The most intriguing rivalry in hockey these days isn't Penguins v. Capitals or even the long-running Bruins-Canadiens saga.
Those have got nothing on the bickering waged daily between the analytics (i.e. fancy stats) crowd vs. the old guard (i.e. the eyeball test). Over the last few years, this ideological divide has been the NHL's fiercest battle, with both sides firmly entrenched in their respective corners. There seems to be no room or appetite to bring these two divergent parties together.
In many ways, the analytics community and the old guard have become almost caricatures of themselves. One side seems to believe that the game is a sterile mathematical equation that can be solved by heat maps and raw shooting percentage data. The other argues "watch the game," leaning heavily on traditional statistics like plus/minus and gut instinct. Think Don Cherry, Mike Milbury and every network TV analyst.
But if we're being honest, Cherry and Co. have been taking a few roundhouses to the chin in this debate as the league has evolved and grown further away from its "if you can't beat 'em in the alley, you can't beat 'em on the ice" roots.
If you wonder where the physicality and violence has gone from the game, look no further than analytics, which views hitting as a wasteful pursuit that runs contrary to the objective of puck possession and thus less shot attempts, less goals and less wins. The game is relying more on data-driven analysis and has given way to new-school thinkers like Kyle Dubas, the Toronto Maple Leafs' new 32-year-old general manager, and countless others who are now in positions of power in nearly all of the league's front offices.
The Penguins are certainly no exception.
Jason Karmanos has been on the front in the battle and the Penguins are as responsible as any organization as to how the game is being played stylistically on the ice due in part to analytics. Few teams place a greater premium on puck possession and a lightning-quick transition game like Pittsburgh, winners of two of the last three Stanley Cups.
Of course this progressive, new-look approach runs counter to the image of Jim Rutherford. The reality is quite a bit different, though. He is not as old school as it would seem at first glance (he even texts). At 69 years of age, he's the NHL's second-oldest GM, behind only the Islanders' Lou Lamoriello, who was ousted this spring in Toronto in favor of Dubas. Rutherford played in the rough-and-tumble 1970s, but he's also not static or set in his ways. He's grown with the game, too.
The use of analytics helps explain how the Penguins have been able to consistently ice a team with top-end talents surrounded by a younger, smaller (and cheaper) supporting cast. In that sense, they are the model franchise in today's game.
But even their use of analytics has its limitations.
That was evident on July 1 and the start of free agency when the Penguins signed 31-year-old defenseman Jack Johnson to a five-year, $16.25 million contract. To date, Johnson is the biggest free agent signing during Rutherford's four-year tenure as almost all of his acquisitions have come via trade. There's a reason why he's not called "Free Agent Jim," you know?
But Johnson's signing came after last summer's head-scratching trade for Ryan Reaves, who is likely to go down as the NHL's last true heavyweight enforcer. Even as the game has gotten away from fighting dramatically, the Penguins dealt for Reaves, only to subsequently flip him to Vegas at the deadline as his ice time was dwindling.
But back to Johnson. Of all the players that Rutherford has brought on board and had success with as a reclamation project, no player has been more polarizing from the outset than the former third overall pick in 2005.
Obviously a large part of that is because Johnson has never been able to live up to the massive hype that preceded him into the league. He's been a very good defenseman but never a game-breaker. His underlying analytical numbers, however, would indicate he's been far worse.
During his 11-year career, his shot-attempt ratio has never been better than 50 percent and his Corsi For Percentage, a barometer which measures the ability to drive possession, is 47.3. That's not good, at all. By comparison, Matt Hunwick was lowest among Penguins defensemen last season at 48.3.
Troubling to many was that Johnson was a healthy scratch at the end of last season and through six playoff games on a Columbus team that bowed out in the first round. Johnson had been a leader with the Blue Jackets, though he requested a trade in midseason due to his financial situation. After the Jackets acquired Ian Cole at the deadline, Johnson was effectively passed up on the depth chart on a defense corps that included Seth Jones, Zach Werenski, Ryan Murray, David Savard, Markus Nutivaara and Cole, the former Penguin.
So given all that, why would the Penguins sign Johnson?
In this case, the eyes had it. He passed the eyeball test of not only Rutherford but also Mike Sullivan and Sergei Gonchar, who viewed his final three games and liked what they saw (Johnson was a plus-3 with an assist). Old school scouting prevailed over what Johnson's Relative Fenwick For Percentage at 5-on-5 was.
“We use analytics as one of the checkpoints, and we use our people who go and watch the player,” Rutherford said. “We try to separate things out as to why the analytics say one thing and why the guys say one thing. Before we signed him, the coaches were looking at video all the time and there was a real comfort level there.
“None of this is foolproof. As humans watching players, we make mistakes on them sometimes, and the analytics are not always accurate.”
As useful a tool as analytics are, they do not yet measure intangibles like leadership and character, two of the selling points on Johnson. They do not measure clutch because, well, the analytics community is not a fan of the concept.
But the fact is that Johnson has been a 0.35 points per game player over his career in the regular season but that number jumps to .93 in the playoffs (21 points in a relatively small sample size of 23 games). Still, that average is better than Erik Karlsson's .77.
Regardless of what Johnson's other stats say, that's a significant reason the Penguins signed him. As Sullivan has stated ad nausea, the strength of his team lies in its ability to elevate its game in the postseason (see Guentzel, Jake). For the Penguins, the postseason is all that matters. Flaws and all, Johnson is a considerable upgrade over Hunwick or Chad Ruhwedel.
On a team that has a legitimate chance to contend for the Cup, Johnson surely won't lack for motivation to win and to silence his many critics.
We shall see how this plays out in the months and (five) years to come but the Johnson signing will be an interesting case study in the ongoing debate about analytics vs. the eyeball test.
