These advanced metrics could mean madness in NCAA Tournament
Here are eight different statistical categories that weighed heavily on the 2018-19 college basketball season.
8 statistical categories that could stir up March
Updated
Duke's Zion Williamson goes up to dunk against Florida State during the first half of the NCAA college basketball championship game of the ACC Tournament in Charlotte, N.C., Saturday, March 16, 2019.
Time is running out to complete your NCAA Tournament bracket. For some, picking winners based on, well, wins isn't enough.
Welcome to Analytics 101. Digging a bit deeper into statistics can help explain why teams — we're looking at you, Gonzaga — have been so darn good this season. Certain measures confirm what fans' eyes tell them, like the fact that Zion Williamson might be college basketball's best player. Other numbers are counter-intuitive. For instance: Chase Jeter was Arizona's highest-rated player in a handful of categories, even though he missed significant time with a back injury.
Here are the advanced metrics that could help you gain an edge in your pool:
1. Offensive (and defensive) efficiency ratings
Updated
Texas Tech guard Jarrett Culver drives past Iowa State guard Nick Weiler-Babb, left, during the first half of an NCAA college basketball game, Saturday, March 9, 2019, in Ames, Iowa.
What it is: Estimate of the points a team would score (or allow) over the span of 100 possessions.
Why it matters: The statistic blends the majority of the possessions together and then adjusts for the team's pace of play to determine the overall efficiency. Put more simply: It gauges how many points a team will score, or allow, on a given possession.
How it's calculated: The stat factors in shooting percentages, rebounding rates and free throw ability. Then, after points per possession are found, the stat is multiplied by 100 to find the rate. Ratings can also be adjusted to account for the opposition's talent.
National leaders: Gonzaga leads the nation with an adjusted offensive rating of 125.1; Texas Tech is tops with an adjusted defensive rating of 85.9.
How Arizona did: The Wildcats finished 157th nationally and 10th in the Pac-12 in adjusted offensive efficiency, and 70th in the nation and fourth in the Pac-12 in adjusted defensive efficiency.
2. Turnover percentage
Updated
Auburn's Horace Spencer holds the ball away from Tennessee's Admiral Schofield (5) and Grant Williams (2) in the second half of the NCAA college basketball SEC championship game Sunday, March 17, 2019, in Nashville, Tennessee.
What it is: The frequency in which an offensive possession results in a turnover.
Why it matters: Dean Oliver, author of Basketball on Paper and prominent statistician, divides basketball into four critical categories — shooting, rebounding, free throws and turnovers — as a determinant of team quality. Using turnover percentage, rather than raw turnover numbers, helps factor for pace.
How it's calculated: Turnovers / field goal attempts + free throw attempts + turnovers
National leaders: Notre Dame has the lowest turnover percentage, at 11.9 percent.
How Arizona did: The Wildcats turned the ball over just 14.7 percent of the time, good for 69th nationally and second-best in the Pac-12.
3. Rebounding rate
Updated
Michigan State's Nick Ward (44) battles for a rebound during the first half of an NCAA college basketball championship game against Michigan in the Big Ten Tournament, Sunday, March 17, 2019, in Chicago.
What it is: The rate at which a team grabs a rebound after a shot is missed.
Why it matters: Raw rebound totals can be skewed by pace of play or an opponent’s quality (or lack thereof), but rebounding rate essentially measures how likely a team is to snag a rebound every time a shot fails to go in the hoop.
How it's calculated: Total rebounds / (Total rebounds + opponents' rebounds)
National leaders: East Tennessee State leads the nation with a 57.5 percent total rebounding rate.
How Arizona did: Arizona's 50.5 percent total rebounding rate was 168th nationally and eight in the Pac-12.
4. Adjusted Efficiency Margin (AdjEM)
Updated
Virginia center Jack Salt (33) blocks a shot by Louisville during the first half of an NCAA college basketball game in Charlottesville, Virginia, Saturday, March 9, 2019.
What it is: The creation of college basketball statistician Ken Pomeroy to find the difference between the total points a team is expected to score minus the points a team is expected to allow. Adjusted Efficiency Margin is the core calculation Pomeroy uses to create his yearly rankings.
Why it matters: Basketball comes down to scoring more points than your opponent, and AdjEM predicts just how likely that is to happen. Only five teams — Clemson, Texas, NC State, Nebraska and Penn State — finished in Pomeroy's top 40 and failed to punch a ticket to the NCAA Tournament.
How it's calculated: Pomeroy explains on KenPom.com that AdjEM “represents the number of points the team would be expected to outscore the average D-I team over 100 possessions.” The exact calculations are proprietary, but the model does factor in things like “luck” and “consistency” to try and reduce the noise that arises from a relatively small sample size and unbalanced schedules.
National leaders: Virginia is No. 1 in Pomeroy's ranking with a AdjEM of +35.66.
How Arizona did: The Wildcats' AdjEM of +6.85 ranked 99th in nation and seventh in Pac-12.
5. Effective Field Goal Percentage (eFG%)
Updated
Iowa State players, from left, guard Nick Weiler-Babb, Cameron Lard, Lindell Wigginton and Talen Horton-Tucker celebrate from the bench during the second half of an NCAA college basketball game against Baylor in the Big 12 Tournament in Kansas City, Missouri, Thursday, March 14, 2019.
What it is: A shooting percentage calculation that accounts for the added difficulty of 3-point attempts.
Why it matters: The number of 3-point attempts has spiked throughout college basketball and the pros — thanks, Steph Curry and Steve Kerr. Regular field goal percentage fails to account for the inherent difficulty of shooting behind the arc, and the additional point a 3 earns. Keep an eye on underdogs with high eFG Percentages or 3-point attempt rates, as these Davids can topple Goliaths when they get hot from deep.
How is it calculated? Field goals made + (3FG x 0.5) / Field goal attempts
National leaders: Gonzaga leads the nation with a eFG percentage of 59.6 percent.
How Arizona did: The Wildcats' rate of 49 percent was 273rd in nation and 11th in Pac-12.
6. True Shooting Percentage (TS%)
Updated
Gonzaga's Rui Hachimura dunks against St. Mary's during the first half of an NCAA college basketball game for the WCC Tournament title, Tuesday, March 12, 2019, in Las Vegas.
What it is: Like eFG, true shooting percentage tries to better measure a player’s ability to shoot and score points.
Why it matters: True shooting percentage is still another way to gauge a player’s shot. The statistics factors in free throws, unlike eFG. TS%, along with a player's free throws attempts rate, is an effective way to see how efficiently a player can score.
How it's calculated: Points / 2 x True Shot Attempts
National leaders: Gonzaga leads the nation with a true shooting percentage of 62.7.
How Arizona did: The Wildcats' true shooting percentage of 52.6 ranked 267th in nation and 11th in Pac-12.
7. Win shares
Updated
Duke forward Zion Williamson reacts after a dunk against Texas Tech during the first half of an NCAA college basketball game on Tuesday, March 12, 2019, in New York.
What it is: According to Basketball-Reference, win shares are “an estimate of the number of wins contributed by a player.”
Why it matters: Baseball analysts learned decades ago that measuring a win is one of the most effective methods of evaluating a player’s importance to a team. As a result, WAR — wins over replacement — is a commonly used stat in baseball. Win share formulas can differ slightly based on the site or statistician. In general, the offensive win shares statistic calculates a player’s marginal offensive ability (the difference between the points Player A scores compared to the average), their turnover rate and their offensive rebounding rate. Defensive win shares factor in marginal points allowed, steals, blocks and defensive rebounding rate. The amount of “wins” created by each are added together for total win shares.
National leaders: Furman's Matt Rafferty has been worth 8.3 wins above replacement this season, tops in the country. If adjusted per 40 minutes, Duke’s Zion Williamson contributes a national-best .350 wins per game. Oregon State's Tres Tinkle (5.4 win shares) was the Pac-12's best.
How Arizona did: Center Chase Jeter led the UA with 3.1 wins above replacement.
8. Player efficiency rating (PER)
Updated
Murray State's Ja Morant (12) is the country's eighth-leading scorer this season with 24.6 points a game.
What it is: Conceptually, PER is similar to win shares. The creation of John Hollinger, the former ESPN columnist turned Memphis Grizzlies vice president, PER “sums up all a player's positive accomplishments, subtracts the negative accomplishments, and returns a per-minute rating of a player's performance."
Why it matters: PER gives a very clear baseline of a player’s ability, and can be compared over time. In the NBA, for example, a PER of 30 or more represents an all-NBA season; every player in the top 29 in career PER are in or headed to the Hall-of-Fame. PER is not as reliable in college basketball, but it still gives a clear ranking of player performance.
How it's calculated: Simply put, PER blends together everything measurable to for an individual player, and then calculates how much more (or less) efficient each player is when compared to a replacement-level player.
National leaders: Duke forward Zion Williamson leads college hoops with a PER rating of 42.55. Oregon State's Tres Tinkle (27.66 PER) was tops in the Pac-12.
How Arizona did: Center Chase Jeter (20.8 PER) was Arizona's highest-rated player.
