How Moreyball and advanced analytics changed the NBA forever

How Moreyball and advanced analytics changed the NBA forever

The roots

Probably one of the very first times when a broader audience could have heard about Moneyball was 2003 when Michael Lewis published his book Moneyball: The Art of Winning an Unfair Game. It is a true story about the Oakland Athletics baseball team and its visionary general manager Billy Beane who used a completely new approach to find valuable players for his team.

Before we move further, I need to make one remark - to keep this post digestible and readable within a reasonable attention span, I had to make some shortcuts and simplifications. But believe me, it wasn’t easy as I also love all the small details that build the story.

In 1997, Beane was promoted to a general manager position of the Oakland A’s team which had one of the smallest budgets in the league at that time. Beane didn’t give up and decided to build a competitive roster on a limited budget by focusing on baseball statistics that measure in-game activity (called sabermetrics) to hire undervalued players. This method preferred to look deeper in the data to find players who were efficient and predictable and not favor players who looked good in basic measures like stolen bases or batting average.

This way of building a baseball team was completely new and not recognized yet by other franchises. Other GMs continued to manage their teams in a traditional approach which relied on coaches, managers, players and scouts whose opinions and overall experience in baseball were the key factors when the teams were deciding on contracting new players.

Beane’s method paid off as the Athletics reached the playoffs in four consecutive years from 2000 through 2003. In 2002, they became the first team in the history of the American baseball league to win 20 games in a row. And we must remember that during this season A’s had $44 millions for salaries, while for instance Yankees’s payroll was over $120 millions.

‍

Source: Wikipedia

The vibe from these days

Source: Sportingnews.com

Evolution

We need to make a pause in Oakland's history for now and let’s get a closer look at what was happening in Texas between 2007 and 2020. In 2007, the Houston Rockets (a basketball club in the NBA) hired an expert in the data field - Daryl Morey as their new general manager. Throughout the years in Huston, Morey managed to build a team that moved from being a mediocre to a serious contender in the race for NBA championship. It happened not only because James Harden (2012) and other big names joined the team and that raised the team’s capabilities, but the real game changer was Morey's plan to leverage data into every aspect of running the Rockets team.

The franchise has invested big amounts to build their edge on data. They have developed a video system that captures all the moves of every player during the matches to analyze literally everything. Their analytics team came up with machine learning algorithms to better understand what skills are most valuable (contract value), who possesses them (prospecting) and what competitors are doing while playing against the Rockets (scouting).

Based on Morey’s analytical approach he was able to break basketball down into two things. One was how to maximize the number of points per possession and the other was how to get an extra possession.

During his tenure in Huston, the team stopped shooting from the mid-range, as data proved that they are most ineffective in scoring 2 points. Instead, the players were advised to move closer to the basket as dunks are statistically the most efficient shots that are worth 2 points. Also, data showed that a 3 point shot can be a game changer, because (simply speaking) you need 33 shots for 3 points to get the same result when you score 50 times for 2 points… and when you miss a 3-pointer, you can get an offensive rebound and finish it with a dunk.

Similarly, like in Oakland’s story, the Rockets became one of the pioneers in using advanced statistics in their field and thanks to the progress in technology they leveraged big data, IoT, ML and all of it  together gave them a huge advantage over their competitors.

During Morey’s time (2007-2020) as the GM, the team played 9 times in the NBA playoffs. In 2018, they played 82 games and won 65 times. This season, in the last game of Western conference finals they missed straight 27 shots for 3 points (NBA record)  and still managed to lose the game by only 9 points. They lost against the Golden State Warriors (who later won the NBA championship), because they didn’t want to play against their own tactics. And the probability that they would miss 20, 21, 22…27 shots in a row was decreasing with every shot, so it was kind of reasonable to carry on with 3-pointers, because the odds were on their side.

‍

One of the NBA fans summarized it with these words “The Rockets hit 36.2% of their 3’s in the 2017-18 season. The chances of a team with that 3 point % missing 27 in a row is 1 in 186 141’. If you're a basketball fan, then you must wacht it.

Conclusion

Morey had a big influence on how Rockets played and managed their team. Even a new term was coined Moreyball to highlight his unique approach. In consequences, all the NBA teams had to adapt to new standards, because it became obvious that teams like Huston would dominate the NBA. Since 2015 advanced data analytics has become a natural partner for GMs, coaches and players. And now, it’s a must-have in every franchise.

‍

The art of data and intuition

Beane’s Athletics and Rockets led by Morey are groundbreaking examples of how data can become a powerful tool, but these cases also reveal that sometimes it’s not enough and models can be accurate and appropriate, however they may still miss some important parameters that are not reflected in the data (not yet?) or they deliver the correct answers too late (real-time calculations are already knocking at the NBA’s doors).

Currently, there is an ongoing discussion that Evidence-based practice (simply: making decisions based on indisputable data and facts) that we can see in the above examples may be insufficient and new forms are being tested in e.g. business management. One of the most promising ones is called Evidence-Informed Practice and it relies on ideas introduced by Daniel Kahneman and Amos Tversky (pioneers of behavioral economics) who analyzed how human brains work and they defined two parts that are responsible for decision making. System 1 is used to make fast, based on experience and intuitive decisions in split-second. System 2 is characterized as a slower and more analytical process of reasoning to get a conclusion.

Many companies around the world have already implemented that way of managing their organizations e.g. Morey did in Huston before he change the team and now he continues his mission in Philadelphia 76ers. Leveraging data proved to produce outstanding results and one thing is certain: in today’s world, running a business that is based only on experience and intuition may be sufficient to survive the next 1-3 years. In the long-term, if these companies won’t adapt, then they will be replaced by either new entities or the organizations that have already begun incorporating data into daily operations.
Combining data with experience and knowledge is inevitable and it will shape our future, so if your company is still making decisions based on gut feeling, it's high time to change it.

(The same post has been published on our friends' blogs at: Faro.team.)

‍

Sources:

We automate the boring
text-processing stuff.

Privacy Policy
all rights reserved