Keith Law’s new book, Smart Baseball, isn't about process batch manufacturing, but it may as well be. I picked it up yesterday and was stunned to realize the degree to which his insight on the national sport overlaps with mine on Electronic Batch Processing . Have I mentioned that a physicist doesn’t recognize artificial boundaries between things? (Everything is connected to everything.) When I first got deeply involved in process manufacturing I was surprised to see the durability of ancient customs: PLCs have been around since the 1970’s. And vendors of this technique have adopted an even more antiquated – medieval – notion of the guild: only the anointed may install, program, modify and fully exploit its power. Imagine if Henry Ford had that silly, high-brow idea and applied it to driving your car. Hey, you can propel a ton of metal at 65 mph, and you can’t make a few hundred pounds of emulsion on your own? Does this make any sense?
Law describes his book right on the cover: “The story behind the old stats that are ruining the game, the new ones that are running it, and the right way to think about baseball.” Let’s share a few ideas to underscore the right way to think about batch manufacturing. And, yes, I’m being symbolic. (Quarks don’t really have flavor, nor are they charming.)
Stolen bases, like total number of batches produced … the more the merrier? Ricky Henderson, who entered the Hall of Fame as soon after retirement as he qualified, stole more bases than anyone else in the history of baseball (1406). By contrast, another baseball great, Tim Raines didn’t make it to Cooperstown until his last possible year (23), having stolen a mere 808. But pause for a moment to consider that although Raines was caught stealing 146 times, Henderson was snuffed out more than twice as many times - 335. Ouch that hurts. And it’s hard to focus only on the successful steals when you take into consideration that percentage of failures:
Henderson – 19.2%
Raines – 15.3%
Suppose those failures were “bad batches”, your bad batches. The cost of a blown steal is a potential run. What is your cost for a blown batch? Why have any?
Then there is the save, a totally fabricated concept sprung, like Athena from the head of Zeus, out of the creative mind of a sports writer. Do you know any sports writer statisticians? Me neither. Oh so many games are mismanaged because the team’s best pitcher is reserved for the save. The effect of “save”, as Law documents exhaustively – and amusingly – is often a distortion of the game, which needs to be played to win, using your best pitcher when needed, not at the end of the game to collect a save.
Rework is worse than a save. “Oh, we don’t have waste. We can always do a re-work.” And what, exactly, is the cost? Zero? I don’t think so. The extra manufacturing time spent, the need to use the skills of experienced techs and managers, the delay of product, the mess-up of schedules … all these have a real cost as well as an opportunity cost. What else could you be doing with your plant facilities, your most skilled technical people and your own time? It’s obvious that playing to win in the batch manufacturing game is having the right tools to do the job right the first time. As the Tiffany ad said, “the best is the cheapest in the long run.” And when the best – the means to create perfect batches - is available inexpensively, the message sparkles like a Tiffany diamond.
To me, the most interesting one of all the old statistics is Fielding Percentage simply because if the player doesn’t get close to the ball, he isn’t charged with an error. Happy with the batches you’re asked to make, the ones you can nail with no mistakes? Consider the high-value-added, highest margin batches you don’t make – and wouldn’t touch because you’re playing for a high success rate. The greatness of Ozzie Smith was not his fielding percentage, but the number of putouts and assists he made … these are the plays others couldn’t make (or didn’t even try to make). What products might you be manufacturing if you felt you couldn’t fail?
And why does baseball measure the rpm of a curve ball rather than stick to the old “seems to have good rotation”? They want to know for sure, to calculate the 2700 rpm hurler from the 2400. Because what you measure, you can control. It’s about the money, the money plays and the money players … Shouldn’t that be the rule in process batch manufacturing too? Measure and know for sure.
With systems built to put you safely in the driver’s seat, why depend on the old slogans and excuses? There’s an open road to the World Series for batch manufacturing. Consider taking it.