Lessons From A 102% One-Day Gain

|
Includes: AYSI, TWTR
by: David Pinsen
Summary

Prediction markets can be resources for investors, in giving them odds on macro events that affect their investments, and in serving as a "sandbox" for experimentation, as we've noted before.

We explicate a "sandbox" example of a market inefficiency here; one that led to a one-day gain of 102% before fees.

We draw three general investing lessons from this trade, and close with a note on using the Kelly Criterion to size bets.

Screen capture of former Democratic candidate Rob Quist via Bing. The Folk Singer Who Ran For Congress

An Actual Market Inefficiency

Back in April, we mentioned the value of prediction markets to investors (Lessons From A 60% One-Day Gain):

In previous articles, we discussed tips investors could take away from the prediction experts at the Good Judgment Project ("Interview With A Superforecaster"), and how to use "the wisdom of crowds" to screen out bad investments (2 Screens To Avoid Bad Investments). PredictIt, the prediction market run by Victoria University, involves elements of both, while simulating some aspects of stock and option markets. As such, it can be useful tool for investors, both as a source of macro predictions related to your stock investments and as a "sandbox" of sorts where you can try out investing strategies with small dollar amounts.

In that article, one of the lessons we drew was to look for market inefficiencies. Here, we'll elaborate on one such market inefficiency we found last week, which led to a quick 102% return before fees. As before, we'll keep the discussion of politics to the minimum required to understand the trade, as our focus here is on lessons for investors.

Setting The Stage

The prediction market where the inefficiency occurred was the one for Montana at large special election last Thursday, in which a Democrat, Rob Quist, a folk singer (pictured above) faced off against Republican Internet entrepreneur Greg Gianforte, pictured below.

Photo of Greg Gianforte, via Bing. Republican Greg Gianforte

Montana is one of those states with a small enough population that it only has one Congressional Representative, hence this "at large" election was statewide. Up until the day before the election, Gianforte was favored to win: He had the endorsements of the local papers, and he had the endorsement of President Trump, who won Montana by 20.5 points last November. Consequently, Gianforte shares on PredictIt were trading at around 80 cents, implying an~80% chance he'd win the election (winning shares are redeemed at $1).

Twitter Moves The Market

We've written before about how Twitter (NYSE:TWTR) can move the stock market, and we suggested that an easy way for Twitter to generate revenues would be for it to delay its feed by a few minutes, and charge users for a real-time feed (Make Twitter Great Again). Last week, Twitter showed it could move the prediction markets too. After the tweet below, by the U.S.-based Guardian reporter Ben Jacobs, shares of Gianforte dropped like a stone.

Screen capture via Twitter.

The Market Inefficiency

Reportedly, Ben Jacobs thrust a recording device in Gianforte's direction, while asking him a question about the Republican healthcare bill, and Gianforte threw him to the ground. There was no video of this, but there was audio of it, which you can listen to here. Jacobs' claim was largely corroborated by Fox News reporters who witnessed the altercation, and Gianforte was swiftly charged with misdemeanor assault. The newspapers that endorsed Gianforte withdrew their endorsements. So far, so bad for Gianforte's prospects, as his shares dropped as low as the mid-30s. But it was clear Mr. Market was inefficient here for two main reasons:

  1. Two thirds of the voters had already sent their ballots in by mail.
  2. Given the antagonism between the press and President Trump, as harsh as it sounds, it was unlikely that many Trump/Gianforte supporters would change their vote based on Gianforte allegedly assaulting a member of the press.

Seeing that inefficiency, and remembering the adage to "buy when there's blood [broken glasses] in the streets," we grabbed our wallet, added funds to PredictIt, and then went to buy Gianforte shares. By the time we were ready to buy, they were trading at 46 cents, so we placed a limit order to buy them there. When that went unfilled, we raised our limit to 48 cents, and purchased the maximum dollar amount allowed by PredictIt, as we noted in this tweet at the time.

Screen capture via PredictIt.

We sold our shares for 97 cents as returns came in Thursday night, for a profit of 102% before fees in less than 24 hours.

Screen capture via PredictIt.

We'll conclude with a few lessons from this trade.

Market Inefficiencies Are Rare

It's unlikely that you will find a market inefficiency this glaring in the stock market, outside of nano-caps. With the dollar amounts at stake in liquid stocks, there are high-speed algorithmic trading programs that will likely find and exploit any inefficiency before you do. With nano-caps, you may have the opposite problem: Finding an inefficiently priced stock that remains inefficiently priced for years (a possible example of this is Alloy Steel International (OTCPK:AYSI), which we've written about in the past, most recently here).

If You Find One, Don't Be Greedy

We would have preferred to buy these Gianforte shares in the 30s, but were happy to get shares at 48 cents. Had we held out for a lower price, our order would have gone unfilled and we would have made nothing.

Base The Size Of Your Bet On The Odds

This is something we did intuitively here, as we were pressed for time. We considered the odds of Gianforte winning to be similar to what the market showed before the incident with Ben Jacobs: about 80%. So we put the maximum amount on Gianforte. But if you're in a similar situation, you can use the Kelly Criterion to precisely size your bet.

How much should we have bet according to the Kelly Criterion? Using Albion Research's handy Kelly Calculator after the fact, we entered $850 as the size of our bank roll, 1.08 to 1 as our odds (48% expressed as odds), and 80% as our estimated probability of winning, and got an answer of $522, as the screen capture below shows.

Screen capture via Albion Research.

Disclosure: I am/we are long AYSI. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.

Editor's Note: This article covers one or more microcap stocks. Please be aware of the risks associated with these stocks.