I went down a research wormhole last night and haven’t fully clawed my way out of it, so there will be nothing today. Have no fear. Shootout week will continue.
Measuring shootout effectiveness using fancy stats and why it should matter to NHL teams, including the Red Wings
On Monday, we determined that on average a team has 10 extra points* available to it per season due to the shootout.
The goal for NHL teams then should be to put them in the best position to gain as many of those 10 possible points total in the shootout.
How do they do that?
Well, the first step is to find a measure of success.
The problem with shootouts is there aren’t an equal number of opportunities for each team. The Capitals had 21 shootouts last year. The Predators had six. Because of these types of discrepancies, we need to deal strictly in percentages: win percentage, shooting percentage, etc.
I looked at a number of different stats to find the one that correlated best to winning percentage. The best one is a stat I’m calling Shooutout PDO or SPDO.
If you’re a new-school advanced-stat hipster like me, you already know what PDO is. If you’re not, you can read about it here. In short: it’s the addition of a team’s shooting percentage and save percentage. Thus, SPDO is the addition of a team’s shootout shooting percentage and shootout save percentage.
I calculated an SPDO for all 270 teams (30 teams a season for nine seasons) that have ever participated in the shootout and graphed it against that team’s winning percentage:
The R^2 value is .908, which is quite high. That means that just under 91 percent of a team’s expected shootout win percentage comes from SPDO. The other nine or so percent comes from other, undetermined factors (luck, etc.).
This can help us determine exactly how much improvement a team needs to make in order to gain more shootout wins.
Our average team plays in 10 shootouts a year, so a 0.100 increase (or decrease) in win percentage is roughly equal to one more shootout win (or loss).
Late one night about a month ago, I mindlessly clicked a link that I had favorited on Twitter. My browser went to a month-old Yahoo! story about some professor who had been teaching a class on Sabermetrics — “The Study of Baseball as a Science” — at Tufts University. He was now teaching the class online, for free.
I signed up.
It wasn’t so much the baseball as it was the accessibility to data analysis. Baseball was already too refined. I wanted to figure out hockey. I paid attention to Fenwick and Corsi and quality of competition and PDO and all the other stats provided by ExtraSkater.com but didn’t have the skills to venture too far into the world myself.
When I initially tried to get more advanced with hockey, it didn’t work too well. I ran into all the common pitfalls: How do you adjust for the differing quality of a player’s linemates, or who they’re matched up against? How do you break up such a free-flowing game that relies on teamwork into the individual one-on-one battles that is in baseball’s DNA?
On one particular night, I watched the whole hockey panel from the 2014 Sloan Sports Analytics Conference (video here), searching for clues. In it, Don Fishman, the Assistant GM and Director of Legal Affairs for the Washington Captials said something that made a lot of sense.
“We’re a fluid game, we’re a complicated game, that’s why we love our sport,” Fishman said. “Let’s stop being obsessed with trying to capture the entire game and trying to come up with some magic formula to capture that whole game, because we can’t. It’s a beautiful piece of art. It’s this creative piece of art with hundreds of colors in different ways. Let’s learn about discrete areas and focus on different areas of the game and let’s see if we can nail those. I’d love to see someone nail the shootouts. Come up with something.”
I’ve thought a lot about the shootout since then. I’ve searched for some analytical breakdowns of NHL shootouts and haven’t found much other than on whether to go first or second in the shootout (Conclusion: it doesn’t really matter, but this is an old study).
If you’ve seen something I missed please pass it along, but this lack of info shows that it’s time for me to take the walking stick. I will attempt to “nail the shootouts”, and no that’s not slang for something else.
The first step is showing they matter.
Sunday Red Wings rumor roundup: What it might take to land Mike Green and why Detroit might be down two assistant coaches
New feature! Every Sunday I’ll bring you a recap of all the legitimate Red Wings-related transaction and signing rumors from the previous week. I see all this stuff anyway. I figured I might as well share it with you.
• Danny DeKeyser chose not to file for arbitration, according to MLive.com’s Brendan Savage. He remains a restricted free agent and will have to agree on a new contract before training camp. DeKeyser and Tomas Tatar are the Red Wings’ only restricted free agents remaining from the NHL club.
• Washington Capitals right-handed defenseman Mike Green is the Red Wings’ top trade target according to Fox Sports Detroit’s Art Regner, who cites a “team source.” Green has a limited no trade clause, so the number of teams both on Green’s trade list and in the market for a defenseman could be small. Regner says the source indicated the Red Wings intend to move quickly.
Green, 28, is slated to become a free agent after next year. He’s got a cap hit of $6,083,333 for 2014-15. The Capitals indicated that they intend to keep Green, but MLive.com’s Ansar Khan says he’s “destined to be traded” because of the Capitals are near the salary cap ceiling and just signed defensemen Brooks Orpik and Matt Niskanen, giving them eight defensemen on one-way contracts. Khan says Green would be an ideal fit with the Red Wings as a right-handed, puck moving defensemen, but it might be tough to work out a deal, because Washington wouldn’t want Jakub Kindl and the Red Wings likely wouldn’t part with one of their young NHL-ready forwards.
The Free Press’ Helene St. James says that if a deal for Green was done, Tatar is the young NHL-ready forward most likely to go to the Capitals.
• As you probably heard, the Red Wings re-signed Daniel Cleary to a one-year, $1.5 million deal. What you might not know is that Cleary got a $250,000 signing bonus and a no-trade clause. He also gets a $1 million bonus if he plays in at least 10 games next year.
• Assistant coach Tom Renney is in the final stages of leaving Detroit to become the president and CEO of Hockey Canada, according to MLive’s Ansar Khan. That would leave the Red Wings with two assistant coach vacancies after Bill Peters left to become head coach of the Hurricanes earlier this summer.
Khan says the Red Wings have interviewed Tony Granato for one of their assistant coach positions. Granato has spent the past five years as an assistant for the Penguins but is without a job after Pittsburgh fired head coach Dan Bylsma this offseason.
This summer some people have earned Stanley Cup rings and All-Star nominations but there’s no honor quite as special (and quite as meaningless) as being named to the Detroit Tigers organization’s All-Name team.
Last year 25 men had their hard work rewarded by making the inaugural team. From Al Alburquerque to Jesus Ustariz they’ll all be remembered for their role in that first team. But time moves on. It’s time to name another squad.
Here is the 2014 Tigers organization 25-man, All-Name roster:
C Duncan McAlpine, Low A (First appearance) – McAlpine hasn’t played at all this year. It’s unclear why but the only two explanations are either injury or steroid suspensions and I’m going on the latter. He’s still on the Connecticut Tigers’ roster, though, so he’s eligible for this team and he earns a starting spot because he sounds like an excellent Canadian Mountie.
1B Giancarlo Brugnoni, Rookie Gulf Coast League (First appearance) – As an undrafted free agent out of Grand Valley State, Brugnoni’s got an uphill climb in the baseball world. But with a name like Giancarlo Brugnoni he’ll be able to glide through the general world.
2B Gregoris Hidalgo, Dominican Summer League (Second appearance) – Hidalgo deserves a better fate. If he played in a system that didn’t have as much depth he’d probably have earned an MVP in one of the past two years. Instead, he gets to be one of the few to earn back-to-back starting nods on the All-Name team. As long as he’s in the Tigers’ system, he’s going to be this team’s starting second basemen. Unfortunately he might not be here for much longer. He’s hitting just .181 in the Dominican league.
SS Eugenio Suarez, MLB (First appearance) – The last man cut from last year’s team, Suarez’s play this year earned him a spot on the All-Name team. Last year, I assumed his first name was pronounced like it was spelled: Eugene with an ‘io’ at the end. Then he got called up and the world learned that the letters and the sounds they produce have no correlation: ay-oo-HEY-nee-oo. Even Baseball Reference gave up, offering only the pronunciation of his last name.
Two weeks before the NHL free agency began, the Red Wings signed backup Jonas Gustavsson to a one-year extension. The move immediately looked bad. As I pointed out at the time, Gustavsson was below-average last year. Six backup goalies were slated to hit free agency that performed better.
When the team disclosed the money, the signing looked worse: $1.85 million. Gustavsson got a $305,000 raise.
Now, a week into free agency, the signing looks terrible. Four of the six goalies slated for free agency actually hit the market. All have already signed elsewhere. Looking at those contracts, it’s clear the Red Wings paid a premium for consistent mediocrity.
Thomas Greiss signed a one-year, $1 million deal with the Penguins. Chad Johnson got two years at $1.3 million per season from the Islanders. Al Montoya went to Florida on two-year contract worth $1.05 million per season. Justin Peters will make $950,000 per season in the next two years with the Capitals.
The Red Wings couldn’t have been concerned about being priced out of the backup goalie market. Heading into free agency none of those mentioned above made more than $750,000 and they all signed for at least $550,000 less than Gustavsson.
Maybe they were concerned about striking out and being left with nothing. Detroit isn’t an attractive spot for backup goalies. Jimmy Howard is the clear No. 1 now and Petr Mrazek is the No. 1 of the future. There’s no chance to earn a starting spot and no hope that there will be in the future. This might’ve bee a problem for the Red Wings — except they had $1.85 million budgeted to give to a backup goalie. Thomas Greiss would’ve come to Detroit had the Red Wings offered 50 percent more money than the Penguins.
Maybe the Red Wings were concerned about giving a backup more than a one-year deal. Set aside the Greiss signing and the we’ll-offer-you-more-money-for-one-year angle. What’s wrong with giving a goalie a two-year deal? Mrazek is under contract until 2016 and then he becomes a restricted free agent. If the Red Wings have to let him marinate in Grand Rapids an extra year, they can, and they still could sign him out of restricted free agency with the promise of a roster spot.