Razorback women's basketball's assistant coach uses advanced statistics to set lineups and check his own decisions.
While sports analytics certainly add a layer of depth to coverage, their origin lies with the teams and athletes themselves. In the college sports realm, wins and losses directly correlate to whether a coaching staff keeps the trust of the university brass paying its salary. At Arkansas, a university where the college programs are essentially treated as professional entities, that notion is even stronger, as a fan base loses patience when hope turns to lip service and wins turn to losses. With jobs and livelihoods on the line, any competitive advantage within the legal rules and regulations of the NCAA is welcome, and advanced statistical tracking plays directly into that. Todd Schaefer, an assistant coach for the Razorback women's basketball team, has coached for over 22 years, according to Razorback Athletics. As he progressed in his career, he recognized the need for a system of statistics that could accurately depict how players were performing. Mike Neighbors, now head women's basketball coach for Arkansas, had the same thought, and the two charted lineup efficiencies even before they worked together. In Schaefer's current role, he tracks players' analytics and uses them on a weekly basis.
Numbers helped Arkansas coaching staff move past poor first season
Coming into the 2018-19 season, coaches didn't know much about how the Razorback women's basketball team would look. Sure, Malica Monk, the squad's top scorer, was back, but several newcomers and transfers were finally getting their chance to play for the Hogs. The biggest problem was the coaches lacked any analytics for these players, and all the charts from last season were somewhat useless considering the team would be fundamentally different. Players would be playing different roles, and they didn't want to be a repeat of the staff's first-season team that went 13-8 and 3-10 in conference play. "What they did last year really isn't indicative of what we need them to do this year," Schaefer said. "This I think to start the year, way back during the summer when we went on our Italy trip, we felt like we kind of had a lot of players who were kind of the same. We played a lot of kids." This made it difficult to know which players should be in the starting lineup, how minutes should be divvied up, and who'd be slated to receive the most shots.
For a while, coaches subbed five players in and out each time they made substitutions, but they realized this technique wouldn't work once competition became increasingly difficult as the season continued. That's why analytics were needed to make those decisions. "In probably late November, we started looking at reducing the number of combinations that we were going to play in the games," Schaefer said. "We wanted to figure out which one of those worked the best instead of playing a large number, which is what we were doing in our Italy games, preseason games, and early non-conference schedule."
In terms of available analytics, the digital service "National Statistic" provides deeper analytics than the majority of regular statistics-based websites. Several of the categories listed on service's page represent why the 2018-19 team succeeded more than the previous year's squad. The O-PPP statistic takes into account how efficiently the offense scored each possession, while the FT/FGA compiles free throws and field goal attempts to generate a statistic that represents how effectively an offense is putting itself in scoring position. In 2018-19, the Razorback women's basketball team's O-PPP registered at 0.955, which was up from 0.891 the season before. For FT/FGA, the team posted a .212, which improved from .161 in 2017-18. These are just two individual analytics that show how Arkansas' offense improved. Of course, much of this improvement could be determined by simply watching the two teams and noticing one clear difference: the addition of Chelsea Dungee to the lineup. While this certainly is evident without any advanced statistics to back it up, the numbers support her importance to the squad. She finished the season as a second-team all-SEC player and set the conference tournament scoring record, and many of the situations she was put in to succeed came about because of the coaches' insistence on checking the numbers.
Through statistical forecasting, Schaefer and Neighbors predicted that competition would get stronger, and they noted that their efficiency wasn't where it needed to be. Once they crunched the numbers, they came up with a starting lineup of Malica Monk, Chelsea Dungee, Alexis Tolefree, Jailyn Mason, and Kiara Williams. With this group, the team significantly improved from the staff's first year, picked up several upsets throughout the regular season, and even advanced all the way to the SEC championship game after an impressive conference tournament performance.
In terms of available analytics, the digital service "National Statistic" provides deeper analytics than the majority of regular statistics-based websites. Several of the categories listed on service's page represent why the 2018-19 team succeeded more than the previous year's squad. The O-PPP statistic takes into account how efficiently the offense scored each possession, while the FT/FGA compiles free throws and field goal attempts to generate a statistic that represents how effectively an offense is putting itself in scoring position. In 2018-19, the Razorback women's basketball team's O-PPP registered at 0.955, which was up from 0.891 the season before. For FT/FGA, the team posted a .212, which improved from .161 in 2017-18. These are just two individual analytics that show how Arkansas' offense improved. Of course, much of this improvement could be determined by simply watching the two teams and noticing one clear difference: the addition of Chelsea Dungee to the lineup. While this certainly is evident without any advanced statistics to back it up, the numbers support her importance to the squad. She finished the season as a second-team all-SEC player and set the conference tournament scoring record, and many of the situations she was put in to succeed came about because of the coaches' insistence on checking the numbers.
Through statistical forecasting, Schaefer and Neighbors predicted that competition would get stronger, and they noted that their efficiency wasn't where it needed to be. Once they crunched the numbers, they came up with a starting lineup of Malica Monk, Chelsea Dungee, Alexis Tolefree, Jailyn Mason, and Kiara Williams. With this group, the team significantly improved from the staff's first year, picked up several upsets throughout the regular season, and even advanced all the way to the SEC championship game after an impressive conference tournament performance.
Sports analytics proved one senior's positive impact despite questions
Of the three seniors on Arkansas' 2018-19 women's basketball team, one was second on the team in scoring and was considered amongst the greatest players in team history. One didn't play too much, but she was always good for a big shot when she came in. Lastly, one seemed to consistently get playing time despite not making her presence felt in the scoring column, on the glass, in the passing game, or on defense. As numbers often state, some things aren't always as they seem.
Bailey Zimmerman was No. 6 on the team in playing time, meaning she was the Razorbacks' first option off the bench. Typically, these roles are reserved for high-volume scorers or energy-laden impact players. With the time she played, one would expect the statistics to jump off the score sheet. Up to March 10, 2019, her average was 3.5 points and 3.6 rebounds per game, which wasn't anywhere close to the numbers of other teammates playing similar minutes. Her inclusion in the lineups, Schaefer said, was entirely due to how efficient the team was when she played. "Bailey played in the mid-30s last year, and her minutes are down this year only because we have more players to play," Schaefer said. "It's not because she's been less efficient with her play."
Coming into the season, Schaefer and Neighbors discussed whether Zimmerman would even find significant playing time with all the talented newcomers ready to eat up some minutes. After seeing how the numbers stacked up, it was clear that she made every single lineup combination better, and it only further improved the combinations of players that involved talented, score-first teammates who took that responsibility away from her.
"That's a testament to who she is as a player and how she can affect all kinds of players around her," Schaefer said. "Bailey's that kind of kid who's going to make the pass that's going to lead to the pass, like a hockey assist. That's not a stat in our game." Zimmerman also boxed out on the defensive end and made sure bigger teammates could get rebounds, sacrificing her rebounding stats but aiding many of the efforts that resulted in wins. "If it wasn't for Bailey Zimmerman going in there and tipping the rebound to begin with to create a loose ball or create the wild, long rebound, that doesn't happen," Schaefer said. "Again, that doesn't show up in a box score or a statistics line, but it shows up in the well that our video guy breaks down defensive measurement."
Bailey Zimmerman was No. 6 on the team in playing time, meaning she was the Razorbacks' first option off the bench. Typically, these roles are reserved for high-volume scorers or energy-laden impact players. With the time she played, one would expect the statistics to jump off the score sheet. Up to March 10, 2019, her average was 3.5 points and 3.6 rebounds per game, which wasn't anywhere close to the numbers of other teammates playing similar minutes. Her inclusion in the lineups, Schaefer said, was entirely due to how efficient the team was when she played. "Bailey played in the mid-30s last year, and her minutes are down this year only because we have more players to play," Schaefer said. "It's not because she's been less efficient with her play."
Coming into the season, Schaefer and Neighbors discussed whether Zimmerman would even find significant playing time with all the talented newcomers ready to eat up some minutes. After seeing how the numbers stacked up, it was clear that she made every single lineup combination better, and it only further improved the combinations of players that involved talented, score-first teammates who took that responsibility away from her.
"That's a testament to who she is as a player and how she can affect all kinds of players around her," Schaefer said. "Bailey's that kind of kid who's going to make the pass that's going to lead to the pass, like a hockey assist. That's not a stat in our game." Zimmerman also boxed out on the defensive end and made sure bigger teammates could get rebounds, sacrificing her rebounding stats but aiding many of the efforts that resulted in wins. "If it wasn't for Bailey Zimmerman going in there and tipping the rebound to begin with to create a loose ball or create the wild, long rebound, that doesn't happen," Schaefer said. "Again, that doesn't show up in a box score or a statistics line, but it shows up in the well that our video guy breaks down defensive measurement."