11/22/24

How Ranking Systems Influence Our World and Reflect the Goals of Their Creators

Ranking systems are not value-agnostic; they reflect the priorities and goals of their creators

by

Ranking systems control our world.

The more digitized society has become, the stronger the grip ranking systems and algorithms have on our lives. They have stretched their tentacles into nearly every facet of human life, dictating our behavior and shaping our choices. Restaurant rankings influence where we get our meals. University rankings guide ambitious students in their educational pursuits. Golf rankings determine not only who we consider the best player in the world but also which golfers qualify for the most prestigious tournaments, the value of television contracts and sponsorship deals, and more.

A ranking system is anything that assigns a priority order to items within a list, though many of the rankings we interact with on a daily basis are not immediately recognized as such. For example, the Google PageRank algorithm determines the order of search results on the world’s most popular search engine, shaping the way we access information. On X (formerly Twitter), the “For You” timeline is the product of a ranking algorithm, which curates user feeds based on a number of factors, including how likely the user is to engage with each post. Content likely to evoke an emotional reaction is rewarded by the algorithm’s methodology, therefore users are incentivized to create polarizing and divisive content. These lines of code have a profound impact on how humans experience the world.

The prevalence of ranking systems isn’t a problem unto itself; however, problems arise when the way we talk about rankings puts distance between the ranking and the human being responsible for the logic that produces the ranking. Phrases like “That’s what the data says” or “One thing analytics tells us” misrepresent who the subject is. Data doesn’t say anything. The person interpreting and communicating the data is the messenger. Language that assigns agency to data abdicates responsibility from the person producing the insight.

Ranking systems are not value-agnostic; they reflect the priorities and goals of their creators. Given their pervasive influence, it’s important to scrutinize them, call out their flaws, and, at a bare minimum, understand how they work and what behaviors they incentivize. With that in mind, let’s examine two prominent ranking systems in the golf world and what their methodologies reveal about the values and goals of the organizations behind them.

Lottia Woad, the No. 1 women's amateur in the world, after winning the Augusta National Women's Amateur.

WAGR

The World Amateur Golf Ranking (WAGR), administered by the R&A and USGA, ranks elite male and female amateur golfers. Details of WAGR criteria and methodology can be found here. WAGR is an important institution within the golf world, especially from the perspective of an amateur golfer. One’s WAGR ranking determines qualification status into many elite amateur tournaments, like the U.S. Amateur, the Walker Cup, and the Augusta National Women’s Amateur. WAGR also impacts other rankings within the amateur golf ecosystem, like PGA Tour University.

WAGR’s published mission statement provides a threefold purpose:

  • Accurately rank players as they compete in competitions
  • To provide a ranking system that enables players to compare with each other even though they may not directly compete against each other in events
  • To stimulate ambition in players and federations to succeed in development of their game and golfing ability

One of the greatest challenges WAGR faces is pursuit of the first bullet without compromising on the third bullet and vice versa. A system engineered solely to rank players as accurately as possible will often disadvantage players from developing countries who lack the resources to compete outside their home region. For instance, imagine a scenario in which a player consistently wins every tournament in his home country, but the rest of the golfers competing against him are all substandard players by any objective metric. How would we know how good that player is? And should he have the opportunity to prove his talents on bigger stages? These are two questions with which the world constantly grapples, both inside and outside of golf, in different contexts.

In the state of Texas, lawmakers have addressed those two questions within the context of college admissions. In 1997, Texas passed a Top 10 Percent Law, dictating that students who finish in the top 10% of their high school class receive automatic admission to state-funded universities. If admission to public universities in Texas were predicated entirely on a strict, objective ranking of students’ performance, using some measure like a combination of standardized test scores and GPA adjusted by high school, certain areas of Texas would never have representation at the state-funded institutions their tax dollars support. Proponents of the policy argue that high school students in some regions of the state don’t have access to resources that afford them the opportunity to compete with students in some of the wealthiest areas of Texas. In implementing the Top 10 Percent Law, Texas has adopted the philosophy that the best performing students within their peer groups deserve the opportunity to prove themselves on bigger stages, instead of selecting the best performing students across the state in the eyes of a standardized metric.

Does this policy result in a student body that would shoot the single lowest test scores of any population from which Texas could draw? Likely not, but Texas has decided to prioritize stimulating ambition across the state, even if it comes at the expense of assembling the strongest field on paper. The Top 10 Percent Law is a useful analogy for understanding the tradeoff between accuracy and accessibility and its application to amateur golf.

At the 2024 Asia-Pacific Amateur Championship this past October, amateurs Alexander Yang (71-73-65-70) and Chi Chun Chen (67-72-72-68) received 7.71 WAGR points for finishing T-16 out of 120 competitors in the field. In March 2024, amateur Rachid Akl received 7.91 WAGR points for winning an event titled Jordan Golf League 2 organized by the Jordan Golf Federation, a higher point total than T-16 at the Asia-Pacific Am registered.

Full leaderboard of Jordan Golf League 2

As you can see in the leaderboard displayed above, only three players are listed in the field for this event in Jordan, and each player received at least 5.85 points for their respective finishes. Currently, the minimum field size for WAGR-accredited events is eight players. Therefore, the correct interpretation of the above leaderboard is that at least five unlisted golfers finished with worse scores than the names displayed above. These additional five golfers do not appear on the leaderboard, though, because they are unranked by WAGR. A player becomes ranked in WAGR once they’ve earned at least 6.5 points in a WAGR-accredited event. The WAGR system treats unranked golfers as unknown entities, and for the purposes of evaluating strengths of fields, WAGR assigns unranked players a default rating just below the lowest-ranked player in WAGR’s system.

As a reference point, at this year’s AJGA’s Junior Players Championship hosted at TPC Sawgrass, the four junior golfers who finished T-16 out of 77 golfers, a group that included 16-year-old phenom Miles Russell, each earned 5.69 points, just shy of the 5.85 points earned for the above T-2 finishes in Jordan. Salem Alabdallat (90-87-89), one of the T-2 finishers at the Jordan Golf League 2, finished 77th at the Junior Players, 16 strokes behind the 76th-place finisher.

It is irresponsible to make grand conclusions from cherry-picked data points. Golf is a fickle sport, and nobody is impervious to a few bad scores. Nonetheless, when you zoom out and examine leaderboards across the amateur golf landscape, it does not require a sophisticated database to see that many of the golfers earning spots in elite amateur tournaments by way of small-field, WAGR-accredited events in developing countries underperform against their peers. Simply put, these players are ranked more highly than their performance warrants. Akl, the aforementioned winner of Jordan Golf League 2, shot 96-94 to finish 120th out of 120 players at this year’s Asia-Pacific Amateur Championship. The other two players who earned points at Jordan Golf League 2 also competed at the Asia-Pacific Am, finishing 109th and 115th. These are symptoms of a system that fails to award points properly in commensuration with performance.

Moving forward, modifications are on the horizon for how WAGR accredits tournaments and awards ranking points. In an announcement released on November 4th, WAGR unveiled changes targeted at addressing issues observed in small-field events. The release states, “With the increase in events worldwide, there has been a noticeable rise in smaller tournaments with weaker fields. Many of these events have minimum field sizes but feature few, if any, ranked players. As a result, a large portion of participants are unable to post scores that meet an elite standard…”

Beginning in 2025, minimum field size requirements will be doubled for male and mixed-gender events from eight to 16 players. In theory, this change should also make minimum-sized fields tournaments more competitive and strengthen the validity of the results. It also adds a layer of difficulty for tournament organizers hosting events with the goal of gaming the system. Additionally, points allocated to small-field events with weak fields will be reduced “by as much as a third.” I spoke with Ross Galarneault, a Director of Championships at the USGA, whose responsibilities include tournament scoring and WAGR. Galarneault explicitly stated that WAGR is making these changes in direct response to observations from the country of Jordan.

The situation in Jordan illuminates the difficulty of striking the balance between accurately ranking players and promoting access within the sport, priorities that Galarneault acknowledged as “somewhat competing ideals.” Deficiencies in the system were identified, and WAGR is taking steps to fix the leaks. It’s a cycle that repeats itself until a system is perfected. The question the golf world faces, both at the professional and amateur level, is whether such a system can be perfected. Or, are accurately ranking golfers and promoting accessibility at such odds with one another that perfection can never be achieved? If so, what is the right balance to strike?

OWGR

Discussions about the tradeoff between ranking accuracy and accessibility may remind people of the evolution of the Official World Golf Ranking (OWGR). The OWGR’s stated mission is narrower than the mission of WAGR. The OWGR aims “to administer and publish, on a weekly basis, a transparent, credible, and accurate Ranking based on the relative performances of players participating in male Eligible Golf Tours worldwide.” The OWGR purports to be solely focused on ranking male professional golfers accurately; unlike WAGR, it is not attempting to stimulate ambition in players and tours worldwide.

Fans of professional golf might recall the structure of the OWGR system before the changes announced in August of 2022. Previously, OWGR distributed elevated points totals at accredited tours’ Flagship Events. Ultimately, the OWGR Technical Committee discovered bias within the system, partially due to the elevated point totals at Flagship Events. Essentially, predominantly America-based PGA Tour players were disadvantaged within the system because elevated points totals in Flagship Events artificially overrewarded performance on tours such as the DP World Tour and the Japan Golf Tour. A study spearheaded by Columbia professor Mark Broadie, popularly known for his role in the creation of Strokes Gained statistics, concluded that the OWGR was, in fact, biased against PGA Tour players. What one person classifies as bias against PGA Tour players could be considered stimulating ambition in players and tours around the world in the eyes of another.

A comparable dilemma continues to unfold in the college football world, where rankings debates often mirror those in golf. This December, the inaugural 12-team College Football Playoffs bracket will be announced. In advance of the finalized bracket, the debate over the ranking position of the undefeated Indiana University Hoosiers (10-0) has reached a fever pitch, conjuring memories of a similar situation that divided college football fans last year when undefeated Florida State was omitted from the College Football Playoff. Indiana hasn’t faced a tough schedule of opponents this year, nor do they compete in the strongest conference in college football, the SEC. Should the College Football Playoff Selection Committee prioritize stacking the playoff bracket with teams from the most competitive tour, so to speak? Or is stimulating ambition in conferences outside the Power Four a priority worth considering? Moreover, what are the second-order effects of failing to stimulate ambition in teams outside of the strongest conferences? Will non-Power Four conferences continue to lose their most ambitious universities to conference realignment, just as the DP World Tour has been reduced to a feeder system to the PGA Tour? The College Football Playoff selection conundrum is simply a different flavor of the same problem decision-makers within the golf world navigate.

Discussions about ranking systems – whether in college football, golf, or any other context – quickly turn into fierce debates because they are about more than numbers in a list. Those who control the methodology behind rankings wield immense influence over the ecosystem they oversee.

Since LIV Golf’s inception, its most contentious issue has been gaining OWGR accreditation –  a cornerstone of the professional golf power structure. While LIV Golf’s tension with the PGA Tour tends to dominate headlines, arguably the more significant battle has been between LIV and the institutions that sit upon the OWGR governing board, namely representatives of the four major championships (the USGA, R&A, PGA of America and Augusta National).

You don’t need a tin-foil hat to suspect that LIV Golf’s ambitions have extended beyond supplanting the PGA Tour. Early in the fight for OWGR points, Majed Al Sorour, the since-ousted CEO of the Saudi Golf Federation, told The New Yorker: “If the majors decide not to have our players play? I will celebrate. I will create my own majors for my players.” On October 12, 2023, Phil Mickelson, one of the most prominent and vocal players on LIV Golf, posted that LIV was “NEVER going to get (OWGR) points” in part because “it’s a monopoly run by all the governing bodies” who are “protecting themselves.”

A Phil Mickelson X post from October 2023

Mickelson’s comments suggest he believes the OWGR has ulterior motives outside of administering and publishing an accurate ranking of male professional golfers. Evidently, he believes the board members’ goals include protecting themselves and solidifying the dominance of the four biggest tournaments in professional golf.

Mickelson and others who share his viewpoint need not look further than the OWGR website to find potential evidence of this claim. Currently, the OWGR awards 100 fixed points for winning a major championship while it awards 80 fixed points for winning the Players Championship.

A screenshot from the OWGR website

A curious piece of logic! Without these fixed point totals hard-coded into the OWGR system, The Players – featuring a larger and stronger field – would award more points than the Masters, for example. This feature of the OWGR methodology begs the question: why did the OWGR decide it was necessary to manipulate the point totals distributed at specific tournaments? If the changes introduced in 2022 were intended to base rankings solely on field strength, why do the governing bodies’ own Flagship Events retain elevated points? Does this not create a bias favoring the rulemakers themselves? And is such bias only deemed acceptable when it benefits those writing the rules?

Reasonable minds can debate what’s happening beneath the surface of the most important ranking systems, both inside and outside of the golf world. What cannot be disputed, however, is that ranking systems have immense power over all who are subjugated to their influence, and the methodology of each ranking system reflects the values and the goals of the human beings who are responsible for writing them.