:09]. However, as results came in, disparities became apparent.
A key criticism of traditional pollsters is their perceived lack of “skin in the game” [01:22:12]. According to Shane from Polymarket, pollsters are “just some person trying to go and buy for attention” with “no consequence for being wrong” [01:22:05]. This contrasts sharply with prediction markets, where participants literally “put their money where their mouth is” [01:22:57].
Challenges with Polling Methodology
Several factors were suggested to explain why polls were “so far off” [01:01:56]:
- Undercounting certain demographics: The Amish turnout in Pennsylvania, for example, involved “miles and miles of Amish in their buggies all with their Trump banners” going to vote, but they were not showing up in pollster reports because “they don’t have phones” [01:02:00].
- Reluctance to disclose vote: Some voters may have been unwilling to publicly disclose their vote for certain candidates [01:02:45].
- Technological changes: The decline of landlines makes it difficult for pollsters to reach a representative sample [01:02:55].
- Youth turnout: The potential impact of young men’s votes, possibly influenced by figures like Elon Musk, was questioned [01:03:02].
A specific example of a questionable poll was the Selzer poll in Iowa, which predicted Kamala Harris would win Iowa less than 24 hours before the election, but Donald Trump won by 14% [01:50:25]. The pollster’s video interview was described as making her “out to be like a total quack” [02:02:15].
The Rise of Prediction Markets
Polymarket is highlighted as a significant and accurate prediction market platform, often calling elections months in advance [00:30:04]. It is described as a “funnest site to sweat these things” [00:30:09] and has facilitated “a couple hundred million dollars” in trading on election day, marking an “all-time high” for prediction markets [01:15:02].
How Prediction Markets Work
Polymarket’s founder, Shane, explains that the platform operates on a “peer-to-peer” model, not as a “house” [01:17:31]. It’s a venue where “the market interprets and ingest that information” to find equilibrium [01:53:46]. The market price for an outcome share is derived from the probability of that outcome, with each share paying out $1 if the outcome occurs [01:18:16].
Advantages of Prediction Markets
- Accuracy through “Skin in the Game”: Unlike pollsters, traders in prediction markets have financial incentives to be correct, leading to more accurate price discovery [01:18:30], [01:22:57]. The accuracy is attributed to people “putting their money where their mouth is” [01:22:57].
- Transparency: The platform is transparent, allowing users to see who is participating, their positions, and even their entry prices in comments [01:54:06].
- Real-time Information: Users can observe market price changes in real time as new information emerges [01:19:00]. This provides a “better way to understand what’s likely to happen” than traditional vote counting or pollster’s roundabout answers [01:21:26].
- Alternative News Source: Polymarket is described as an “alternative news source” or “information markets platform” that distills essential information, making it “intuitive and digestible” for people to understand likelihoods [01:19:05].
- Arbitrage Opportunities: While Polymarket aims to be the primary market, discrepancies with other less liquid venues can create arbitrage opportunities, allowing traders to profit from price differences [01:15:37].
Criticisms and Challenges of Prediction Markets
- Misinformation and Discrediting: The platform faced “criticism in the press” and “a lot of memes that were about discrediting the accuracy of the free market” [01:19:21]. This includes narratives about a “French whale” making large trades that allegedly skewed odds [01:16:31], and the platform’s use of stablecoins (USDC) leading to it being labeled “crypto” [01:30:19]. However, these are seen as “characteristics of a free market” [01:30:19].
- Legal Status for Americans: Polymarket is “not yet for Americans” [01:16:31] due to regulatory hurdles [01:46:26]. The company has to “navigate system and things like that” without the luxury of waiting or lobbying due to initial lack of capital [01:47:46]. However, there are ongoing efforts to legalize prediction markets in the US [01:48:19].
Other Market Indicators
Beyond dedicated prediction markets, other financial instruments are seen by some as indicators of political outcomes. The price of Bitcoin, Dogecoin, and the Mexican-US Dollar exchange ratio were observed to track Trump’s likelihood of winning the election [02:52:15]. These are considered “truest biggest markets” [01:03:39] and can move “way before the prediction market start moving” [01:37:37].
Impact and Future
The significant divergence between traditional polls and prediction markets on election night suggests a major shift in how information is consumed and trusted. The success of prediction markets is seen as a “10x plus improvement” over traditional polling [01:22:36], potentially signaling the end of reliance on the “Legacy pollster model” [01:50:10]. This shift highlights the growing influence of alternative media and data-driven platforms in shaping public understanding of political events.