Bob Costas And Baseball Analytics

I want to ask your indulgence this morning.  I would like you to listen to a commentary by Bob Costas on a recent episode of The Rich Eisen Show.  Please follow this link; click on the picture in the Tweet and listen to Costas’ comments.  It takes about 7 minutes; it is worth your time; I will wait here for you to come back.

That commentary would have been excellent if it had been written out beforehand and “read into the record”.  Rather obviously, that was not the case; that was an extemporaneous analysis and opinion piece.  And … there is so much truth in there.  Those two “standing ovations” from the producers and staff of The Rich Eisen Show were perfectly appropriate.

Even though many folks had said previously what Costas said in his comments, his eloquence adds weight to the arguments.

• Being a slave to analytics is not good strategy or tactics in baseball.
• Fans do not really care about newfangled stats such as launch angle and exit velocity and spin rate.  They cannot see any of them on TV or at the ballpark.
• Pace of play is working against MLB as an entertainment product.

The Rays got to the World Series by adhering – – for the most part – – to a tactical situation where their starting pitcher did not go through the opposing lineup more than twice.  Obviously, that had worked for them because they were in the World Series and 28 other teams were not.  In Game 6 of the World Series every “analytics acolyte” would choose to pull Blake Snell because “the numbers” show that doing so is a path to success.  The problem is that “the numbers” they rely on allow them to ignore some other numbers:

• Snell had thrown only 73 pitches.  I understand that starting pitchers in 2020 are far more fragile than pitchers in the past – – but if 73 pitches becomes a “burdensome load”, there is almost no reason to differentiate between starting pitchers and relief pitchers.
• Snell had struck out 9 batters in 5 innings – – and had struck out the guys who were due to come up to face him in the 6^th inning.
• Those numbers are not statistical trends; those numbers represent what was happening on the field in that game on that night.  They were ignored so that the same guys who brought you “stadium-adjusted OPS+” would be happy.
• That analytical decision backfired spectacularly.  No one will ever know if Blake Snell would have crashed and burned in an even more spectacular fashion had he been left in the game until it appeared that the Dodgers had figured out how to get some hits off him.

Bob Costas makes a great point in that rant when he referred to the old Strat-O-Matic baseball game.  Every player has a card that represents his statistics for a season or a career.  The “Ted Williams Card” had far more chances to get a hit or draw a walk than did “The Sparky Anderson Card”.  The “Sandy Koufax Card” produced more strikeouts than did “The Humberto Robinson Card”.  Every player in Strat-O-Matic brought the same “effort” and “energy” to the table in every game.  And that is simply not the case in real life.  Every great player has an off-day; every mediocre hitter has an at-bat that produces a grand slam (e.g. Bill Mazeroski).  Analytical stats come from the past and the game on the field is happening in the present.  That is why a slavish addiction to analytics has not worked in baseball even though it has been around a LONG time.

Take a moment and recall Billy Beane and “Moneyball”.  It was all the rage about 25 years ago; Michael Lewis wrote a book about it and they made a movie based on the book.  It was a totally different way to evaluate players as an executive went though the process of building a roster.  And it worked – – to a degree.  The “Moneyball” teams in Oakland never got past the Division Series in the AL let alone get to or win a World Series.  The teams were very “cost effective” in terms of wins per dollar spent on player salaries – – but that is not how baseball seasons are decided.

To my mind, the problem is not “analytics” per se; the problem is using them to the exclusion of what is happening in front of one’s nose.  Another aspect of analytics that is troubling is that it is spreading into the world of football.  Players and teams now get a DVOA rating; DVOA stands for Defense-adjusted Value Over Average.  When you tune into a college game this Saturday or possibly to an NFL game on Sunday, send me a note telling me where you observed someone’s DVOA on the field.  I often miss it…

One more thing here…  Another problem with the “analytics acolytes” is that they cannot ever admit they were wrong.  A part of the reason for that is the infallibility of the calculations of past performances that led to the critical judgment of what needed to be done.  If you do not believe in that infallibility, then you are only partially committed to analytics and/or a reactionary.

Analytics has a seat at the table in terms of evaluating players and game situations.  Unfortunately for the “analytics acolytes”, human behavior and performance cannot yet be completely modeled by mathematics.

Costas’ other major point was that pace of play was hurting baseball as an entertainment product.  Given the revenue streams for MLB, television is the critical element of the game as we know it today.  Too few people seem to recognize the criticality of television to MLB – – and to the NFL and the NBA and other sports as well.  The owners, the players, the umpires, the announcers and everyone else involved with bringing the game to the public has to understand that they are all in a partnership to produce a TV series that attracts viewers.  If MLB ceases to do that, it will be relegated to a sporting status alongside things like horseracing and boxing and tennis.  All those sports can be enjoyed on TV – – sporadically, but not as a steady diet.

Would that Bob Costas – or someone who thinks as he does – had been named Commissioner of Baseball somewhere along the path of the last 30 years or so.  We can never know how the game would be different – and presumably better – had that happened because one can never describe fully events derived from a contrafactual statement.  Too bad…

Finally, a former colleague – and an avid Denver Broncos’ fan – sent me an email back in the middle of the summer as we were all going through the potential for restarting sports and dealing with COVID-19 and witnessing the ramping up for our Presidential election.  He was bemoaning the chaos of 2020 and wondering how and why we got to where we were.  The way that Game 6 of the World Series played out leading to the Bob Costas commentary here would seem to add another layer to his outcry:

“Remember, everything was under control until the Chiefs won the Super Bowl.”

But don’t get me wrong, I love sports………