A few weeks ago, I got into a discussion on Twitter about the merits of actual time that an NHL team spends in the offensive zone, compared to more widely-known shot attempt metrics, like Corsi and Fenwick. Anthony Martial Manchester United Jersey . The issue at hand isn’t so much that one particular measure has to be more valuable than another, it’s more about what we know based on available evidence and what to do about that which we don’t know. I spoke with John Chayka, a co-founder of Stathletes, a Canadian company that has been providing comprehensive data analysis to NHL players and teams. While they track zone time, that’s a tiny portion of the data that Stathletes compiles. Now and then, zone time shows up on a TV broadcast. It’s an interesting tidbit, but it comes without context. Outsiders don’t know league-wide numbers (leaders, averages, etc.), so it’s difficult to compare and if you can’t compare, there’s not much statistical value. It’s one thing for me to say that I drove my car 100 km/h. It’s another matter entirely to do it in the left lane on a wide-open highway and another to do it in a school zone in a quiet neighbourhood. Without that context, you don’t know much about how I drove my car, beyond the stated 100 km/h. The puzzle that I was trying to resolve was about whether the Colorado Avalanche could simultaneously be one of the league’s worst by shot metrics last season, yet somehow one of the best when it comes to time spent in the offensive zone. We’ll get to that. STATHLETES One of the things that Chayka said, several times during our interview, was, “We don’t know what we don’t know.” Stathletes takes a Big Data approach, to the point that they have game-trackers analyzing every NHL game in minute detail. It’s not quite fair to say they track everything, because Chayka notes that they keep adding new categories, but they have built a formidable database. Zone time or shot attempts are merely pieces of that puzzle, but once there are years of data in the bank, that’s when statistics start to reveal their significance. It’s one thing to have suspicion, for example, that a team exiting the defensive zone with control of the puck has such-and-such a value; it’s another to have measurable data that can either confirm or refute that idea in tangible terms. Chayka acknowledged that, in compiling their massive repository of information, that Stathletes may very well track some things that don’t appear to have great value or correlation to winning but, since he’s been surprised about the effects of some of the categories they’ve chosen to track, it’s hard to really know what impacts winning until the data is pulled together. As Stathletes started to grow as a company, they talked with a lot of hockey executives, trying to learn what those experienced hockey minds felt was important to being a successful NHL player. Chayka noted that some could be resistant, saying, “None of this data matters. I just want to know if a player competes.” To which a company collecting data responds, “What does it mean to compete?” Answers to that question helped them try to find a way to measure what hockey experts consider being competitive. “Some of the most analytical minds out there are those that think they dont like analytics,” Chayka said, referring to old-school NHL coaches and general managers that have provided the company with valuable insight. “Coaches know a bunch intuitively,” Chayka continued. “How a player handles pressure, whether he takes contact. It can be tough to quantify.” The challenge for a company involved in analytics is to take those insights and intuitions and somehow use them to provide objective measurement. STATISTICAL SECRECY The public is reluctant to blindly accept information that hasnt been evaluated out in the open. On Twitter, it’s regularly referred to as a Black Box, where there is a cloak of secrecy over results and statistical findings, so there is immediately skepticism if those numbers dont fit with what is more widely known. There’s a reason for the secrecy. As this summer made abundantly clear, there is a market for selling data and analysis to NHL teams and if the research and data hasn’t been refuted, then there may be teams that find it valuable. Like it or not, this may be what the future holds when it comes to statistical findings in the NHL. It has already happened in the NBA and MLB, where pro teams have hired a lot of talent that was operating in the public space, and suddenly their research isn’t available for the world to see, it goes only to their employer. That lack of public scrutiny casts doubt on findings that are developed in secret. It doesn’t mean there can’t be value there, only that the public doesn’t know the value one way or the other. However, when there are pieces in the public space which compare zone time to shot attempts and they present a very strong correlation, it becomes more difficult to just accept the differences. TRACKING Those who are critical of using shot attempt metrics, rather than the detail of measured time in the zone, expect that once optical tracking technology is prevalent in the NHL, as it is in the NBA, there won’t be such a need for puck possession proxies. Tracking, it is assumed, will be the holy grail of data collection. Undoubtedly, player tracking will provide more detail and mountains more information than is currently available, but it’s not like player tracking will start one day and untold hockey truths will emerge the next. “Its going to take time for teams to understand what it means and sort it out,” said Chayka. “Were already on the way. It will be a big evolution for everyone, but we feel were at the forefront. Having multiple years of data should be an advantage.” Already knowing how to work with a mountain of data does figure to help when a new mountain shows up. This is a big factor when it comes to the evolving acceptance and use of analytics. When talking to Columbus Blue Jackets Director of Hockey Administration Josh Flynn for Episode 3 of The Hockey Analytics Show on TSN Radio, one of the common threads between Flynn and the guys at Stathletes, is that they recognize the need to get years-worth of data in order to have useful sample sizes. Flynn noted that the Blue Jackets are looking into tracking systems and how that is going to apply in the future, but acknowledged that it requires a cost-benefit analysis. “It’s expensive to gather it, right now,” said Flynn “And you’re going to basically be investing money for a few years to get data that is going to be at least a few years before you can use. It’s a big sunk cost before you actually get value back.” BACK TO THE AVALANCHE Back to the initial question that sent me down this rabbit hole. How can we reconcile that last season’s Colorado Avalanche were one of the top teams in offensive zone time, yet were one of the worst when using shot metrics? Is the answer that the Avalanche hold the puck for an abnormally long time at one end and allow shots at a really quick rate at the other? That kind of stylistic hypothesis doesn’t appear to my naked eye, when watching the Avs play, and it doesn’t appear to have any correlation to ever-elusive shot quality, as the difference between the shots taken by the Avalanche and their opponents didn’t appear to be dramatic. Without definitive answers to how a team could be ranked so highly in offensive zone time and so poorly in shot differential, here are some theories: The Avalanche played a lot of their games at 5-on-5 last season, ranking fifth in the league in total 5-on-5 ice time, so if the context is merely raw zone time during 5-on-5 play, then that would give Colorado a leg up in terms of raw time counted, though it doesn’t explain how they would still manage to get outshot so significantly. As Chayka pointed out, zone time and possession time are not necessarily interchangeable. Stathletes also (separately) tracks stick-on-puck possession time and it’s theoretically possible that the Avalanche spent more time in the offensive zone without actually possessing the puck. Over the course of a full season, though, that’s also a tough logistical sale. Using the data being compiled by Corey Sznajder on zone entries, David Johnson (of Puckalytics fame) found that the Avalanche had a strong net carry-in percentage and it’s not unreasonable to infer that carrying the puck effectively into the offensive zone could lead to more time with possession in the zone, but it still doesn’t account for the shot discrepancy. Ultimately, I don’t have a resolution for how the Avalanche could have top of the league time in the offensive zone and bottom of the league shot metrics, but that stalemate brought a follow-up question: If given the choice between measured zone time and shot attempt metrics, which has more value? To those of us in the public space, we already know that there is a correlation between shot attempts and future goals, so knowing that correlation makes it valuable. Not knowing the correlation between offensive zone time and future goals, or the repeatability of zone time metrics, we’re left to wonder. Maybe zone time is more valuable, but maybe it’s not. Assuming definitively either way would be irresponsible. Baseball statistics pioneer Bill James recently reflected on his career and noted how he was probably too critical of what could not be measured when he started into the field. “I have to take my share of responsibility for promoting skepticism about things that I didn’t understand as well as I might have,” James told NBC Sports’ Joe Posnanski. “What I would say NOW is that skepticism should be directed at things that are actually untrue rather than things that are difficult to measure.” Like Chayka said, “You don’t know what you don’t know.” Some might look at measured zone time skeptically and, citing the lack of context to give it real meaning, there’s reason for doubt. It reminds me of Blackhawks GM Stan Bowman’s quote to Mark Lazerus in this Chicago Sun-Times piece about the Blackhawks’ use of analytics “People can say we’re drawing the wrong conclusions, and that’s fine. But I don’t think we are. I’m not looking to get it adopted league-wide. I like what we have, and I believe in it.” So it goes with the evolution of analytics in NHL circles. So long as the team using them believes in their value, then there is a place for them in the decision-making process. As teams get more and more secretive as the use of data evolves, how would a team know if their belief was misplaced? With more analytics work being done in private, the answer is that teams will have to figure a lot of it out on their own. Scott Cullen can be reached at scott.cullen@bellmedia.ca David de Gea Jersey . Ricciardo made it only halfway around the Jerez track before his RB10 rolled to a stop and began spouting smoke from the back. After team mechanics tended to the car, Ricciardo went back out for a mere two trips around the circuit before calling it quits. Eric Bailly Manchester United Jersey . The CFL unveiled its 2014 schedule Wednesday and the Redblacks will play their first-ever regular-season game in Winnipeg against the Blue Bombers on July 3. http://www.jerseymanchesterunitedsoccer.com/womens-eric-bailly-manchester-united-jersey/ . The third-seeded Bouchard, from Westmount, Que., was coming off a semifinal appearance at the French Open. She took the opening set before the unseeded American came back for the 2-6, 6-3, 6-3 victory.Montreal Canadiens goaltender Carey Price has been placed on Injured Reserve retroactive to February 26, the club announced Monday. Price joins wingers Brandon Prust and Michael Bournival on the injured list, meaning the team has one roster spot open and creates room for a potential addition. "Carey is still day-to-day," head coach Michel Therrien confirmed following Mondays practice from the Staples Center in Los Angeles. "As soon as the medical staff feels hes ready to start practicing, hes going to go on the ice, but obviously he was not today." Price traveled with the team for its four-game Western road trip that begins Monday against the Kings, wherre Peter Budaj will get his fourth-straight start. Luke Shaw Manchester United Jersey. Budaj has already helped the team pick up five points out of a possible six since taking over starting duties with Price. Dustin Tokarski from the Hamilton Bulldogs in the American Hockey League will back up Budaj. Price can be returned to Montreals roster as early as Tuesday and he could play against the Ducks in Anaheim on Wednesday if hes ready to return. Price was hurt last week after facing just one shot at practice. He left the ice in discomfort and appeared to be favouring his leg. Price has a 26-17-5 record this season for the Canadiens, with a 2.33 goals-against average and .925 save percentage. ' ' '