I’m not the smartest guy on the planet and this i know. But I also know that I’m not the dumbest either. I used that to preface the following statement: “I don’t like this statistical revolution that’s going on in sports. I don’t like it because I don’t get it.” If you’re like me, you long for the days when scouts were overweight men with grease stains on their un-tucked shirts, juggling a hot dog, radar gun on both arms and a book with a pen on their lap. Those guys have been replaced by young Harvard grads in their Docker pants and polo shirts, laptops in hand with radar guns.
The latter are the new wave of front office execs that are all the rage in sports today. These math majors have revolutionized the sports data collecting field over the last 20 years. Their godfather is Bill James who writes several influential books detailing the metrics with which he values a player and how he stands to do in the following season. His opinion is so well respected, his numbers in many circles are considered bankable. Let’s forget for a second that twenty years ago and even while I was growing up, the initials that every baseball fan had to look at were HR’s, RBI’s, and BA. But nowadays, with the advent of this new scientific model, OBP, OPS and OPP* have all made the leap into baseball statistical commonplace vernacular.
*= What, you thought I was going to stay away from the obvious Naughty by Nature joke right there? C’mon.
I suppose that it has to do with one indisputable fact that we must remember about sports in general: its more business than ever before. Teams have reacted to lagging ticket sales and fewer profits by being creative in how it spends its money. If a team has a specific set budget, they may be limited in who they give their money to so they have to figure out a way to assess a player’s value and somehow find a player who shows promise to do better than what his statistics say he did the previous year.
Ever year, the waiver wire is chock full of guys who couldn’t hack it in the current year. But teams that are wise enough, have made it a business to chart how the player did in context. Did he hit the ball hard? Was he hurt with bad luck? Was his defense bad because his team was putting him out of position? Does he have a knack for getting on base? Perhaps the player was released for personal reasons. We’ve seen guys who have major ability brought to other teams purely on that fact alone. One team’s trash is another’s treasure.
But each year, figuring a way to find this player is difficult. Because teams have embraced this new technology and new way of thinking, most who were on the cutting edge have had to find newer ways to evaluate players so as to remain in the advantage. Whereas some teams stick to old school methods, more and more teams are beginning to understand the nature of this new style and are looking to hire these highly trained guys who normally would have taken high paying corporate jobs to come and be baseball scouts and train them to be execs.
Its something that was unheard of back in the day where scouts did have technical reasoning for their ability to pick players but really trusted their instinct on most of their decisions to draft. These young guys relate everything back to science and can graph a player’s performance following a specific model but all relating to how a player does in a current fixed environment. As situations change (for example, managers, team philosophies, and teammates), it becomes more difficult to understand.
This past weekend the Sloan Conference was held in MIT. What’s the Sloan Conference you ask? Think of the biggest collection of geeks that have nothing to do with Star Wars, fantasy movies or Star Trek. Yes, its the Sports Analytic Conference. Its a place where the leading minds in technical analysis of sports come together to discuss the future of their field in sports. Where does it fit in and how do things like the Fielding Bible and Plus Minus all fit into the bigger puzzle of forecasting a player’s ability and the more pressing need to assess value to a player based on it.
To understand the growing belief in sports analytics one must understand that the business of sports has played a major role in its growth and popularity. Teams have looked for ways to grow its team value by putting out a better product on the field and sporting a smaller payroll. That is the goal of every team. Small payroll, big winner, big money at the box office. Teams believe its all related. I believe in that too. I mean the second part. I believe that good teams are what people pay for. Fans will not pay to watch crummy teams. Ask the Mets who are really going to be dependant on a very fast start by their ball club to offset sagging pre-season ticket sales.
I do however believe that no matter what, you have to pay for talent. Teams like the Twins have done well to build good teams and certainly you can win on a very small payroll (ask the Florida Marlins in 2003), but they are the exception and not the rule. When the Yankees won last year, it was as if a great burden had been lifted from that franchises’ shoulders. For all the bickering by their fandom about where all the money was going, their $400 million spending spree in the midst of one of the greatest economic downturns in memory resulted in a world championship which is the expected end result for any team out-salarying everyone else by almost 40 million every year. But the great task remains to lower payroll and remain competitive at the same level.
Thus you have guys like Johnny Damon leaving and Hideki Matsui leaving and bringing in Curtis Granderson who’s on a much friendlier long term contract while being a young commodity. They have some of the highest paid athletes in sports but one of their biggest hopes is that one of their two burgeoning aces Phil Hughes and Joba Chamberlain will emerge. They are breaking the bank (in our pedestrian terms) as thousand-aires surrounded by guys who have enough salary to keep Haiti afloat, for a couple of years.
But the task becomes even bigger for teams like the San Diego Padres. They are about to let you in on one of the worst kept secrets in all of sports, that’s the impending trade of Adrian Gonzalez who figures to price himself out of San Diego. While the team has his rights at very reasonable salaries, in two years he will be one of the most sought after first basemen because of his age, and his ability. San Diego with its limited payroll flexibility will not be able to afford to keep its hometown hero despite doing everything it can. Of course on the other side there’s the Minnesota Twins who hope to have a contract done for their own hometown product Joe Mauer. Mauer has stated that he wants to stay a Twin and not go into his free agency year where the Twins will be easily outbid by the Yankees, Red Sox, Mets, Angels, and perhaps the Dodgers. These major market teams with their own networks to boot have money coming in at a rate that other small market teams can’t compete with. While the League front office can’t have an opinion on things of this nature, they quietly hope that Mauer resigns with the Twins so they can show all the doubters that their current salary structure works.
That is the end result of sports analytics. To do a better job at analyzing bargain bin athletes and finding those guys in someone else’s trash bin that can be productive in a different environment. But how does evaluation work? Well, its too technical for me to even explain. But now when we consider who gets drafted and who doesn’t the obvious test subject is Tim Tebow.
Tebow’s college resume speaks for itself so no need to rehash his Heisman Trophy candidacy, championship year and all other things that make the legend of Tim Tebow as wild as a modern day Davy Crockett. But Tebow is projected as a third round, or fourth round talent. Now, there’s no doubt that his effectiveness will be limited unless he learns a new delivery. His old one is not pro-friendly because the mental clock must be shortened by a couple of seconds on that level and he will get hit by the time his arm gets to his shoulder. He’s not being viewed as a QB though he wants to be seen as one. His value may be as a half back or TE where he would be a physical specimen who with time can learn the position. So how come he can’t learn the QB position? It can’t be a matter of intelligence. He’s a work in progress on any level you think of it but many consider him to be a pro player. His attitude and all other mental aspects of his game are not questioned, its his ability to pick up the pro league system that is being questioned. All based on metrics that see him not producing on that level.
The danger with a player like Tebow is that a GM can look like a fool if he pays too much attention to the facts of just his rawness and he can be smart to avoid him if he ends up being what they say he will be; at best a marginal back up QB in the NFL. However, with Tebow there’s an area of intangibles that’s impossible for these stat geeks to compute and thus makes the Tim Tebow career one to watch. The jury won’t be out till five years from now when most assume that a player will show he can either get better or he just doesn’t get it. Tebow will be given that opportunity but a person of his stature and status going later than a second round is a shame in my opinion. He may not be your starting QB now, but his work ethic and his drive make him a coachable asset that most would love to have. That has to have some kind of value in any system of evaluating a prospect. Most don’t.
Which brings us to the Combine in the NFL. What do running 40 yard dashes and broad jump and high jump do than leave you with very specific idea of how athletic they are? Great they are athletic, but can they handle a pro style defense or offense? Do they show an ability and willingness to learn and have a work ethic? Those are things that are for me, more important than anything a math/stat geek can tell us. What do sports stat analyses show us that our own intuition cant see.
Used to be that we could see a Usain Bolt and describe his speed as that of a cheetah and say something to the effect that while he looks like he’s running in slow motion he’s running faster than anyone else and be able to go on that alone to assess it. Sure its not accurate and may be subject to a specific view but it is good enough to describe that Bolt is the fastest man alive. We can see how quick money Mayweather is with his blows and ability to dodge. We can see with our own eyes the decimation of Mike Tyson’s skills. All in context.
For me sports analytics have a place but not one of highest importance. Human intuition is often a better friend than any analysis of how a batter hit in specific conditions in a specific day. Our intuition guides us. Both are capable of making mistakes but the intangibles are still something that need to be weighted heavier than I feel is being weighed today. The Combine in the NFL is going to convince some exec that a player is worth taking in the first round despite the fact that there are several red flags about his character and work ethic dating back to college days. The hubris that some of these front office execs show in their belief that they can right a certain player within the locker room is comical. Some guys like Vernon Gholston, who some were questioning his work ethic dating to his college days, are red flag guys who front office execs ignore because of faulty projections.
I’m not saying that analysis is bad, in many cases its good and it shows its effectiveness. No team outside of Chicago was going to sign Julius Peppers to that contract. No team had more of an urge to sign him since they didn’t have draft picks in the first two rounds this year because they traded for Jay Cutler last year who again was subject to wonderful projections in Chicago but fell flat and had a terrible first year in the system. Something most could’ve seen coming if they had just looked at their roster. They had no proven weapons outside of Greg Olsen who had a down year as well.
The analytics business is a growing one but one with many faults. Teams that rest their entire payrolls on it can only expect moderate success. I haven’t seen Moneyball expert Billy Beane win any championships employing that system. He’s willing to risk oft injured Ben Sheets will be a valuable commodity based on expectations he probably got from some machine. One can assume that the Mets with their boat load of injuries will automatically be better because they get all their superstars back, but there’s no certainty to that. Its possible that even with a healthy line up, they could still end up second fiddle to Philadelphia who admirably stuck to their principles and resisted the urge to forego their payroll and future in order for one or two years of glory.*
*= read absolutely relieved!
Much of sports analytics is based on the principle that things can be quantified to a certain percentage point of accuracy and yet that percentage point is never 100. The projecting is based on certain things being the same all the time which is impossible. A point was made during the Sloan conference about it being impossible to judge quarterbacks because you would have to have a QB throw in the same system over ten years to the same recievers. Think about the likelihood that a reciever and a QB running the same system with a coach who’s liable to call the same play at the same situation every time over a ten year period is? Can I say negative percent?
Even then the certainty of those stats is impossible to negotiate. My point simply is that statistical analysis has a place in sports but the more teams employ it, its clear that its for one purpose only. To cut payroll and put a competitive team out there. Not a championship one. Can a team that employs this system win a championship? Sure. But the probability of that is very hard to calculate even for these sports geeks.
acebook
witter
The Lazy Eye of Sports 