Why has baseball become so hard to predict?

November 1, 2012  |  Sports

“Until a salary cap, MLB outcomes will remain predictable” – Frank Deford, Sports Illustrated (April 2010)

Complaints about payroll disparity in Major League Baseball have been leveled frequently over the years. The fury reached a crescendo following the 2009 season, when the New York Yankees won the World Series after paying dizzying amounts for free agents such as C.C. Sabathia. The “baseball is dead” narrative picked up considerable steam thereafter.

So when we sifted through our sports data this week, we were struck by the following: baseball has been the least predictable of the major sports over the past three years.

We previously highlighted the fact that not one of the 58 baseball pundits we track from ESPN and Sports Illustrated picked the San Francisco Giants to win the World Series. This was a fitting end to a woeful season for the MLB pundit community, which collectively generated a $0.67 yield in 2012. Recall that the Yield measures the average payout (using Vegas odds) had you placed $1 bets on each of the pundit’s selections.

What’s notable is that this simply continues a trend: baseball pundits have significantly underperformed their brethren in NFL and NBA in each of the past three years. Here is the aggregate data for each sport for the following prediction categories: Division Winners, Conference/Pennant Winners, and League Champion.

Payout if you bet $1 on all of the pundits’ predictions:

League2009201020112012Average (2010-12)
NBA$1.00$0.93$1.62$1.18
NFL$0.85$1.07$0.83$0.92
MLB$1.02$0.65$0.37$0.67$0.56

So if you had bet $1 on each of the baseball pundit picks over the past three seasons, you would have lost 44% of your money. This compares to an 8% loss for football and an 18% gain for basketball.

Another way to analyze the data is to look at the consensus favorites (as deemed by Vegas). Again, the data shows that baseball has been the least predictable sport among the three:

Payout if you bet $1 on all of the consensus favorites:

League2009201020112012Average (2010-12)
NBA$1.73$1.11$1.46$1.43
NFL$0.90$0.72$1.54$1.05
MLB$1.65$0.17$0.40$1.06$0.54

If consensus/Vegas is doing its job well, these yields should be hovering right around $1.00. The fact that the MLB Yield has been 54 cents reveals that the the ‘experts’ have dramatically underestimated the odds for the underdogs. The opposite is true for the NBA, whereby Vegas has significantly underpriced the odds for the favorites.

Breaking down the picks by division, the AL East and NL West have been vexing the pundits the most:

AL EastAL CentralAL WestNL EastNL CentralNL West
Yield$0.51$0.90$0.87$0.83$0.89$0.41

 

So why has baseball defied the “payroll disparity = predictability” argument espoused by Frank Deford and others? And why are football and basketball pundits not having the same handicapping woes (at least not to the same degree)? We have no definitive answers, but here are a few possibilities:

(1) Payroll disparity has been shrinking: See this article by Baseball Daily Digest.

(2) Cash-strapped teams are being managed better: Could the Moneyball phenomenon be at work here?

(3) Cash-laden teams are being managed worse: Alex Rodriguez and Bobby Valentine, anyone?

(4) Randomness: Perhaps this data is just noise and not a signal, as Nate Silver might say.

Let’s open it up for discussion. What’s at play here?

 

Think you can predict better than these sports “experts”? Here’s your chance to show the world… and win prizes while doing so! Learn more by clicking here or go straight to PunditTracker.com to start making predictions now. Less than one week left to get your NBA picks in, and NFL weekly picks are coming soon.

Comments

comments

 


2 Comments


  1. Seems clear to me that it’s just a result of the fact that in Basketball and Football, the better team is pretty likely to win, while baseball is more of a toss-up.

    In the NFL and NBA, the best teams may win 80% or more of their games, while the worst teams will win 20% and under. In baseball, only the very best teams will win more than 60%, and even a historically bad team like the 2012 Astros only lost 34%.

    Over the 162-game regular season, the best teams rise to the top. But in the playoffs, where you might have a team which wins 58% of its games playing a team that wins 60% of its games, the result of a 7-game series is going to be more a matter of chance than anything else.

  2. It has nothing to do with payroll. Part of it is that people do in fact tend to overestimate the predictability of the MLB playoffs. Part of it is that your sample is tiny (3 years).

Trackbacks

  1. The MLB Pundits Are Back. Will They Be Less Of A Disaster In 2014? | The Myrtle Beach Post
  2. The MLB Pundits are back: Will they do any better in 2014? | Pundits and Predictions Blog | PunditTracker
  3. Which Experts Predicted a Red Sox-Cardinals World Series? | Pundits and Predictions Blog | PunditTracker
  4. MLB: Revisiting the Pundits’ 2013 Division Picks | Pundits and Predictions Blog | PunditTracker
  5. MLB: Revisiting the Experts’ Preseason Predictions | Pundits & Predictions Blog | PunditTracker
  6. MLB Pundits: Rounding Up the 2013 Baseball Predictions | PunditTracker Blog

Leave a Reply

Current month ye@r day *

  1. Have a pundit you would like us to track? Or some general feedback on the site? Let us know!