How the NCAA's analytic formulas allow college basketball teams to game the system, start postgame drama

For many years, the NCAA Tournament field has been picked by a Selection Committee, tasked with seeding teams based on a combination of factors to create the best March Madness possible. To help this process and remove controversy, many analytic models have been introduced to the closed meetings.

The results have been largely positive, but create a unique dichotomy between traditional sportsmanship and maximized outcomes. From a change in schedule philosophy down to the controversial decision to dunk in a 20-point blowout just before time expires, coaches have altered their strategy to appease the computers.

Most recently, in a rare February non-conference matchup between Memphis and Utah State, Penny Hardaway took exception with MJ Collins’ decision to score two more points late, forcing Jerrod Calhoun to explain why.

Incapable of emotions, the computer numbers simply do not care about the feelings efficiency invoke. They also lack the ability to differentiate crunch-time performance from blowout fun without human help to isolate the data. But at the core of the controversy, many still wonder, does it matter?

Analytics in college basketball

College basketball features many analytics which aim to predict outcomes and rank the best teams in the sport. With more than 360 teams competing for spots inside the NCAA Tournament field, finding ways outside the eye-test and traditional records is critical.

From KenPom to Torvik, and now the NCAA’s official NET Rankings — plus more specialized numbers like BPI, KPI and WAB, there is no shortage of ways to sort through the performance of each team. At the core of these models is a simple goal, to use data to predict which team is the best in college basketball. To achieve that goal, they must condense each win and loss into numbers which can be compared.

This leads to a separation between offensive efficiency and defensive efficiency, along with strength of schedule and the quadrant system. The first two help determine how successful a team performs on both ends of the floor, while the second two weigh how impressive that dominance should be in context.

The latter impacts how each team schedules around their circumstance. Many home-and-home series have been abandoned over recent years in favor of neutral site games, which minimize the risk of a home loss becoming detrimental to the resume. It has also incentivized teams without an abundance of Q1 and Q2 opportunities in conference play to schedule one-off matchups with teams in similar situations — and a handful of such games between top-tier programs looking to create an edge in the battle for a top seeed.

The former is the one which creates the most drama on the court and in press conferences throughout the season. Whether a high-major coach is running up the score on a mid-major opponent, letting the scoreboard reach big numbers before calling it off, or the coach of a team calls for his players to score even after the opposition concedes in the final seconds without fouling. The diversion from commonly practiced sportsmanship often does not sit well.

Sportsmanship in college basketball

This perceived lack of sportsmanship often does not come from a malicious place, although it does sometimes cloud the ability to tell whether the actions are intended to come off aggressive. However, these actions always place a coach and players already having a bad day in a position where they have been taught over and over to feel disrespected.

As many coaches will point out, every possession counts toward a teams offensive efficiency. That means taking a late shot-clock violation or even dribbling out the final seconds will ding the team across the computer numbers. In another isolated issue created by the pursuit of efficiency, some coaches are now less inclined to put walk-ons and end-of-the-bench players in for a 90-second curtain call because of the drop they could cause.

The counter to this argument is simple. In a game which features significantly more than 50 possessions on each end on average, the impact of one empty possession should only impact the game score by a minute percentage. Multiplied over 30-plus games in the season, it has a fraction of a percentage difference which is ultimately unlikely to create serious movement across the board.

More than a tangible benefit it provides to NCAA Tournament seeding, it is part of a program-wide commitment to follow the analytics and maximize every opportunity to capture those fractions of a percent until they add up to a big number. But the question left to be decided by each coach and their opponent in the meantime is whether those gains are worth the death of a generations-long show of sportsmanship.