In making sports predictions, we look for variables that are predictive. For example, we ask whether this variable has a strong correlation from season to season. Usually, this relates to an underlying skill. However, some variables, such as a player's 3 point shooting percentage, do not follow the usual rules. This episode explores predictability vs skill and provides a framework to study the latter.
Michael Caley, soccer analytics expert and co-host of the Double Pivot podcast, joins me to talk about the beautiful game. His writing has appeared in FiveThirtyEight, The Athletic and SB Nation. He explains expected goals (xG) and the arduous process the data companies go through to calculate this quantity. He compares the current and long term prospects of big clubs like Bayern Munich and Liverpool. He also tells us whether the US men's national team has any hope.
Paul Schulz of the University of Michigan joins me to take a data based approach to the likelihood of college football and the NFL in 2020. We discuss how COVID-19 has progressed in the United States, and how the heterogeneity of the disease will impact football. He tells us about the difficulties of modeling the likelihood of a second outbreak. Finally, we put some numbers behind the probability of college football and the NFL in 2020.
Host Ed Feng talks about using pictures to explain exponential growth and how it relates to COVID-19. To check out the video with this explanation, click here.
Evan Silva of Establish the Run joins me to talk about his most recent mock draft. We get into Tampa Bay's prospects with Tom Brady and who the Bucs might draft in the first round. In discussing this year's quarterback class, Evan tells us whether he likes Tua Tagovailoa or Justin Herbert. He gives us two surprise predictions for the draft before ending with his thoughts on Astroball after the uncovering of the cheating scandal.
Jonathan Bales, co-founder of Fantasy Labs and author of multiple books on fantasy sports, joins the show. He begins by telling the story of how he did 2400 push-ups in 12 hours. Then we get into Daily Fantasy Sports and the reason you should still start playing in 2020. Finally, we discuss his counter intuitive tips for starting a business, and how a world health pandemic can create opportunities.
Benjamin Robinson uses the wisdom of crowds and mock drafts to predict the NFL draft. He describes his methods and tells us whether the "experts" are better than others at predicting the draft. Then we discuss quarterbacks and Tua Tagovailoa might go despite his injury. Finally, Benjamin gives us a surprise prediction for the 2020 NFL draft.
Cliff Sargent helps you find a great book through his YouTube channel Better Than Food. He joins host Ed Feng to talk about his story and some great books. For those that like football and data, Cliff offers some recommendations from authors such as Leonard Gardner, Tom Kristensen and Yukio Mishima.
Suppose West Virginia is elite at offense rebounding but Iowa State is awful at defensive rebounding. Can we use this to make a more accurate prediction? Host Ed Feng digs in by first explaining the four factors of basketball. Then he describes the work of Jordan Sperber on matchups in college basketball.
Ever wondered how to make a college basketball prediction based on efficiency? Host Ed Feng digs into points per possession and how to estimate this quantity from the box score. Then he gets into the least squares algorithm used to adjust for strength of schedule. This leads to the college basketball rankings on Sports Reference and kenpom.com. Finally, he talks about going from rankings to a prediction for a game.
Michigan State might not seem like championship contender given their record. They lost 6 of 10 games at one point this season. However, a surprising predictor of tournament performance likes Michigan State. Host Ed Feng breaks down the team on offense and defense and explains their poor record because of an exceptional Big Ten conference.
You got the memo on using analytics to predict the NCAA tournament, and you're ready to win. Not so fast. If you get this one thing wrong, you might as well light your entry fee on fire. Host Ed Feng explains through an analogy with Steph Curry.
The word parity has been used to describe college basketball this season. But does it actually apply? And if so, does it imply even more Madness in the tournament? In this first episode of the Bracket Wisdom series, host Ed Feng compares win probabilities for this year to the past few seasons. He gets into his latest methods for predicting college basketball, and discusses Dayton, San Diego State, Arizona and Michigan State along the way.
Adam Stanco, producer at the Pac-12 Network and hoops expert, joins me for a wide ranging college basketball conversation. He describes the landscape for the 2020 NCAA tournament and determines whether a mid-major like Dayton or San Diego State could win. He also talks about Michigan State, Arizona and Seton Hall among other teams.
Ken Pomeroy, college basketball analytics guru, joins the show to talk about his work. He discusses the statistic that he uses to evaluate home court (no, it's not points). Then Ken tells us about his favorite research project out of the many he's done over the years. Finally, we discuss teams that can win the 2020 NCAA tournament and whether mid-major programs like Dayton and San Diego State belong on this list.
To give yourself the best odds of winning your March Madness pool, you need the right strategy. Host Ed Feng discusses how this worked in last year's tournament.
Nick Kostos, sports betting personality at Sports Illustrated, joins me to discuss the Super Bowl. He begins with his journey from writing articles at Bleacher Report to CBS to videos at Sports Illustrated. Then we break down the Super Bowl match up between San Francisco and Kansas City. Nick also shares his thoughts on social media, food, books and the important things in life.
Host Ed Feng breaks down the Super Bowl between San Francisco and Kansas City. He uses snap data to analyze match ups in this game, and discusses an important factor that could swing the game. He ends with a prediction based on his analytics at ThePowerRank.com.
Bob Stoll, a pioneer in using football analytics for betting, joins me to talk NFL analytics. He describes his model and how he incorporates statistics like air yards and QB pressures. Then Bob tells us about the surprising resource he uses for injury adjustments. Finally, we discuss the conference championship games: Tennessee at Kansas City and Green Bay at San Francisco.
JJ Zachariason, content guru at FanDuel and numberFire, joins me to talk analytics and the NFL Divisional Playoffs. He tells us about the surprising statistic for college running backs that he uses to project NFL performance. Then we get into the 4 playoff games. Among other insights, he discusses the situation that makes San Francisco even more potent and the truth about Kansas City's defense.
New Orleans might be the most complete NFL team on both sides of the ball. Host Ed Feng breaks down this team and their odds to win the Super Bowl.
Should you fade teams with interim coaches? Host Ed Feng looks at the data on bowl games with one interim coach.
Michigan plays Alabama in the Citrus Bowl. Host Ed Feng digs into the story behind these two teams this season and gives a prediction for the game.
You know to use data and analytics to predict bowl games. But what about another resource, one that doesn't involve data from the current season? Host Ed Feng tells you about a resource and what it says about Washington State vs Air Force.