“Numbers never lie.” This mantra has become the all-encompassing truth that business executives, government technology leaders, security analysts, academic gurus and even sports historians universally accept.
Experts tell us to focus on “just the data.” Acceptance into college, current salaries, future promotions, global investments, political campaigns and even our selection of sports heroes are dependent on a few all-important statistics. In some cases, a small percentage can mean the difference between world-class success and embarrassing failure.
Of course metrics matter and we need trustworthy scales to measure our progress. Who can argue with keeping track of where we've been, where we are and where we want to be? Whether improving the assembly line, balancing the checkbook or losing weight, good numbers are needed.
But while there is no doubt that compiling reliable statistics is an essential 21st century activity for business and government, what if the right numbers tell the wrong story? What if you get the right answer to the wrong question? Or, what if “the system” is busy pumping out metrics that reconfirm what you wanted to hear in the first place?
Mark Twain once said: “There are three kinds of lies: Lies, damned lies, and statistics.”
Our New Normal in Numbers
Over the past century, American management has fallen head over heels in love with numbers. Computer-driven scorecards, metrics and dashboards are red hot within business and government right now. More than that, the new trend is to comb through terabytes of “big data” to find that key indicator that accurately measures the past and can foresee the future.
Political leaders make strategic campaign decisions based upon swing state poll results from a single demographic group. Will our next president really depend upon the opinions of white, middle-age, middle-class women from the Midwest? Other analysts and campaign experts predict that the November Presidential election will hinge on one key number: the national unemployment rate.
But can we trust these “all-important” numbers? Some say no. As quickly as the unemployment rate is announced each month, we are told that the “real unemployment rate” is much higher. Experts offer very different numbers when everyone who is looking for work is counted. If you add in those who stopped looking for a job, have a job but are underemployed, or include the normal growth of the labor force, the percentages change radically.
The same tendency to love particular statistics is seen in business and technology. Whether measuring a company’s future prospects, the performance of individuals or the likelihood of a project’s success, executives strive to boil things down to a few all-encompassing numbers. While it’s true that dashboards or “balanced scorecards” often contain multiple metrics from different aspects of the business, we are told to abbreviate as much as possible.
For example, we are pushed to give an immediate verdict based on the combined numbers: Is he a winner or a loser? Is the project red, yellow or green? What will the return on investment (ROI) be? How much did she sell last quarter? Or, what’s the stock’s price to earnings ratio?
And the numbers tell all. Management typically does whatever is necessary to boost, analyze and predict future metrics. In extreme cases, key numbers are even unethically or illegally manipulated – which is an important topic for another day.
Could there be situations where the correctly calculated numbers mean something different than experts or analysts think? Is the future of the economy, your child’s academic success, your friend’s business franchise or your next promotion really about one or two all-important metrics? Even when this is true, are there exceptions we should care about? Can numbers be twisted inappropriately or given some power they never should have had in the first place?
Or, what if the “right numbers” bring the wrong overall results? Might other factors unexpectedly change the outlook for your favorite stock? For example, could a Greek exit from the Eurozone cause a global recession and a drop in the value of your portfolio, regardless of how a key metric looks at your favorite company?
More specifically, public and private sector technology leaders need to ask: Can new innovations be measured successfully by accurately analyzing old metrics? Are there compelling examples of seemingly poor traditional statistics that actually led to better results than expected? I think there are.
A Decades Old Example - Sports
To spice this topic up a bit, I’d like to examine an area of life that lives and dies by the numbers – sports. If you take the total annual spending of all the global sports, we are talking about over $620 billion spent annually. From attendance at games to television viewing share, from baseball box scores to quarterback completion percentages, no business in the world lives and dies by the numbers more than professional and major college sports.
By Dan Lohrmann
- What do business, government and sports have in common: Numbers
- What Tebow-mania can teach us about measuring success
- Don’t let bad numbers fail your good project
For example, multi-millionaires are made with Major League Baseball (MLB) batting averages over .300. But if you bat less than .250 for long, you’re probably heading back to the minor leagues soon and may struggle to make ends meet.
Growing up in Baltimore, I was regularly treated to some great MLB seasons. The Baltimore Orioles were stacked with four twenty game winners in 1970, along with superstars like Boog Powell, Frank Robinson and Hall of Fame, gold-glove winning third baseman Brooks Robinson.
Brooks was known as Mr. Clutch because of his late game hitting heroics. No pitcher wanted to face Brooks with the game on the line and runners in scoring position. As a kid in the 70s, Brooks was the one superstar all my friends imitated. Yes, he was also called the “human vacuum cleaner” because he played great defense at third base, but I always wanted Brooks up at bat at the end of close games.
Many years later I discovered that the Hall of Fame 3rd baseman had a lifetime batting average of a mediocre .267. Brooks had several sensational seasons, great games and plenty of defensive awards. But his overall batting average misrepresents his value to the Orioles and his amazing career story.
If you look closely, you can find many other statistical anomalies in sports history. There are stories of men like Kurt Warner, who was stacking grocery shelves before he became the MVP in National Football League (NFL). Warner is probably heading to the NFL Hall of Fame, but you never would have predicted his results based upon his earlier numbers.
There is the evolving Josh Hamilton baseball story which defies traditional baseball logic. He came out of high school with incredible potential and off the charts stats, only to run into trouble with drugs and other problems and drop completely out of baseball. Later, after being written off by everyone, his life took another dramatic turn - this time for the better. He was recently selected as the top player in MLB and given an ESPY Award. Josh’s life story is now being made into a movie.
But as we head into the new NFL Football Season in September, 2012, no statistic in sports in America today gets more attention than former Denver Broncos starting quarterback and current New York Jets backup quarterback Tim Tebow’s passing completion percentage. What can we all learn from Tebow's stats?
Later this week, in part two of this three-part series of blogs, I’ll dig deeper into this topic and address what government, business & technology leaders need to learn from Tim Tebow’s statistics.