Reds’ Riggleman gets it done by going with flow, not data

CINCINNATI – Jim Riggleman is 65. It’s a neither Here nor There age in real life. In Major League Baseball, it’s edging into fossil-hood. Naturally I went to Riggleman to learn about Bayesian statistics and their use in baseball. “What?” said Riggleman. Bayes-ian statistics.

“Spell it,” he said.

B-A-Y …

I was surfing the ’Net the other morning when I checked in at Fangraphs .com. FanGraphs is a baseball statistics website. It is to baseball numbers what Coke is to sugary sodas. Supreme and supremely addictive.

The Reds had placed a want ad of sorts on FanGraphs. They were looking to hire a “data scientist.” One of the qualifications — along with “3+ years experience in a computational field” and a “strong knowledge of predictive modeling” — was “experience with Bayesian statistics.”

In 2018, there’s nothing unique about that. The game has been (slide) ruled for a decade by people whose love of spreadsheets knows no bounds. Your team is in deep these days if it doesn’t have a few keen minds in its Bayesian statistics department. We say this with all due respect, knowing a numbers revolution when we see one.

All Bill James’ chillun’ are spreading the news. Get hip or get run. But Bayesian statistics … is that a new way to define fielding success? Is it the new BABIP? It was only last week I learned that BABIP was not a character from “Lord of the Rings” but rather Batting Average, Balls In Play.

I looked up Bayesian on Wikipedia: Bayesian probability is an interpretation of the concept of probability, in which, instead of frequency or propensity of some phenomenon, probability is interpreted as reasonable expectation representing a state of knowledge or as quantification of a personal belief. Yeah, OK.

Sam Grossman is a Reds assistant general manager. He’s also their top Numbers Guy. When analytics were young, Grossman and current Reds GM Nick Krall were mining stats for meaning.

“It’s conditional probability,” Gross- man explained.

Of course it is.

“Like strikeouts per plate appearance,” said Grossman. As he explained it, if the Reds are looking at a player “who has struck out eight times in 10 plate appearances, 80 percent, we’re probably going to believe (he is) closer to the league average of, say, 20 percent.”

So you’re assuming a player who K’s 80 percent of the time in a short sample is going to strike out 20 percent of the time over a longer time frame?

“It has to do with regression to the mean,” he said. “That sort of thing.”

The Reds have “eight to 10” people in their analytics department, Grossman said. Four or five have advanced degrees, in diverse fields, including kinesiology. “A few are on their way to doctorates,” he said.

How does someone with a masters in kinesiology help determine the efficiency of a starting pitcher, I asked.

“The methodologies you learn can apply to a lot of fields,” Grossman said. “Rest and recovery” of players, for example. “We’re just looking for smart, energetic people who do things creatively.”

Meantime, Jim Riggleman goes with the information flow. When you’re 65, an open mind is a big asset. He says the data guys don’t push anything on him. “They’re a little bit shy and reluctant because I could be their grandfather,” Riggleman said. “They have some decent baseball ideas. But I’m not being forced to run ballgames based on that. Some organizations are way heavier into it.

“My college coach was doing this back in the ’60s and ’70s. Batting orders, defensive positioning.” The new breed is “pointing out statistically what our eyes have been telling us.” Namely?

“Starting pitchers, even the good ones, get a whole lot worse after six innings or three times through a lineup. In five years, because of that, we won’t be spending as much money on starting pitching.”

Riggleman has brought a calm, wise head to the Reds dugout. The team is 4242 since he took over. His chances of being here next spring grow with every scrappy win. And so I had to ask:

“Do you ever anticipate saying at contract time, ‘I simply can’t work here anymore unless you get me a few decent Bayesian statisticians’?”

“No,” Riggleman said. “I really don’t.”