November 8, 2002
Starvation Through Force Feeding
Despite what you may have heard or read over the past several years, the information age has yet to actually arrive in business. Not a single company in the Wilshire 2000 has done anything near optimize how their organizations acquire, process, generate, and use information. Hundreds of billions of dollars have gone into investments in information technology in enterprises of every shape and size throughout the world, but overall productivity gains have been marginal.
The issue doesn't really have much to do with hardware or software. The computer you're reading this on is probably hundreds of times more powerful than you actually need for your purposes, and the same is true in most organizations. Product researchers know that even "power users" of software like Microsoft Word or Excel typically use less than 5% of the capabilities of those products. For more complicated and specialized tools, usage is slightly more intensive, but in most cases, the capabilities of the tools we use to manage information are not an obstacle preventing us from using information to do our jobs more effectively.
Most organizations with information problems, including MLB clubs, don't suffer from a lack of information. Quite the contrary--they have more information than they can effectively use to make better decisions. For several years, organizations of all size and purpose have been geometrically increasing the amount of information they gather. If there's any doubt about whether a piece of data will be useful, an organization will probably collect it. The marginal storage cost is generally low, and it can always be ignored later if there's no use found for it. Furthermore, the risk of not collecting all available data is high--if it turns out later that the information is crucial to saving or making a bunch of money, and you don't collect it, it's a very painful missed opportunity.
The end result of all this information is predictable - overload. One former GM had this to say on the subject:
"Occasionally, a fan who knows all the numbers will ask me if I was aware of some statistic or other from someone's minor league record. He thinks that I don't have some piece of information that he has. Of course I have that piece of information available. I've got everything the public can get, plus current and past scouting reports, as well as his agent dogging me every chance he gets and stretching the truth every which way to get me to give the kid more playing time and money. Fans really think that we can't go to Baseball Prospectus or ESPN.com and read the same things they can? There's no question we have all that information. But we also have 100 other pieces of information, presented by a dozen other people, each with a different idea of what's important. And we have what we know about the person beyond the numbers.
So when we get an email like this…
Gary, how come the Orioles signed David Segui? Segui's not difficult to replace, and the Orioles aren't going to be in contention while he's around, so why pay that kind of money for a first baseman that might be average? Don't they know what he's going to do? Do they not read his stats for the last few years before inking a contract?
Well, they of course read his stats before inking him to a deal, BY. It's not as if there's no one in the Oriole front office who had access to a STATS Red Book at the time. The problem is that there's so much information, and the information that's presented is usually presented by someone with an agenda, be it an agent, a staffer who thinks a guy is going to break out, and information can be selectively presented or withheld to advance an agenda.
Just like in politics and every other business in the world, facts are advocacy.
Don't make the mistake of believing that just because some GM makes what every baseball analyst universally decries as a rotten signing, that it's because of a lack of information. It can be, but the issue isn't usually the information, it's the awareness of the decision maker about the importance and predictive value of the available information. Let's look at a signing made by the Arizona Diamondbacks a few years ago that I mentioned in a previous column…
At the end of 1997, the Arizona Diamondbacks signed Matt Williams to a 5-year deal worth $45 million, weighted slightly towards the back end. (Williams is scheduled to receive $10 million during the 2003 season.) What information was available?
Information presented to Diamondback GM Joe Garagiola, Jr. is always presented by those who wish to inflate the importance of the information that advances their agenda. Williams' agent, Jeff Borris, has a fiduciary responsibility to present his client in the best possible light. So where I provide you with this selected information:
Borris might say this: "Yes, Matt's numbers haven't been where they once were, but look at the injury record. He was fighting some nagging injuries during that time, and your own scouting reports now show that his range in the field and bat speed have returned. You can expect his numbers to bounce back up to where they were in his prime."
The verbal judo of using the injury record (a negative) to turn the performance record (another negative) into an expected positive (expected increased production) is just an example. I presented the information that I thought was important to make my point, and I'm certain that Borris presented the information he wanted to make his point, in the mode that was most advantageous for his client. I wanted you to think about performance forecasting and financial responsibility as a way to make a club more successful and profitable so you'd be entertained, return to the web site, and buy books. Jeff Borris wanted to make his client happy through a combination of increased wealth and more enjoyable living conditions. (One of us is probably not using our time optimally.)
It is not enough to have the right information when making a decision. You also need to identify what the right information is, and incorporate it into the decision making process properly. That's not as easy as it sounds, and it's why it's going to be a long time before every front office "gets it." There is an entire system of people whose livelihood is dependent upon making sure that people with the ability to make multi-million dollar decisions don't "get it." Front office professionals need to have a system of verification of their data sources over time. They need to track which information sources end up closest to reality, so they can weight those sources more heavily in the future. This kind of feedback loop is critical to any enterprise, but it's also one of the easiest things to overlook when resources become scarce.
Baseball is filled with competitive, confident people by nature. These people tend to have strong opinions, and the ability to hold onto one's opinions in the face of overwhelming evidence is a human strength. Add in the fact that sometimes, the evidence isn't as overwhelming as many think, and we're likely to see many more years of multi-million, multi-year contracts for Pat Meares, Rey Ordonez, Derek Bell, and hundreds of others. Some will bear gloves and play on the field, and others will wear suits and ties and look out at the field and wonder why their team's not doing better. The answer isn't necessarily more information, but instead better selective usage of what's already available. And the willingness to try and test new ideas.
"You can make mistakes with stats just as easy as you can with your eyes", says another GM, "so why not go with the guys that have experience? Any stat can be made to say anything. I need people who can watch a field full of ballplayers without numbers on their backs, and tell me who the great ones are going to be."
A different kind of information yet, already with a perceived high level of importance.