Thoughts on the Oilers: Chasing Optimism

Dead last in the conference. Sitting at 6-10-2. Brutal.

img003Quick rundown of the season:

  • At home: 1L, 4W, 4L
  • On the road: 5L, 2W, 1L
  • Overall: 5L, 4W, 4L, 2W, 3L.

It’s easy to pick this team apart. For one, they’ve struggled at different areas like goaltending, defence and centre. And second, we have experience dissecting a bad team as it’s been the same story since 2006. We know what a bad team looks like, right?

But unlike the past few years, there has been progress. The 12 (!!) forwards look like a legit NHL group. RNH is growing into a star. And the underlying stats are pointing to an improved puck possession team. Unfortunately, the wins aren’t coming, so the frustration is starting to build among the masses. And preaching patience isn’t going to work on a fanbase that has been waiting for a competitive NHL team since 2006.

Sapp made a fantastic point that pretty much summed up my general outlook on the Oilers:

The last three losses have been extremely frustrating. Each game featured stretches of terrible play by the Oilers. The first period in Nashville where they got outplayed and outshot, resulting in a 3-0 defecit was a perfect example. The Oilers did manage to score a couple to make it close, but they couldn’t recover from the hole they dug themselves into. When a team comes out flat like that, I tend to fault the coaches whose sole purpose is to implement a strategy and get the right guys deployed. My concern this past week has been that adjustments appear to be made at the intermission rather than during the period. Regardless of the reason, the club is losing a lot more 1-goal games and just aren’t finishing their chances. You’d think this will turn around soon, but then the club keeps giving the cherry minutes to Shultz, scratching Marincin from the lineup, and musing about adding Gazdic to the roster.

Tough stretch of games coming up against western conference teams who the Oilers, for whatever reason, can’t beat. Absolutely bizarre.

Quick point on analytics: It’s a highly personal thing. There are common methods to uncover patterns in data, but the value we put into the data depends on our own biases and interpretation of the game. For instance, Corsi/Fenwick are great. But I tend to value shots on goal and shot differential higher. Just do.  And also keep in mind that analytics has, in my opinon, enhanced the overall discussion of the game. It won’t explain everything, and it might not always make sense, but it’s a much more engaging starting point than “Hemsky is the first guy off the ice at practice…”.

Recommended Links

Oilers are taking positive steps – TSN

Season fancystats updateafter 15 games – Boys on the Bus

If the Oilers’ advanced stats are so good, why don’t they win hockey games? – Oilersnation

One of these nights – Lowetide

The Yandle Treatment – Woodblog

Expanding the Scope: Insight from Gabriel Desjardins of Behind the Net

fehr-behind-netIt’s been great to see hockey analytics grow this year, but it’s still perplexing that more people aren’t looking past it and asking tougher questions about it’s relationship with other concepts and fields of research. The numbers and metrics are a part of a continuous discussion, which has intensified with more and more people joining in the discourse. For instance, possession stats have gained prominence, and that’s lead to more questions about the game. Projects tracking zone entry stats and puck retrieval stats will uncover new information, which will likely spawn off more questions. That’s the thing about analytics in any field…there is no finished product.

So if we look past the actual stats and the ensuing discussion, there’s still a lot to be understood about how exactly hockey analytics has impacted the different facets of the game (management, scouting, players, etc) as well as it’s relationship other fields such as information technology, business and society in general.

Behind the Net was one of the first websites that collected advanced stats, with its owner Gabriel Desjardins leading a lot of the online discussion in the early days. He sent out a series of tweets last week that gave some insight into hockey analytics, but also touched on some topics that have yet to be fully explored.

As we approach the trough of disillusionment for hockey analytics, here are a few helpful thoughts…

#1) Hockey insiders have been using “analytics” for decades. +/-, Sinden/Corsi shot/pass/touch counting, video aggregation. These stats had the imprimatur of cigar-chomping insiders, so nobody dug too deeply or cared too much

#2) One day, members of the general public found out what insiders had been doing and slowly worked through the value of this data. Somehow people popularizing the league’s internal metrics became outsiders as far as fans/media were concerned. It’s a classic obtuse battle of ideas. e.g. Obamacare was conservative for Romney but socialist for Obama. It took “analytics” predicting an unavoidable Leafs collapse to push people to the “Peak of Inflated Expectations”. It’s amazing – people promoted ideas the NHL used for decades w/o press caring but these ideas then needed to be proven publicly

#3) Now that analytics have been re-proven externally, teams have been getting PR boosts by announcing various hires but teams were already using analytics. So there’s no new benefit. Except I suppose people will expect the Leafs to benefit, hence the trough of disillusionment.

#4) Now here’s the missing piece: teams need to know how to interpret these stats and correctly use them to drive decisions. It’s statistical parallel to @Lowetide_‘s “saw him good” principle. Teams don’t understand regression will cut guy based on 3 games. Teams need to take long view to get analytics benefit and incentives don’t align. Today’s best GMs still only have 14-day outlook

#6) May see benefit for poorly-run teams [TOR, EDM] but the inflated expectations are that “stats guys” will take them on a 2015 playoff run. Unless @mc79hockey is making $1M a year, we need to seriously temper expectations

#7) There’s very little low-hanging fruit in analytics and most of it has been harvested in hockey. There’s no Matt Stairs or Roberto Petagine waiting to be freed. Helps that the KHL will put a pile of cash in your suitcase at the end of every game. Nobody needs to toil in the A for $75k/year

#8) The thing that initially annoyed me was Pierre Mcguire’s comments to @wyshynski about firing coaches for using analytics. But teams already sign and play guys because “Coach knew him in junior” or much worse.

Potential Jobs in the field of Hockey Analytics

The recent NHL hirings are a good indication that there will be more jobs out there for hockey bloggers, especially those who have a good understanding of analytics. Analytics really is a continuous discussion, so following the flow of information that comes from the process can shed some light on what areas of the game will likely hire more individuals.

So far we’ve seen a few bloggers get hired to work on the analytic departments of NHL clubs. My assumption here is that they’d sift through all the of the available data and provide recommendations on things like player personnel and possibly game tactics.

But as we’ve seen the Leafs do, recently hiring Darryl Metcalf, there’s also a need to collect data and then publish it in an easy-to-use, dashboard format for management and coaches to access. As we’ve seen online, there are a lot of fans who have developed tracking software and reporting tools, like Super Shot Search, Player Usage Charts and Shift Chart, that might be of interest to NHL clubs

Taking it one step further, teams will likely need some in-game analysts who can gather and share information to fans and the media. Teams typically provide some sort of game file with an assortment of stats and other information. These types of communication pieces will now likely include some high level stats that could be of interest to the public.

Teams may also look for people with coaching experience, who might have experience applying the analytics to coaching tactics. For instance if a club is finding that their defenceman is struggling and the opposition has been found to be strategically attacking his side of the ice, how does a team respond?

Teams may also want simple data trackers, and could rely on external companies, such as the one in Europe that specializes tracking soccer games

And of course, like in any professional sport, there are many, many people outside of the league who want to dissect the play of a team and players, including journalists, broadcasters and of course fans.

There will definitely be a huge demand for writers, specializing in data journalism, as major sports websites will want people who understand the full spectrum of analytics and its application to the game. But tv and radio broadcasters (in-game color commentators, talking heads, etc) will also have job openings, as there will likely be more discussion pertaining to the new questions from viewers about what impact the analytics will have on team success.

Following the flow of information, there will be a very high demand for hockey analytic expertise outside of the game. This would include groups involved in fan engagement such as video game developers, fantasy league service providers and even gambling sites. These external groups have always relied on hockey information to supplement the fan experience, and will likely look to expand their operations to include additional advanced stats (if they haven’t already).

And keep in mind, others that you may not expect to care about analytics just might be looking for some help to understand teams and players. Maybe there’s an advertising company out there that wants to align itself with a team that will  have a talented young core (that gets lots of ice time), for branding purposes. It’s a stretch, I know, but analytics really is a continuous discussion that can lead down many different paths.

Any job that is related to hockey analytics will require an individual to possess more than an understanding of analytics. Taking a step back and following the flow of information that can come from analytics, there will be a growing need for individuals with various backgrounds and experience. Organizations will likely search for people with information technology, broadcasting, or communication skills, among others, to really solidify their approach to analytics. The good news is, anyone who is interested in working in this field can begin honing their skills online using blogs and other publicly available applications.

Keeping the NHL Data Open

legosIt would make absolutely no sense for the NHL to restrict fans access to any level of stats, whether it be the traditional set or the advanced possession ones (i.e., Corsi, Fenwick, etc).

The NHL has changed it’s Terms of Service to reflect that. On Thursday, Zsolt Munoz of The Copper and Blue provided an excellent summary of the changes and raised concern that this might impact the go-to websites like Behind the Net and Extra Skater.

As of Friday morning, Extra Skater has become inactive, much to the dismay of its many users. The website does scrape data from the NHL website but extends it to calculate new information and “advanced” stats. Darryl Metcalf, the websites administrator, puts it all in a fantastic dashboard format making it easy to navigate and use. I’m really not sure what’s happening with ES, but I’m pretty confident that the NHL will not be preventing anyone from using and extending its game data. Couple points:

  • Any sort of data analysis, or finding patterns, whether it be simple counting of goals or calculating Corsi%, requires an individual’s time and effort. The more time and effort a person puts in, the more engaged they become with the game. It requires critical thinking and gets people into the game a lot more than if they just read static stats from a game report. Fantasy league is a prime example of something that gets fans into the game and is linked to data analysis. Fans are growing an attachment to teams and players they typically have no interest in because of fantasy leagues. The NHL has got to love that.
  • There’s a been a lot of weight put into the concept of “open data“. The idea is if you provide anyone and everyone with easy to use data, individuals may be able to come up with new information that can be shared with the community at large. For example, the City of Edmonton publishes data sets, which have been used by individuals to develop new apps available for public use.
  • And, as mentioned by Tango Tiger, the online ecosystem today, supported by stats from the NHL or the data scraped by Extra Skater, have served as a fantastic training ground for analytic folks. Teams like the Edmonton Oilers and New Jersey Devils have hired individuals who have had access to this data and demonstrated its value.

I really see the NHL really opening up the flood gates when it comes to data. They may push to be the gatekeeper of data, which may be a fair trade-off if fans can extend it into accurate information. Past attempts by Major League Baseball to stifle the analytics growth have either failed or been held up in court, which I think gives hockey fans some hope that whatever data they want will be available to them free of charge. I also think the NHL will develop a process to support the development of new applications that use the data in some creative way. The more people toiling away with hockey data, the more time they’re spending on NHL related stuff…I assume that’s what the NHL would want,

And hey, if for whatever reason the league restricts data or even tries to charge for it, you will definitely see more tracking projects pop-up, similar to Corey Sznajder’s All Three Zones Tracking Project, who has found an excellent way to track his own stats and receive compensation from the larger fan community.

Information and knowledge can’t be restricted in the modern society as far too many tools and the cooperative nature of individuals will support its growth. The momentum of information growth is giving fans a new role when it comes to consumption as a shift from passive consumption has been replaced by active, collaborative, engagement.

Data Journalism in Hockey

old-time-reporterHockey analytics is doing some pretty remarkable stuff for the game. Not only are there new methods of understanding and following the game, but we’re beginning to see some of the more unintended consequences of its growth.

One thing to keep an eye on will be the evolution of hockey reporting and journalism. Analytics has received widespread attention this summer with NHL teams hiring experienced professionals and discussing the new statistics quite publicly. All of this is great, but now the people following the game, especially those that are relatively new to hockey analytics, will demand supporting evidence, in an easy-to-use, storytelling format.

Travis Yost brought up an excellent point, citing the potential for new data tracking technology as a reason why the demand for better sports writing will likely grow. There will definitely be teams looking to hire individuals with experience in analytics to keep up, but there will also be media outlets looking to enhance their coverage of the game and these advanced stats.

This will put the onus on anyone developing information about the game, whether it be newspaper writers, hockey analysts on television and fans online. The game has changed because of analytics, and now the information surrounding the game has to evolve.

What this means is that the distribution of  information has to shift from simple reporting to more data journalism. It won’t just be about grabbing quotes from coaches and players and then referencing some of the new stats like Corsi and Fenwick etc. But it’ll also mean having to embed graphics and information in an interactive and engaging way to tell a unique story.

The good news is, there are a handful of newspaper writers and bloggers who have been doing this very well for a few years now. James Mirtle of the Globe and Mail, as well as Japers’ Rink and SB Nation’s Outnumbered come to mind. But now, there will be an increasing demand for it. And I can definitely see some slick new platforms or reporting dashboards that can take all of the advanced stats and put together a strong game story.

Blogging platforms have served as a solid foundation and medium to develop hockey analytics. And now, they can be a real training ground for those looking to advance their careers in writing about the growing field.

 

Oilers hire Tyler Dellow of mc79hockey.com

620-eakins-dallas2-thumb-620xauto-329706Happy to see the Oilers hire Tyler Dellow of, the now defunct, mc79hockey blog. Tyler’s been blogging for years about the Oilers, focusing on hockey analytics and recently applying his findings to game footage. The blog was a fantastic source for analysis, really pushing the discussion on hockey analytics, and will definitely be missed. Sounds like he’ll be joining the Oilers hockey operations department and reporting to Coach Eakins.

Related: On Tyler Dellow, his role, and what this can tell us about Eakins – Copper & Blue

I’d love to point out some of the work he’s done including his dissection of Taylor Hall’s Corsi stats and the Oilers performance following defensive zone faceoff losses. But I guess it’ll have to wait until someone can uncover that hidden treasure. A lot of his work picked apart the Oilers and did show some of the weaknesses in their tactics, so I guess it makes sense for the Oilers to ask him to remove the content. His blog was also one of the few sites where the comments section was just as  valuable as the actual post. I found it a little troubling that Tyler was able to wipe out the work of others who trusted that their contributions would be always be accessible. I’m sure most of the commenters are supportive of Tyler’s move, so maybe losing their work was just a necessary trade-off.

Those who have read my work know that more than hockey analytics itself, I’m interested in the social, online environment that fosters the development of hockey information and knowledge. So as much as the growth of analytics interests me, so too does the network and the collaboration of fans.

I think Tyler would admit that a big reason for his success in analytics is the open environment that he was able to work in. He received regular feedback both on his blog and on Twitter, which played a big role in the direction he took with hockey analytics. He’s obviously a bright guy and will more than likely carry on with his ideas and insight as a member of the Oilers. However, the corporate culture that the Oilers have is far different than the interactive environment Tyler has grown accustomed to. Online, Tyler has been extremely open with his data, which makes him open for feedback and criticism. Now that his work will only be reviewed by a select few, I’m curious to see what impact it might have on his own critical thinking. There’s a much different working-dynamic when you’re sealed off with a select few. Biases, groupthink, politics, etc., are all the things that tend to creep in to any bureaucratic system.

It’s almost as if the Oilers should actually hire a few more bloggers that Tyler’s work can be tested by. The good news is, the Oilers do have an analytics working group that can really challenge his work. But it might be in the Oilers best interest to add additional bloggers, or some sort of external advisory group, with little social or financial ties to the team, who can provide Tyler with feedback.

Ice Code

thinkerGreat explanation as to why hockey analytics is important for fans and the game as a whole. Unfortunately I can’t find the full article online to share, but the original source is below.

 

Accepting hockey analytics, then, requires more than brushing off your arithmetic skills and reading a spreadsheet. It requires a shift in thinking, embracing an uncomfortable amount of arbitrariness, accepting that a bounce is just a bounce and dumb luck can have more to do with a win  than grit or heart. Not to make too lofty a comparison, but it can sometimes feel like the difference between believing in a world of a higher order, of divine narratives and preordained plans, and seeing the word as a messy, arbitrary place. The numbers challenge some of the stories we used to explain the world, but they also introduce their own kind of complex beauty.

Fervent irrationality, after all, is what makes a fan a fan. But thinking about the percentages and probabilities doesn’t suddenly drain the magic from the sport – it just means watching the game more thoughtfully.

Hune-Brown, N. “Ice Code”. Sharp. April 2014: 94-95. Print.