Below is a list of tips and insights that I use when deciding how to allocate my bets. I will update this list on an ongoing basis.
Look for Low Volume Markets: Buy up shares in markets where there isn’t much volume. In the absence of many choices, you may be able to sell your shares for a higher profit than what you’d find in a tighter, high-volume market.
Example
“Will Will Hurd be re-elected to Congress in 2018?” I bought no at 55% on the assumption that he is a vulnerable Republican incumbent whose polling would get worse as the midterms approached. Because it was such a low volume market, however, I was able to sell off most of my shares at 98% to a desperate buyer who apparently could not get better prices. The market price eventually recalibrated and gradually plummeted as Hurd’s poll numbers held up through October.
Look for Markets that Attract Highly Ideological/Partisan Investors: A comprehensive study on Intrade data found that 37 percent of volume comes from 7 percent of traders who are “strongly biased.” That percentage can be markedly higher in emotionally-charged markets, creating profitable deals from gamblers who cannot separate their wishes and fears from their betting strategy.
Example
In the French presidential election markets, Marine Le Pen was consistently overpriced, trading at 35 points even when she was down by 20 points in the polls. The cause was almost certainly probably due to a combination of wishful thinking by right-wing, populist traders as well as irrational fear on the part of those who missed the Trump-Hillary outcome—sentiments prevalent in mainstream op-ed pages.
Take Advantage of Panics: Even though everyone knows that surprises are part of the game, they never cease to promote irrational swings in the markets that create profitable opportunities. Keeping a level head in these situations can yield quick, high returns.
Example
After Trump’s victory in the November 2016 elections, the odds of Janet Yellin remaining as Federal Reserve chair plummeted before stabilizing again a few days later. Although I had predicted a Hillary victory, I was able to make up a great deal of ground in this market alone.
Research Creatively: Look for information in sources beyond the standard articles and Google searches on which other traders are relying.
Example
Lawrence Kudlow accepted an offer to become Trump’s Director of the National EconomicCouncil. It was unclear, however, whether Kudlow would start before or after the date when the PredictIt market would close. Ten days before the date, a member of my PredictIt group, knowing that Kudlow had a radio show, went on the website, found the station phone number and asked when Kudlow’s shows were going to be live for the next week. The rep confirmed that Kudlow’s show would be on for two more weeks. He also confirmed that the radio station was hundreds of miles away from Washington,D.C. With this information, he could make a confident No bet and cash out quickly.
Take Advantage of Breaking News: Breaking news tends to produce temporary overreactions in the markets.
Example
When news broke that Trump agreed to meet with Kim Jong Un, the markets shot up from 10% to 90%. A member of my PredictIt group took advantage by considering all the ways the meeting could be canceled and/or postponed and how the news media would harp on any obstacle that might derail the summit. He calculated that the market was overvalued at 90% and bought No shares. Eventually, the price went down to 70% and he cashed out.
Pay Attention to PredictIt Ads: The volume on traders on PredictIt remains small enough that targeted advertisements can draw enough traders from a certain demographic to bias markets.
Example
During the 2016 elections, PredictIt began to run ads on the populist, right-wing site Breitbart. The ensuing influx of die-hard Trump supporters on PredictIt led to a number of yes bets in the “Will a federal criminal charge be filed against Hillary Clinton?” market at inflated prices.
Read the Rules Closely: In every market on PredictIt, there is a detailed description of the precise rules on how PredictIt will decide which position wins. Often, reading the fine print will allow you to see opportunities created by investors who are making bets based on false assumptions.
Example
In the run-up to the vice president debate between Mike Pence and Tim Kaine, PredictIt created a market titled “Will a VP debate moderator ask about ‘religious liberty’?” My initial instinct was to buy No shares at a reasonably high price given that “religious liberty” is a conservative phraseology used predominantly by Christian conservatives with an ideology not typically shared by mainstream media moderators. When I looked closely at the rules, however, I learned that the market would also be settled yes if “a moderator or sanctioned questioner shall utter the phrase ‘religious liberty or ‘religious freedom.” The inclusion of “religious freedom” dramatically changed my calculation not only because it is more ideologically neutral language, but also because Mike Pence, as governor, had vetoed a controversial bill on “religious freedom.”
Be Careful in Markets with Tight Deadlines: Many markets on PredictIt close in counterintuitive ways because of the way PredictIt defines the deadlines.
Example
“Will Angela Merkel be elected German chancellor in 2017?” Merkel was the odds-on favorite to remain as chancellor, which is why the markets, at certain points, were trading yes shares above 95%. What many investors missed, however, was that Merkel would have a hard time finding coalition partners. Because she did not form a coalition until 2018, the market resolved as No.
Pay Attention to Correlated Markets: Traders can profit by identifying markets that are likely to move alongside shifts in other markets.
Example
When Democratic candidate Ralph Northam won the 2017 Virginia gubernatorial race by a larger margin than expected, the odds of Rep. Barbara Comstock, a Republican from northern Virginia, plummeted from the mid-30s to around 10%.
Exploit Arbitrage Opportunities: In a 1995 study professors Kenneth Oliven and Thomas Rietz found that arbitrage opportunities are widespread in the Iowa Political Stock Market (IPSM). This is because traders
regularly violated assumptions underlying the efficient market hypothesis… Traders could have made money by exploiting arbitrage opportunities or setting bids and asks to exploit violations of the law of one price. However, the final prices in the market proved to be very efficient. Indeed, most markets run on the IPSM have proven to be very efficient…
At first brush, this leaves us with a quandary: How can market traders repeatedly violate concepts assumed necessary for any kind of market efficiency and simultaneously drive final market prices to efficient levels? More in-depth analysis shows that active, knowledgeable, experienced and educated market makers with larger orders can drive prices to efficient levels while profiting from other (price-taking) traders’ mistakes. (14-15)
Take Advantage of Favorite-Longshot Bias: Intrade betters during the 2008 presidential elections exhibited “favorite-longshot bias,” in that they tended to “underprice in-the-money options, and overprice out-of-the money options.” Professors John Kros, Enping Mai, and Christopher Keller determined that traders were not willing to pay for “favorites” despite their probabilistic over-performance, but were overpaying for “longshots” relative to their probabilistic underperformance.
Example
In the November 2016 presidential election markets, there
were states where Clinton had a 99% percent chance of winning according to FiveThirtyEight models, and a 20%+ lead in the polls. Yet on PredictIt,
Clinton shares were trading in the low 90s.
Part of the long-shot bias might be explained by leverage.
In a market where one is only permitted to bet $850, those taking the longshot
can exert a much greater influence on the price than someone who takes No. For
instance, if a longshot is priced at 10 cents then you can buy 8500 Yes, but
only 944 Nos.
Look for Discrepancies between Words and Non-Verbal Cues: News reports, especially in print, will focus on a formal statement, which will move the markets. The real insight from a betting standpoint, however, can be gleaned from body language and other indicators.
Example
Asked in an interview whether he would appointed a second special counsel, Attorney General Jeff Sessions replied that he would make cleaning up the Department of Justice and FBI a top priority. Conservative media in particular ran stories about how Sessions was considering appointing a second counsel, which they had been advocating. Sessions’ body language and tone in the video, however, made it obvious that he had no intention of doing so. He avoided answering questions directly and made a general statement that was true—a major red flag that he was being deceptive. When asked about sanctuary cities in the same interview, Sessions became much more passionate and listed a number of specific policies he was implementing to punish those responsible for creating sanctuary cities.
No bets in the second special counsel markets shortly after the interview paid off quickly after Sessions subsequently announced that he would not be appointing a counsel.
Compare and Contrast Market Prices on PredictIt with Prices on Other Exchanges: The market price in other political prediction sites such as Betfair can alert traders to irrational prices.
Be Mindful of Limits on the Number of Total Traders Allowed in Each Contract: There is no way for investors to tell, as far as I know, when a particular market will reach the 5,000 limit on the number of traders allowed. I avoid getting blocked out of markets I’m interested in by immediate buying and keeping at least one share. This has the added advantage of making it easier for me to navigate my contracts as I’m moving money around on a day-to-day basis.
Look for False Comparisons: Traders create profitable opportunities when they make bets based on false/irrelevant comparisons with other events. This often happens when traders draw and overstate parallels between a past election outcome and an upcoming one even if the two elections differ.
Example
On Tuesday, September 4, 2018, The markets were surprised when leftist Ayanna Pressley toppled 10-term incumbent Massachusetts congressman Michael Capuano in her primary challenge in Massachusetts’ 7th Congressional district.. The odds of Delaware’s Democratic Senator Tom Carper surviving a primary challenge two days later from another woman of color Kerri Harris subsequently fell from about 93% into the mid-70s. The likelihood of an upset in Delware was significantly overpriced. Faulty comparisons were skewing the markets. Victories by Alexandria Ocasio-Cortez and Ayanna Pressley were significant, but were the exceptions in a cycle that had generally seen victories by establishment Democrats. Orchestrating primary upsets in far-left congressional districts is an entirely different phenomenon from doing the same in a statewide race with a center-left electorate. I saw little evidence, either in terms of polling or more qualitative indicators, that Carper was in trouble. PredictIt forums were replete with emotional pro-Harris partisans that were indicative of emotional buying. I bought up as many shares as possible of Carper in the 70s and 80s, which rose to 99% within a few hours.