Ooda

2024-07-23 See all posts



For professional gamblers, thinking in terms of expected value (EV) becomes second nature; part of your base decision-making process. That can be a double-edged sword in the less logical parts of life, but it can also lead to identifying extremely +EV betting opportunities where other people just see information. In a short timespan recently I (1) failed to position myself to take advantage of one such betting opportunity, and (2) came across a Xeet that highlighted former military strategist John Boyd’s decision-making model called the OODA loop.

OODA stands for Observer, Orient, Decide, Act. Boyd posited that, “An entity (whether an individual or an organization) that can process this cycle quickly, observing and reacting to unfolding events more rapidly than an opponent, can thereby get inside the opponent's decision cycle and gain the advantage.” A recent credit to the model, also from Wikipedia, “Dominic Cummings, Chief Advisor to the UK Prime Minister Boris Johnson, credited the success of the Vote Leave campaign in the 2016 United Kingdom European Union membership referendum to its faster processing of OODA loops, along with the focus of Vote Leave on disrupting the OODA loops of the opposing Britain Stronger in Europe campaign team.”

How does an entity shorten its OODA loop? (1) Obtain information faster, (2) Analyze information faster, (3) Make decisions based on new information faster, (4) Implement faster. This becomes more and more true as technology evolves since Technology (capital T) is the main means by which we shorten our loop; Boyd conceptualized OODA after watching a traditional war of attrition strategy fail for the United States in Vietnam. I want to note that technology doesn’t have to be Technology. The other important means used to shorten the loop exists in the people realm – all the software in the world won’t save you if your people are not constantly challenging your side’s assumptions; if they are not authorized to act quickly; if they are afraid of delivering bad news to decision makers. The people empowerment half of the equation seems like a tough nut to crack in a military application. Not so if you are a solo trader or part of a small team.

One of my first Python projects a few years back (when the Twitter API was free (rip)) would monitor NBA news-breaking accounts. If a tweet contained the name of an impact player, my program would send the tweet to a Telegram channel that was always visible on my monitor. I would then use my judgment and previous off-the-table number crunching to determine whether to bet the news and how much to bet. Typically I was looking for starting lineup news which must be released by teams 30 minutes or more before game time. I would consider things like how impactful the player is to their team’s spread / moneyline / total and whether the news was a surprise or not. Surprise sits for star players were the most profitable because it impacts the odds tremendously and the information is not priced into the market at all.

I made some profitable bets and was pretty proud of myself. It did not last long though because this is a very uncreative way of betting and books will limit you quickly and severely if and when they notice what you are doing. It is common knowledge amongst professional bettors that finding an edge is the easy part. What eventually got me limited was about $2k in winning bets against the Milwaukee Bucks when Giannis was a surprise sit. Bringing this back to the OODA concept, I won the conflict against the book because my loop was shorter than theirs. I got the news, assessed that I had a +EV bet, decided what bets to make and for what amounts, and then executed. If the book’s NBA trader had seen the Giannis news, all Bucks wagers would have disappeared in seconds with the press of a button. My specialized news feed, availability of funds, and previous number crunching won me that battle.

The most hyper-competitive technical OODA loop wars are seen in disciplines like high frequency trading (HFT) and miner extractable value (MEV, kind of a blockchain subdiscipline of HFT). But again, you do not always win the OODA war by being the most technical. Consider the case of a neighborhood grocery store owner. It is hard to offer prices competitive with big chains as the little guy. But the owner spends time in the store every day building relationships with customers and asking about their shopping habits and preferences. He knows when holidays and local events will take place and stocks oddball items related to them. He tailors his offerings to his shopper demographics better than the big chain. Stack all these agility and psychology edges and he is able to thrive by being different.

To close, a non-exhaustive summary of tactics to shorten your OODA loop: