Comment Text:
Dear CFTC,
I know there has been a large focus on hedging. I’d like to associate myself with the very smart people who have explained clearly why markets on congressional control are a great and necessary opportunity to hedge.
I would like to focus my comment on the greatest value of electoral prediction markets, which is that they incentivize expertise in a field currently filled by people who don’t have an incentive to be correct.
Here is some background on me. I’ve been actively trading in prediction markets since August 2020. What started out as a hobby has become my profession. Because of prediction markets, I can name all 100 senators. I can tell you every cabinet secretary in the Biden administration and every deputy cabinet secretary. I can give you a story on how each one of them was confirmed. I can tell you the story behind every major and minor piece of legislation that occurred in the Biden administration. I can give you a play-by-play of every somewhat competitive statewide primary and give you a current standing of every somewhat competitive statewide general election. I humbly would call myself a self-created expert 1) because I find this stuff fascinating and 2) because I was incentivized to become one.
What follows are several election stories that I believe all significantly emphasize the value of allowing incentivized traders to process data in real time, use this analysis to bet their own money to predict an outcome, and provide the public with true, up to the minute, probabilities on which side will win an election. In 2018, the Democrats trailed in several house seats in California after election night. This included seats eventually won by Gil Cisneros, Katie Porter, and TJ Cox. I wasn’t trading then, but if I recall correctly, the experts were not exactly on top of the fact that these seats were going to be won by Democrats until after several days of vote counting. There is no doubt in my mind that prediction markets were ahead of the pundits. This also was undoubtedly the case in the 2018 Arizona Senate race. Kyrsten Sinema trailed Martha McSally by a significant amount of votes, with only mail left to count. The narrative coming out election night was that McSally was probably favored (WaPo election guy Henry Olsen called it for her mistakenly), but Sinema would likely pick up some votes in the mail-in ballots, but there were also late drop-off votes that were probably favorable to McSally. It is this exact situation where you want prediction markets to exist. There was a major senate race up for grabs and the experts you should want the public to listen to are those who are personally incentivized to acquire each piece of data immediately and figure out where they should put their money. Kyrsten Sinema ended up winning comfortably and there was no batch of mail-in or drop-off ballots that were favorable to Martha McSally.
In 2020, on election day, the markets favored Democrats in Wisconsin, Pennsylvania, and Michigan, while they favored Republicans in Florida, Arizona, North Carolina, and Georgia. All of these states were priced to be close, which is not what the votes said. In Florida, the markets started to drift to Democrats as some Election Day partisan voting data started to come in. These were not actual election results, but just showed that Democrats were showing up to vote. Then, we started to get actual election results. Early voting results were dumped first. Counties started to come in showing a somewhat close race, until prediction market traders got a significant piece of data: Miami Dade County was awful for Democrats. Florida was easily going to go for Trump. CNN didn’t call the race until 4 hours later.
Continuing the 2020 trend, the markets took this piece of data from Florida and began to consider Trump to be a heavy favorite. The initial votes from Michigan and Pennsylvania were election day votes that showed Trump winning. It was clear Trump was going to win in North Carolina. Texas was worse than expected for Democrats including a massive shift in the Rio Grande Valley to Republicans. Georgia was interesting in that the New York Times election needle showed Trump to be favored, but in reality, we didn’t have much data from Atlanta. Still, taking the picture of data, the odds had shifted towards Trump. Brit Hume mentioned this several times on Fox News, and the Fox News round table was starting to taunt Donna Brazile after she said “there was a path to Joe Biden winning.”
Later in the evening, Arizona began to release election results. The results showed Biden and Mark Kelly winning in the all important Maricopa County. Soon after, Fox News called Arizona for Joe Biden to the shock of their own broadcast and honestly the country. The general election odds immediately took a hard turn towards Joe Biden, which again, Brit Hume reluctantly shared multiple times on their own broadcast. As it turns out, this call was borderline ridiculous and the prediction market traders were right to be hesitant at first, but the important reaction for the markets was to shift the general election odds with the information that Biden had probably won Arizona.
Later in the evening, at around 3:30 AM, the election results site Decision Desk HQ had somehow acquired the election results for Milwaukee in Wisconsin before any other news outlet had them. These results showed that Biden was clearly going to win Wisconsin, and the election odds immediately surged towards him.
Finally, as Pennsylvania, Georgia, and Michigan continued to count their votes over the next 24 hours, it became immediately clear that Biden would win the election, despite the media not calling it for 4 more days.
I’m attempting to paint a picture with all of these stories. It’s not that pundits are dumb and prediction market traders are always right. It’s that the public deserves a real-time accounting for what is going to happen. Most people don’t have time to know what each margin in each county means for the outcome of their election. To be honest, I assume most people don’t know who their congressman is. There’s nothing wrong with that, but that’s the world we live in. The best counter to that is to allow people who know what every precinct in every state means for the final election result process information quickly, bet their own money on what that information means, and provide real time odds for the public that the public knows are likely to be an unbiased, strong estimate, of where each race currently stands.
If you deny this opportunity, you will be denying the public a great service that can’t be replaced by pundits. At a time where pollsters are literally terrified of polling Wisconsin because they have been wrong so many times, the public needs these experts in these markets more than ever. Additionally, if you deny this opportunity, you should be honest with yourself that the prediction markets will not go away. They will just continue to be unregulated in which traders will use their VPNs and cryptocurrencies to trade on their elections and legislation in real time. The only difference will be that Brit Hume on Fox News will be less likely to quote the prediction market odds because either he doesn’t know these unregulated markets exist or he doesn’t feel comfortable sharing that information.
Finally, the risks here are minimal. Unless the Supreme Court is planning on overturning Citizens United soon (which isn’t happening), you will not get rid of money in politics. Both parties are aggressively spending money in politics. Mitch McConnell’s Super PAC will clear 100 million dollars spent in 2022 alone easily. The Democrats have spent a significant amount of money boosting candidates that they believe are a threat to democracy in order to give themselves a slightly better chance at winning a few House seats. There is no world in which an election market with $25,000 limits are a risk to election integrity with this current landscape of the amount spent in politics. Respectfully, you are being extremely naive to pretend that’s the case. In fact, the opposite is true. Many people spending their own money to predict an outcome is a hedge in favor of election integrity. If a scenario that a liquid market was pricing at 1% or less happened, that would be something worth looking into to see if anything untoward happened. On the flip side, if the 90% scenario happened and the 10% side cried foul, that would be a data point in favor of a clean election. These markets are a good thing for election integrity.
Therefore, it is in the CFTC’s interest and especially the public’s interest to approve these congressional control markets and to allow electoral prediction markets in general.
Sincerely,
Jonathan Zubkoff
