Let’s talk sports analytics.
Sports action. Moves and counter-moves by players or managers and coaches used to be driven by feel of the game, looking in the eyes of players and gut instinct. A baseball manager who just felt that this was the right move or the right player for the situation and who was successful, was praised as a genius or Manager of the Year, and a manager who made the wrong moves was most-often soon looking for a new job as he was judged to have lost control of the team. That was then. Now, sports are ruled by metrics and analytics. Managers used to gut-feel that “this player was due for a hit.” Now, a manager will look at predicted outcomes in a situation and make decisions based on sports analytics data.
Sports analytics is the process of plugging statistics into a mathematical model or algorithms to predict the outcome of a given play or game. Three basic components to analytics are data management, predictive models that include trends, and information systems. Coaches rely on analytics to scout opponents and optimize play calls in games, while front offices use it to prioritize player development and acquisition. Performance metrics are the specific standards of various physical qualities needed for each sport of participation. These include the physical qualities of strength, speed, agility and power that are necessary for optimal athletic performance. Analytics now are the driving force behind coaching decisions and front office moves. The bottom line is that analytics are used to try to find the true impact a player makes on a game or to help those players make a bigger impact and how to play each at bat on the field – this includes defensive shifts, match-ups and pitch selection.
Back in the day, reporters, analysts, front offices and on-field management relied on the showcase stats or what we called the “eye test,” as they looked good to the eye. These stats included wins and losses, saves, strikeouts, ERA, batting average, home runs, RBIs and stolen bases (or yards thrown or gained, points per game, rebounds, goals or penalty minutes, in other sports). Now, the glory analytics in baseball are such initialese categories as WAR, wRC+, wOBA, BABIP, FIP, Exit velo and launch angle, framing (don’t get me started on umpire calls affected by how a catcher frames a pitch that was out of the strike zone), barrels, spin rate and more.
Bill James, a statistician and member of SABR (The Society for American Baseball Research) may have started the trend with his creation of Sabermetrics in the 1980s. James came up with a mathematical system to evaluate baseball players in his book, “Bill James Baseball Abstract.” In it, he created equations, including “runs created,” that factored in a baseball team’s offensive stats to predict how many runs they’d likely score. He moved on into other equations to objectively analyze players and their true value, and to help general managers optimize their teams.
The concept gained traction in 2002, when Oakland Athletics general manager Billy Beane bought into the analytics model and put together a team by replacing heralded superstars with non-stars who performed under the radar, and they nearly won a World Series – they won 103 games with the sixth smallest payroll. His strategy of optimizing a team through statistical analysis became known as “Moneyball” and quickly became the model for other teams’ operations.
I bring this up because of the selective praise of sports broadcasters. This is a rip on broadcasters and not a player judgment. During a recent Yankees-Rays game, Aaron Judge bombed two massive solo home runs. He also struck out on a bad pitch in the dirt with the bases loaded. I am not singling out Judge. He is one of the best and most productive players on the diamond who is having a historic season, and the following game, he came through in the clutch with a ninth-inning walk-off single to defeat the Astros. Three days later, he beat the Astros again with a 10th-inning walk-off homer. My gripe is not with Judge, it is with broadcasters. In the game in question, the broadcasters praised Judge’s home runs (with no one on base), while in a clutch bases-loaded situation, he went down swinging poorly. He was huge with less pressure and failed in a pressure situation. I am only saying that broadcasters should report on his successes AND his failures. Broadcasters shouldn’t be able to play it both ways. They shouldn’t spout analytics as Gospel, then report “eye-test” performance as equally Gospel. It is the broadcasters and analytics I find at fault, not Judge.
There is no perfect analytic for coming through in the clutch in high-leverage situations, though WPA/LI, or context hitting in high-leverage situations and wRC+, or weighted runs created plus, try.
Judge is a great and exciting player. All I am saying is that I saw Judge whiff on a bad pitch with the bases loaded, and he unloaded a bomb blast with the bases empty. It helped the Yankees get close in a game they eventually won. Bottom line to me … if broadcasters live by analytics, they should fully live by analytics and not mix in the eye test. A player who is praised for his successes should be equally subjected to reporting of his failures in the same game, and I find that overuse of analytics by broadcasters makes me turn down the sound and just watch the game in silence.
What are your thoughts? Do you prefer analytics or the old eye-test stats? Let me know at mike.blake@mountvernonnews.com.
See you next time.
