Goldman Sachs bestows the esteemed title of “partner” to about 500 people in its organization, a number matching the total number of active players in the National Basketball Association.
However, there is a key difference between how these 500 partners and 500 players are evaluated.
Professional athletes are some of the most scrutinized workers in the world. Everything they do is recorded, measured, and analyzed. Data is at the heart of every decision from promotion to compensation.
Despite what you may assume, professionals on Wall Street are the opposite. While their work creates troves of financial data, only some of it is recorded for regulatory reasons, and most daily individual activities of bankers are not measured or analyzed.
The dearth of so-called “stats” on Wall Street means that politics, relationships, and subjective perceptions are often the primary determinants of who moves up in an organization. The consequences are real, ranging from a suboptimal workforce to gender gaps in opportunities.
In a human capital-intensive industry, it is critical to have a comprehensive understanding of people and workforce analytics. Leaders must make data-driven decisions on compensation and promotion—and for that, we can look to professional sports.
Before we get to the sports pros, we need to look at the Wall Street pros. Bankers today are evaluated once a year through a combination of performance reviews and revenue attribution.
Performance reviews typically follow this format:
- Peers grade you on criteria like “teamwork”, “commercial effectiveness,” and “culture.”
- Your points are tallied and compared with others.
- You are quartiled based on grades and feedback.
Revenue attributions typically follow this format:
- Which clients do you cover?
- How much did they pay the firm? Was it up or down?
- What percentage of those revenues should be attributed to you?
There are three main problems with these “stats.”
- The performance reviews are highly subjective and susceptible to confirmation, hindsight, and conformity biases. In addition, the way these metrics are collected makes comparisons between people almost statistically meaningless; what you deem to be a good score might be very different from someone else’s.
- These “stats” measure individual performance but ignore the reality that companies are complex entities with dynamic linkages between individuals and teams. For example, a salesperson may have convinced a client to buy a complex security, but the trader who took on the risk may have lost money on the same transaction.
- There is no distinction between “Alpha” and “Beta,” which are terms used on Wall Street to indicate exceptionality (Alpha) or averageness (Beta). For example, a senior trader might have made $50 million in profits, but perhaps a junior trader in the same seat would have done the same. In other words, the $50 million could have been Beta, and not attributable to exceptional individual performance by the senior trader.
Corporate America is just starting to tackle these issues—but professional sports franchises have already spent years addressing people analytics.
Major League Baseball (MLB) was the original pioneer in rethinking the conventional metrics used to assess player performance. Introduction of Sabermetrics (the use of statistical analysis to determine player performance) made managers realize the old ways of measuring success were flawed. People analytics changed how teams chose players as well as the strategies employed to win games. (The concept eventually entered the mainstream through Michael Lewis’s 2003 book, Moneyball.)
As a result, baseball broke away from its long history of subjective player assessments and embraced data-driven decision-making to complement traditional scouting reports and in-game strategies. Sabermetrics forced teams to change the measures they valued or come up with new ones all together.
It also forced them to think about how individual performance was linked to the success of a team. Whether on the field or in the boardroom, the strongest teams consider various facets of players’ abilities and how they contribute to the team, not just individual statistics in a vacuum. To learn this lesson about the virtue of smart people analytics, we can turn to Shane Battier.
During his 14-year career in professional basketball, Shane Battier was traded multiple times. Statistically, he was mediocre. Subjectively, scouts thought he was a marginal player at best. In other words, Shane was an NBA player that didn’t tick any of the traditional boxes of success. He wasn’t a star.
Toward the end of his career, as more rigorous analytics got deployed in basketball, astute observers began realizing that Shane Battier was an asset to any team he played for. Despite putting up mediocre individual numbers, his team once won 22 straight games. As Michael Lewis wrote in a 2009 feature on this no-stat star, “Battier’s game is a weird combination of obvious weaknesses and nearly invisible strengths. When he is on the court, his teammates get better, often a lot better, and his opponents get worse—often a lot worse.”
He had many dynamic linkages to positive outcomes that were missed by traditional measurements. Shane was a star, but managers were focused on metrics that favored other individuals and didn’t understand the positive benefits he brought to a team.
Better data analysis will allow companies off the field to soon uncover their own Shane Battiers. And this time, they won’t trade him away.
While robust people analytics have yet to be widely deployed on Wall Street, it is ripe for AI-driven analysis because of the tremendous volumes of data that financial intermediation generates.
The digital exhaust of market data, electronic communication, and trade history provides data scientists with rich structured and unstructured data sets. In addition, advancements in deep-learning techniques and computing power mean that companies can learn new things about people and workforces.
Machine-learning models like neural networks can find patterns, trends, and connections within an organization that humans simply will be unable to see. Advancements in natural language processing (NLP) also enable unstructured text data to be accurately captured, indexed, and structured. This means the daily activities of bankers can now be quantified and analyzed, much like a professional athlete.
What is a trader’s average response time to a client’s inquiry, and how does that correlate with revenues? How deep is an investment banker’s relationship at a corporate client, and is there a pattern across her coverage companies that can be replicated? Does a bond salesman in New York actively share information to make his colleagues in London better?
In the era of AI-powered people analytics, some stars will continue to be stars, but some will lose their shine. On the flip side, some overlooked bankers will end up being recognized as critical to the success of a business. In the end, Wall Street will be more efficient, and more people will be rewarded based on merit rather than seat value, perception, or bias-prone metrics.
Ultimately, winning teams are built through a culmination of personnel decisions: Who they hire, who they fire, and who they promote are critical inputs.
AI will help leaders make merit-based and data-driven decisions. It will help measure the unmeasured and quantify the qualitative. It will help companies align opportunity with skill, compensation with Alpha, and promotion with positive organizational impact. It will help firms become more efficient, retain real talent, and give opportunities to overlooked employees.
It will help them win.