Ryan_Gilmore
Community Manager Community Manager

Episode 17 (Monday, May 11, 2020)

 

Segment Description Who's On Air

Welcome

The Monologue

@julian

Featured Program

 

What Can Baseball Data Tell Us About Ourselves?

When you think of analytics in baseball you immediately think about traditional measures like hits, runs, and on base percentage. If you’re a baseball enthusiast you might know about advanced metrics, like FIP, wOBA and other Sabermetric terms. In this session, Sig Mejdal, one of the first analysts in professional baseball, shares with us some less obvious or unusual metrics that influence success for professional baseball players. Learn how phenomena like the California real estate boom; the proliferation of air conditioning in Florida; and westward and southward U.S. expansion affected where major league teams looked for players for their organizations. Along the way Sig will also show us how the average location of major leaguers has changed over time, using an animated bubble charts and examines the “age cutoff” observation analysis Malcom Gladwell made famous in Outliers. Find out how month of birth can influence the early and continued development of strong players and how it can affect current (and perhaps future) major leaguers. Sig ends his talk by showing us yet another “boring data table,” specifically player’s height, and noting what this data tells us about current major leaguers.

 

@Sig
Featured Program

 

Mamba Magnificence: A look back at Kobe Bryant’s career

Basketball has always ruled Principal Research Statistician Developer Clay Barker’s world. As a youngster he fell in love with the Atlanta Hawks and his passion for the game only intensified when he moved from Atlanta to the basketball Mecca of Chapel Hill, North Carolina. So naturally, when it came time to choose a career, all those years studying the back of his basketball cards made the decision to pursue statistics an easy one. In this presentation, Clay looks at scoring over the course of Kobe Bryant’s 20-year career to see if his performance really dropped in that latter stages of his career, as many of his critics claim. Using JMP’s Generalized Regression Platform, Clay uses the lasso to look for changepoints along with other features like home court indicator, opponent, nights off (by season), and playoff indictor to analyze and visualize how Kobe’s performance changed over time. Clay notes some obvious conclusions, particularly when looking at the first decade against the second, and as a bonus, examines how Kobe’s career trajectory compares to Michael Jordan, who many believed to be the best basketball player in NBA history.

 

@clay_barker
Featured Program

 

JMP into Sports Analytics

Dr. Tim Chartier is a professor of mathematics and computer science at Davidson College, where he leads a 100-member sports analytics group that supplies analytics to the college's coaches in various sports. The team also provides analytics to organizations all around the sports world to show them how sports analytics can improve their teams’ performance. In this segment, Dr. Chartier shows how his team used JMP to predict which players would be become MLB All-Stars and which teams would participate in the annual college basketball tournament known as March Madness. Using partition analysis in JMP, Chartier and his team determined the best predictor for becoming an MLB All-Star was total bases, with RBI being a secondary predictive statistic. But that wasn’t all, his team also determined that Twitter followers was a strong predictor as well – specifically, if you have over 100,000 Twitter followers, you’re likely to be an All-Star, even if your statistics fall slightly short of the mark! In the second half of his talk, Chartier shows how his team was able to use four years of game logs, JMP and partition analysis to explain the perfect storm of factors that allowed an underdog of historic proportions, #16 seed UMBC, to defeat #1 seed ,University of Virginia, in the 2018 NCAA tournament.

 

@julian
Featured Program

 

Statistics for Fantasy Football

With approximately 60 million people around the world playing each week, fantasy football has become almost as popular as the actual NFL games themselves. In this JMP On Air segment, Joseph Bryan, founder/creator of Koalaty Statistics shows us how he uses analytics (along with JMP and R) to produce a week-to-week prediction of how players will perform. As an example, Joseph demonstrates how he used a number of metrics, including the critical metric of “air yards,” (yards the ball travels in the air), to help improve his followers’ chances of picking the right wide receiver during the most recent NFL season. Through the presentation, Joseph also gives you additional nuggets to help your game, including his observation that every player in the NFL is always regressing toward their mean and that good games do not imply more good games and likewise for bad games. Want to win your fantasy football league this year? Start by watching this presentation.

 

@julian
Featured Program

 

What's the Value of a Shot in Hockey? 

How does a southerner from Tennessee, now living in Georgia, become a consultant to executives in the National Hockey League? Write a blog charting Kobe Bryant 30,699 career shots. At least that’s how SAS Analytical Consultant Sam Edgemon landed the gig. Combining his love of analytics and sports, and a WHOLE lot of play-by-play data provided by the NHL, Sam helped the NHL team better understand the importance of “shot value.” Specifically, Sam discovered that there are key spots on the ice where the probability of scoring is much higher. Using the percentage of goals scored based on location on ice where the shot was taken, he created a “shot value” metric for each shot. The metric, and the visualizations Sam created in support of it, now helps the team in so many ways – how they attack the net, how they draft, how they pursued free agents, and more.

 

@samedgemon
Featured Program

 

Where Olympic Champions Are Born

JMP Research Statistician Caleb King joins us to give us a look at where Olympic champions are born. Using JMP and data from Olympic Reference, Caleb explored data about where gold medalists are from, geocoded birthplace data for all Olympians and shares his insights. Are basketball and beach volleyball players primarily from the U.S.? Do baseball tend to come from Cuba? Are Alpine skiers really from the Alps? Using simple data filters in JMP, the world map and animated time series see for yourself. Spoiler alert: Clay’s animations show history in action and you might be surprised by the top two cities to produce gold medal winners, and by how much a lead they still hold over the rest of the world.

 

@calking
Tip of the Day

 

Graphlets in Graph Builder

Mary and Pete are football fans of the New York Giants and Denver Broncos respectively, so they haven’t enjoyed a lot of on-field success to celebrate recently. But, how about when it comes to the NFL draft: have their organizations at least been drafting successfully? In this episode of the JMP Tip of the Day Pete and Mary explore NFL data to see how often their teams (and all the other NFL franchises) pick up the 5th year options for their 1st round draft choices. (If a team exercises the 5th year option, that’s usually an indicator of a valuable player they want on the team for another year.) Using Graphlets in JMP Graph Builder, Pete shows Mary if the data says their teams are “winning” off the field, if not on it!

 

@phersh
Closing The Last 5 @julian 

 

Article Labels
Article Tags