Broken Baseball Numbers A Review Of Sabermetrics And What It Means To The Game

Pregame Warm-Up

I have been around baseball all of my life. Some of my best memories are from sandlot games, usually with my older brother Jimmy. Once, I kid you not, I was catching and he was batting, in tight on the plate like always, and when he swung the bat back into position, he hit me square in the forehead. Mum was pretty upset, and I was not too pleased either. The Doc said I only had a mild concussion, and the lump, with all the bruises, was gone in a week or two. I don’t remember.

But I guess my best memory was when I was down in Mexico playing a pickup game with some neighborhood boys. I slammed a home run, a line drive 10 feet, no that’s not right, 15 feet over the head of the center fielder. That was a great feeling. What was even better happened when I had my next at bat. The pitcher stopped, turned around and faced the fielders, and he yelled, Back up, big hitter! Well, I tell you, you feel like King Kong when that happens. I think I struck out.

I remember watching a Yankees and Angels pennant game. We got tickets from a friend, and it was my first big game, with people yelling and the place rocking. Towards the end, a Yankee slammed a hit into left. The runner got cut off coming home, and he hurried back to third. Only the other runner was already on the base, coming from second. He stepped off, to be courteous like, and before either runner did anything, the catcher tagged them both. He looked up at the umpire, and believe it or not, he let the one runner from second keep the base? The catcher, coach, and manager went crazy! Some of them got thrown out of the game for being pretty upset, I can’t remember how many. I still can’t believe this happened, and of course, the Yankees won.

I guess I’ve always loved baseball. My kids think I’m nuts, but I was named after my great uncle who played minor league ball, up around Stockton, I think. And I believe my kids have enjoyed going to the games with me. I remember when my boy and I saw Randy Johnson pitch for the D’backs. Believe it or not, we saw him strike a batter out and then he headed toward the dugout. The whole team followed, and a minute later, the umpire had them back on the field. Yes, there were only two outs. The whole stadium seemed to have a good laugh from that, especially Randy.

I love that baseball always surprises me. Some of the surprises lately have come on the math side. I always did well in math at school. I studied economics and got my fill of stats, models, and number crunching. When I was in school, you had to do it by hand. What a labor of love that was, let me tell you. I love the computers now; it makes it so much easier. But now, in today’s game, the numbers have become another way to love baseball. Unfortunately, it hasn’t always worked out that way.

That’s why I wrote this book. First, the math really isn’t that tough, I don’t think. I believe there are millions out there who would like to understand all these new baseball stats, what’s behind them, and most important of all, what they mean to the game. So this is a pretty simple, straightforward read. An elegant design is usually the simplest, yet it captures the most meaning. Think of Einstein’s E=MC². I also believe most people are smart, so conceptually, we will dive into the deep end. Don’t worry, get up to bat, you can hit.

I think this will be good for the uber sabermetrics gurus too. It’s time for a step back, a different look at what’s coming out of all of these stats. I don’t think it will hurt the gamesters either; it should give you all plenty to think about. As to the baseball professionals, you probably have been waiting on this book. I am hoping that there are a lot you that would like to tell the analysts your point of view, in a language they understand. Also you can understand a little better what they are saying to you. So let’s bridge the gap, and I hope, put things in a better perspective.

I would like to thank all my family and friends for all of your help and support. I could name you, but I’m trying to keep this low key. If the book does sale, I don’t want to enter into the cyber world of endless arguments. The great thing about a book is that you can say everything that is on your mind, take your time and state it as clearly as you can, and polish it to try to give it elegance, like the crack of a bat, the flight of a long ball or the arch of the perfect curve. Afterwards, you’re done and you can walk away. Really, you should have said all you need to say. I hope this gives you, dear reader, some pleasure and insight, and that this adds to your love of America’s pastime.

Casey Dugan

Batter Up!

Sabermetrics has changed major league baseball. The old ways of evaluating the game are gone. I don’t mean that runs aren’t scored, wins don’t count, or that the goal is no longer to win the World Series. The normal baseball fan still enjoys all of those things. He goes to the game and loves to stand and stretch, drink a cold beer, enjoy a hot dog, and root, root, root for the home team. She remembers the games with her dad, likes relaxing in the summer sun, and she can always tell if the umpire’s blind.

What I mean by the expression the old ways are gone is that now there is a science to the game. Numbers are crunched, sabermetrics are created, and with these new stats, opinions of who’s the best in the game cannot be debated – it’s a hardball fact. Mind you, these facts don’t have to be obvious to the average fan, the players, team manager and coaches, the scouts in the game, or the owners and commissioner who run major league baseball. The facts about a player’s performance are all found in a number. One single number can give you all you need to know.

And what is this number? How is it computed? And why, oh why, do the majority of us not understand this number? How is it we have enjoyed baseball for so long without understanding these fundamental truths about the game?

It used to be that the fan did understand the numbers. Heck, even as a kid we could understand a ballplayer’s batting average. We knew about a pitcher’s ERA. It was clear, well pretty clear, that RBIs were a good thing. How many home runs did he hit? How many strikeouts did he get? Who stole the most bases? And we still know a great catch when we see one, don’t we?

Of course, I know that if I argue against the new sabermetrics, then I’m confirming for you that I’m definitely an old fogey, a dinosaur. I’m someone who doesn’t own a computer, a cell phone, or live on Facebook. You, the modern fan, are trying to understand OPS, WHIP, and the mother of all numbers, WAR. You nod your head and agree with the TV analyst who smugly explains why any player over 31 years of age is on his way out. Old ballplayers, like old fans, need to go.

But why is it there are so many fans, as well as the baseball writers doing the voting, who still think the Triple Crown winner for the 2012 season was the most valuable player, the MVP, and not the wunderkind that the sabermetricians have anointed? Are the experts just waiting for the fans to catch on and realize who is right in valuing players? Do the fans not know about the facts of the game? And when will they go along with the guidance the experts are giving? Or do the old men of baseball actually know something the experts don’t? Do they know something not found in the numbers? Do they know something about the game that only comes from experience, observation, and intuition?

Don’t get me wrong, I liked Moneyball, both the book and the movie. Who couldn’t love a team like the Oakland Athletics that had very little money, but against all the odds, figured out a way to beat the rich and dominant franchises of baseball? Their manager, Billy Beane, had to watch several of their star players take lucrative contracts at the end of the 2001 season and leave the team. Power hitter Jason Giambi had knocked in 38 home runs, and outfielder Johnny Damon had scored 108 runs, both playing for the Oakland A’s in 2001. But Giambi went to the New York Yankees and Damon headed to the Boston Red Sox, both being paid far more than Oakland could afford to offer them. In fact, the Athletics’ team salary was ranked 28th out of the 30 teams in the majors for the 2002 season. Yet they went on to a record of 103-59, winning their division and advancing to the playoffs. They also posted a 20 game winning streak towards the end of the season, which set a new American League record. How did they do it? Maybe there is a lot to these sabermetric ideas?

It could be a valid point, which was one of the ideas in Moneyball, to question giving up outs in order to advance the runner. You see, in the past, managers believed that you had to manufacture runs. There is no point in getting on base if you are not going to score. And after a hitter gets a single and is safe on first, how do you get him home? This brings us to the idea of the sacrifice. The next hitter can lay down a bunt. The base runner sees the signal from the coach, knows the bunt is coming, and he can get a jump on his way to second base. The hitter is thrown out at first, but we have advanced the runner by giving up an out. The runner is now on second base and in scoring position with only one out. To take this idea of manufacturing runs even further, the next hitter may try and hit a sacrifice fly to deep right field. Once the fly ball is caught, the runner on second base tags up and races to third. Now we have a runner at third base with a much better chance of scoring. However, it cost us two outs to get him there, and the other team only needs one out to stop the run from scoring. You only have three outs for the inning, and maybe giving away outs just to move a runner is a bad idea?

Another concept developed in Moneyball was the idea that getting on base with a walk is a good thing. When we were kids, we always wanted to get a hit. A walk, however, is just as valuable as getting a hit. Isn’t it? Well, in order to increase walks, the batter has to take more pitches and be patient at the plate. Maybe he will fall behind in the count and see his strikeouts go up? On the other hand, he could get a free pass and be on base with a walk. In contrast to our first example, we advance the runner to second base without sacrificing an out by bunting. Now we have two runners on base with no outs, and maybe, the probability of scoring runs has increased. Who is to say? In any case, this gives managers another approach to managing the team. These ideas shouldn’t be that controversial, but I guess it took the language of sabermetrics to convince the old school baseball men that these ideas are worthy of consideration.

Setting aside the philosophy of sabermetrics in Moneyball, the Oakland A’s had three great pitchers in 2002, and this may be another reason why they won over 100 games that year. Tim Hudson won 15 games with an earned run average (ERA) of 2.98. Mark Mulder won 19 games with an ERA of 3.47. And Barry Zito won 23 games with an ERA of 2.75, making him the Cy Young award winning pitcher for the 2002 season. The Cy Young award recognizes the best pitcher in the league that year. As a side note, ERA is all the earned runs given up by the pitcher, averaged over a nine inning period. If a run scores because of a fielding error, it isn’t counted in his ERA. If the pitcher allows a hit or gives up a walk, and is pulled out of the game, and if these men score, the runs are counted against him. To only give up about three runs a game is very good pitching. With pitching like this, your offense only needs to score four runs to win.

The real power hitter for Oakland that year was Miguel Tejada at shortstop. He was selected American League Most Valuable Player (MVP) after posting a batting average of .308 with 34 home runs and 134 RBIs. Having over 30 home runs and 90 RBIs in a season, and averaging over three hits out of 10 at bats, a .300 batting average, makes you a top, consistent power hitter in baseball, usually placing you in the top 5% of all players. The key issue is none of these players were recruited using the sabermetrics principles. Billy Beane had not started using stats to build his team when this top talent was brought on board. Instead, a player who was used as a specific example in Moneyball of applying the new saber rules was the recruitment of Jeremy Brown. He was a college catcher who was taken in the draft. Jeremy was highlighted as a player who got walks and could get on base, and this is why they used a high draft choice to get him. But he never developed into a major league talent. He retired from baseball after only 10 major league at bats, even though Brown spent six long years with the Oakland A’s organization, most of the time in the minor leagues. So who is to say if a sabermetrics approach caused the Oakland A’s to succeed or not? Are there any other examples of sabermetrics in baseball, besides the Oakland A’s?

Surprisingly, the Oakland Athletics example in 2002 was followed by another underdog story in 2003. The Miami Marlins, or Florida Marlins as they were known then, were ranked 25th in payroll that year or about one third of the New York Yankee’s salary for the season. Oakland was spending more than the Marlins by 2003, and the A’s were ranked 23rd in payroll. The Marlins started the season poorly and had a losing record early on. They fired their manager and hired 72 year old Jack McKeon to take over. He led the Marlins to a 91-71 record and the wild card slot for the playoffs. To everyone’s surprise, the Marlins defeated the Yankees to become the World Series champions. This had a much better finish to the story than what Oakland was able to accomplish. Oakland won their Division in 2002; they were not playing as the wild card team. However, the Athletics lost right away in the playoffs and were eliminated. The Marlins won it all.

Unfortunately for the stats experts, McKeon is the oldest manager to ever win the World Series, and he doesn’t much believe in sabermetrics. A 2005 article in Baseball Analysts quoted him as saying, "Moneyball is basically computer stats. I think my style is more observation and going with your gut. I never learned my baseball out of a book. I learned it by doing it and watching the best in the game do it. I go all the way back to Branch Rickey."

Rickey was the manager in St Louis back in the early 1900’s, but he is best known for managing the Brooklyn Dodgers and bringing in the first black major league player, Jackie Robinson, in the 1940’s. This is not an example of a new sabermetric approach to the game. Regardless, there is more to sabermetrics than Moneyball, right?

I’m also appreciative of what Bill James brought to baseball as the self-made inventor of sabermetrics. With his new, in-depth analysis, and his understanding of the nuances of baseball from this fresh point of view, James wrote his first book, The Baseball Abstract, in 1977. This book is full of stats he created studying box scores. Remember, this was a time before the home computer, so a tremendous effort went into his analysis. He advertised the book in the Sporting News and it caught on. The Society for American Baseball Research (SABR) was formed in 1971 by Bob Davids, a career federal service employee who was also a baseball researcher and writer. Bill James joined the group and coined the term sabermetrics, to describe the statistical analysis they do. These fresh perspectives have given all of us a new way to question how the game is played. Innovative thinking like his is hard not to love. James is referenced in the book, Moneyball, and he was hired as a consultant by John Henry in 2004, the owner of the Boston Red Sox. John Henry was a fan of his work. Boston had not won a World Series since 1918, but after an 86 year wait, Boston won the World Series in 2004, in 2007, and then again in 2013. And as if to verify the significance of his contributions, Time Magazine added Bill James to the Time 100 in 2006, its annual list of the 100 most influential people in the world. Major league baseball revenues have grown to about $7.5 billion in 2013, with over 74 million fans attending a game this season. The sport is immensely popular in Japan, parts of Asia, and throughout Latin America. Bill James work has had quite an impact on an important global sport.

Maybe we don’t understand sabermetrics well enough, but I think what is more important is that we don’t understand statistics. Sadly, in baseball, and in many other aspects of our life for that matter, statistics are taking over. It’s not that I don’t necessarily like statistics. I was schooled in them and have used them all of my life. I will use the numbers approach in evaluating things most of the time. What is very disturbing about our new statistical age is that statistics are not taught in school as a requirement, and therefore, they are not understood very well. We still push algebra in our school system. I think my children took algebra probably three or four times in high school and college, never to be used again. Yet every day we read a business or economic statistic, hear about statistical analysis concerning new health opinions, or get information from the latest political poll, but we were never required to take a high school class in statistics. This lack of general statistical knowledge has allowed some bad statisticians, conducting bad statistical analysis, to flourish. Meanwhile the public has a misunderstanding of the facts, stated as a statistic. This same problem has created confusion for baseball fans, and it has allowed baseball stats to be misleading and confusing. Maybe it isn’t just confusion. It could be that the saber experts assume their audience understands statistics better than they actually do. Perhaps the sabermetricians know better, but some have let their enthusiasm drive them to overstate their case. Or maybe, in some cases, these stats are purposely meant to mislead us, because the person writing the article has a position they are defending, and so they have manipulated the data to make their point. Are some of these sabermetricians this sinister? Maybe the debate has just become as ugly as a game brawl, and intentions aren’t really known.

The Saber Approach

Statistics describe something in the past. They do not predict the future. This is the crucial point that I need you to cement in your thinking. A number, a statistic about a baseball player for instance, simplifies our description of that player’s past performance. It does not tell us what his future performance will be. We can try to infer from this metric what he may do in the future, but we really don’t know. Sabermetricians may assert, and some of them do so with incredible confidence, that they know what a player can do, and will do, in the coming year. But if they were true to the math and the definitions within statistics, they would confess that correlations do not determine cause and effect, and therefore, they cannot predict results.

True, there is a probability that the player’s past performance will be a strong indicator of their future performance. But a strong indicator does not guarantee an outcome. Also, strong is a relative term. What is considered to be a strong indicator in projecting outcomes for baseball would not be acceptable in the medical field. If a new medical drug were able to cure an illness two out of three times, but it had adverse effects one third of the time, it wouldn’t be allowed on the market. Nor would these probabilities hold up in the physical sciences. We want our scientific proofs to be correct as close to 100% of the time as possible. This is what we consider a cause and effect relationship. In many cases, we’ll see that these baseball correlations aren’t much of an indicator at all, even at the level of two out of three times. Yes, showing that a baseball metric is related to winning games two out of three times is a significant correlation in sabermetrics. Given these weak relationships, another person’s opinion of a player’s future performance should be acceptable, even if it comes from an old baseball man and is not based on the new sabermetrics, but instead on the intuition that comes from experience and skilled expertise. It should not be dismissed out of hand or disdained as if they were a rube.

Let’s look at the case of Derek Jeter. Jeter plays shortstop for the New York Yankees and has had a Hall of Fame career. He has a lifetime batting average of .313, which puts him in the top 75 players for lifetime batting average among all players in the history of the game. With well over 16,000 major league ballplayers in MLB’s stats books, this is a pretty impressive accomplishment. He is also one of the top 10 hitters of all time, with over 3,300 hits to his credit – so far. In the 2010 season, at 36 years of age, his batting average fell, dropping all the way down to .270 for the season. Sabermetricians, and many in the New York press, were sure that he was in a steep decline. After all, Derek was now very old, in new baseball terms. One of the early findings of sabermetrics was that peak performance for a baseball player comes at 27 years of age, with an inevitable decline as the player continues to age. And so Jeter was an old ball player who needed to make way for younger talent. He was no longer worth the money they were