DEBATING THE STAT-HEADS ON TWITTER

On Tuesday night I got into a debate on twitter over the value of all the new metrics and stats that have been created in an attempt to analyze baseball players.  Let me say, it is very hard to debate anything on Twitter with the 140 character limit, but both sides tried.  And for the most part we kept from calling each other idiots.

Let me say from the outset having been closely related to baseball for almost thirty years as an announcer starting in the minor leagues I have talked with hundreds of players, managers, scouts, general managers and owners collectively during that period.  I have watched and/or announced thousands of games in person.  I have learned from the old line baseball people and listened to the new breed of numbers guys.   Both have their value.  However, no one will ever convince me that the old line baseball guys-and that includes the folks who started in the game young, but became old as we all hope to do, don’t still have the most value.

There is absolutely no better way to evaluate a player than with the eyes.  However, to get the right line it takes more than just a quick visit to one of his high school or college games.  Hitters have to be seen versus all types of pitching.  Pitchers have to be seen against all types of hitters.  The mental aspects of how they play the game, related to team mates and coaches have to been evaluated.   What they do will produce numbers.  Some will be the traditional numbers that some of the new breed love to denigrate as unimportant.  But even the “new” numbers that can be culled from their performances don’t mean as much as what those scouts actually saw.

What the numbers…any of them… do is to give someone an idea of what he might see before he sees it.  A player with a high batting average may be expected to get some hits.  A pitcher with a high strikeout count may be expected to fan a few.  A high walk total or low on base percentage shows what the player has done in the past.   But none of those numbers show what he has done exactly.  Have his hits been line drives or bloops?  Does the ball rocket off the bat?  Is his bat quick?  Have the strikeouts been on good pitches or simply against over matched hitters?  Can he throw the ball with command?  Does he have a good breaking pitch or change-up?

Today stats for such things as percentage of pitches swung and missed are readily available on the pro level.  Spray charts that show where hitters have hit the ball most frequently can be found.  All are nice guides but they cannot ever be considered fool proof because robots don’t play baseball.  Men do.  And men are not robots.  All statistics are based on the past.  There is no way to predict the future.  Positioning a defense or signing a player based on stats is done based on history.

If your father died early of a heart attack knowing what we do of genetics you might be considered a candidate in the future as well.  But because you are aware of that you change your diet, exercise and keep your weight down, reduce stress in your life and wind up living to 85 or more.  The past did not dictate your future.   That is the same as getting caught relying entirely on statistics that show only the past.  Players can change.  Good ones will.

Admittedly most of the new stats or metrics seem to have been designed to give fans (or sportswriters) a way to rank players.  Why that is needed perhaps goes back to the birth of fantasy baseball.  Certainly everyone knew that Barry Bonds was the most dangerous hitter during his playing days.  We didn’t need someone to add his slugging percentage and on base percentage into a new artificial stat– OPS to tell us.  And did it really matter who was number six and seven on the list?

But the argument the “old guard”has most is not as much with the new useless stuff like WAR…wins above replacement…a stat that has no value except for ranking and an awful name.  Rather it is with those who have devised the new stuff who now declare the stats most dear to the history of the game are nearly worthless.

That is where real War is declared.  It is true that no statistic, old or new, is perfect.  However, the old ones like batting average, runs batted in, wins and losses are the base of the game’s history.  Batting average is what it is.  Does it show how many of the hits acquired were solid smashes or bloops?  No, but over the course of a long season it shows how successfully that player got on base after hitting the ball.  Does it need to be more?

Slugging percentage does show more how many hits were solidly stroked because extra base hits get more credit.  And on base percentage shows how frequently a player reaches safely.  Those two stats have been kept for decades. Then someone decided to add them together.  Why?  To rank players only.  All baseball fans knew Babe Ruth got on base a lot and he had a high slugging percentage.  The same with Ted Williams or Stan Musial, Willie Mays, Frank Robinson and Hank Aaron.  They weren’t all alike, but all had good numbers.  We also knew that it was generally felt an on base percentage of .375 or more was considered good, but the higher the batting average so should the on base percentage be higher.  A .325 hitter with a .360 on base percentage quite obviously rarely walked.    A hitter with a .500 slugging percentage but a .310 on base percentage would likely have a high strikeout total and lower batting average. We all knew that, but didn’t feel a need to rank everyone.

The game has changed even the veteran followers have to admit that.  Starting pitchers are not allowed to go as deeply into games. Relievers pitch more.  Long time guides such as won-loss records and earned run average as valuable statistics have been attacked.  There is some merit, but not enough to declare both are meaningless as some want to do.  For one thing they are the history of the game.  Baseball, unlike any other sport, owes a great deal of its success to its history.  Babe Ruth hit .342 with 714 home runs.  Cal Ripken,Jr. broke Lou Gehrig’s consecutive game played record, Hack Wilson drove in 191 (or 190) runs in one season.  Dizzy Dean and Denny McClain were the last 30 game winners in the two leagues.  And Ted Williams was the last .400 hitter.  Those are treasured stats and memories.  Yet, some of the new breed want to call them meaningless?

I am all for major league teams and organizations using whatever means they can statistical analysis or otherwise to help their scouts get a full look at amateur players or free agent players.  I can assure you though that whatever organization one chooses a major decision will not be made without a thorough eyeball scouting report on those players.  That is still the most important thing.

But my beef is not with the teams, but the fans who contend everything has been done wrong for a long time and old geezers don’t get it.  Fellows I can assure you that up to the point when age brings mental deterioration it brings wisdom first.  A lot of that wisdom comes from experience and good old common sense. Hopefully, you will all be able to experience it someday.

So you say runs batted in as a stat is over-rated.  I say why?  It seems to me to score runs someone needs to drive them in.  I will admit someone needs to be on base (except in cases of solo home runs) for the RBI man to drive them in, but if he doesn’t do it there are no runs.  Conversely if a player gets on base all the time but is never driven in he doesn’t score any runs. Where is the logic that RBIs are over-rated?   Do some players rack up a lot of seemingly meaningless RBIs in blow out games?  Or does he leave way more runners on base than he drives in? The answer to both can be yes of course.  But total RBIs is simply a record of what he HAS driven in.   If so then why is another stat over-rated and simply a matter of chance–batting average with runners in scoring position?  It is agreed that if a hitters batting average overall and batting average with runners in scoring position is essentially the same his success rate is mostly a matter of chance.  But if that RISP average is significantly higher…or lower… over a long enough sample size that shows something.  (On radio and television the average with RISP is often mentioned as simply another aspect of a hitter’s record and not to prove anything.)  Yet some still contend batting average with RISP does not show anything about the makeup of a particular hitter.

The only weakness with RISP average is that it is ONLY average.  Hitters who get on base to sustain a rally may not always drive in a run.  The batting average goes up with an infield single and a runner on second, but no run actually scores.

Let me quickly add that I am not opposed to new stats only some of the convoluted invented stats that are based on trying to rank players.  For instance I would like more light given to how hitters who strike out too much hurt teams.  In my mind frequent strikeout hitters are given too much of a pass.  I want to know percentages of strikeouts in situations in which hitting the ball even if it results in an out would have helped a team more than a K.  Some strikeouts truly are nothing more than just another way of making an out.  But others are worse.  Situation:  Runners on first and third with no outs.  Each of the next two hitters strike out instead of hitting a fly ball, base hit or ball on the ground.  No runs score in the inning.  If the first hitter had hit a hard ground ball it would have like been turned into a double play, but the run would have scored.  Had the second hitter with one out hit the same ground ball and a double play resulted the inning would have been over and no run would have scored.  But if he had hit the ball good enough for a sacrifice fly on even on the ground and was able to beat out the throw from second a run would have scored.  Hitting the ball still is the percentage play versus a strikeout.

I also want someone to start recording plays not made on defense.  You know the plays that should have been made but were not charged as errors.  This includes outfielders getting bad jumps on balls that fall.  It includes plays that would have been made by others players with more speed or skill.  Double plays that are not completed would be on the list.  Catcher’s throws on steal attempts that would have gotten the runner but were off target and plays are are made “safely” instead of properly.  That would include bunt plays during Steve Garvey’s career.  He would never make a throw and always took the easy out at first.

Those stats and others determine a lot of winners and losers.  Knowing what the WAR is of your left fielder does nothing, but clutter the landscape.  Knowing how a players on base percentage compared to players of his type does nothing to help a team win a game.

I have gone on too long here, but in a nutshell let me know a players batting average, home runs, rbis, stolen bases and attempts, slugging percentage, on base percentage, strikeouts and walks…with a little RISP on the side and I know enough.  That is, I know enough until I get a good “eyeball read” on what he looks like on the field.

With pitchers I want to know his W-L record as a simple historical fact…his earned run average for the same…his innings pitched, hits, strikeouts, walks, batting average against, home runs allowed and stolen bases allowed and I have a good initial read.  Then, like the hitter I have to actually see him play.

All of the numbers I use, like the numbers some of the new stats guys use, are only guidelines.  They do not tell me if the hitter can sustain those numbers of if the pitcher can successfully get major league hitters out.  I have to see that first.

Let us all hope the numbers, metrics or whatever that are called do not take over the game and push out the great numbers from the game’s history.  But most of all let us not forget the game is ON the field…not in computers.