In the previous post, I explained how Taylor Hall’s Corsi% has fallen this year. I defined a sample in the 2011-12 and 2012-13 seasons and examined the zone entries and shots. I discussed what’s gone on in the defensive zone and concluded that it was hard to find all that much in terms of results that was different from last year.
The offensive zone is a different story.
As I mentioned in that piece, there’s been about an 18% decrease in the shot attempts for with Hall on the ice at 5v5 this year. That’s a pretty amazing collapse. The point of digging into something like this is to try and identify, with greater specificity, the things that have changed with Hall on the ice that have led to this decrease in shot attempts.
We’ll start at looking at the zone entries. As I explained in the previous post, I classify zone entries into ten different types.
CS – Successful carry
CU – Unsuccessful carry attempt
DC – Dump and change, used when there was no or negligible pressure following the puck being dumped in
DS – Dump-in where the attacking team achieved control of the puck before it was cleared from the offensive zone
DU – Dump-in where the attacking team did not achieve control of the puck before it was cleared from the offensive zone
SS – Used when Hall came onto the ice with the puck already in the defensive or offensive zone
PE – Used when a penalty expired with Hall on the ice and the puck in the defensive or offensive zone
OZW – Offensive zone faceoff win
OZL – Offensive zone faceoff loss
X – Other. This largely consists of two types of play. First, when the defending team brings the puck into the defensive zone, frequently after a faceoff win in the neutral zone or while regrouping. Second, when the attacking team sends the puck into the defensive zone without any intent of making an entry. This might included failed passes in the neutral zone or pucks being fired out of the defensive zone or before the red line to relieve pressure.
One of the things that analytic types have found is that successfully carrying the puck in to the offensive zone tends to lead to better outcomes than dumping the puck in. What’s more, early research seems to indicate that a lot of the difference between players in terms of their team’s shot attempt rates when they’re on the ice is tied in to their ability to gain the zone with possession.
This is an important idea and I want to put it into context a little bit. If you’ve read Moneyball, as I assume most of you have, there’s a passage in which Michael Lewis discusses a new approach to valuing baseball players, which involved assigning run values to minute things on a baseball field:
Ken Mauriello had seen a connection between the new complex financial markets and baseball: “the inefficiency caused by sloppy data.” As Bill James had shown, baseball data conflated luck and skill, and simply ignored a lot of what happened during a baseball game. With two outs and a runner on second base a pitcher makes a great pitch: the batter hits a bloop into left field that would have been caught had the left fielder not been Albert Belle. The shrewd runner at second base, knowing that Albert Belle is slow not just to the ball but also to the plate, beats the throw home. In the record books the batter was credited with having succeeded, the pitcher with having failed, and the left fielder and the runner with having been present on the scene. This was a grotesque failure of justice. The pitcher and runner deserved to have their accounts credited, the batter and left field to have theirs debited (the former should have popped out; the latter somehow had avoided committing an “error” and at the same time put runs on the board for the other team).
There was hardly a play in baseball that, to be precisely valued, didn’t need to be adjusted for the players involved, or the ballpark in which it occurred. What AVM’s system really wanted to know was: in every event that occurs on a baseball field, how – and how much – should the players involved be held responsible, and therefore debited and credited? Answer the question and you could answer many others. For example: How many doubles does Albert Belle need to hit to make up for the fly balls he doesn’t catch?
How to account for a player’s performance was obvious: runs. Runs were the money of baseball, the common denominator of everything that occurred on a field. How much each tiny event on a baseball field was worth was a more complicated issue. AVM dealt with it by collecting ten years of data from major league baseball games, of every ball that was put into play. Every event that followed a ball being put into play was compared by the system to what had typically happened during the previous ten years. “No matter what happens in a baseball game,” said Armbruster, “, it has happened thousands of times before.” The performance of the players involved was always judged against the average.
The system then carved up what happened in every baseball play into countless tiny, meaningful fragments. Derivatives. “There are all sorts of things that happen in the context of a baseball play,” said Armbruster, “that just never got recorded.” A tiny example: after a single to right field a runner who had been on first place, seeing that Raul Mondesi is the right fielder, stops at second base instead of dashing to third. Runners seldom tried to go from first to third on balls hit to right field when Raul Mondesi was playing there. That was obviously worth something: what? Just as it never occurred to anyone on Wall Street to think about the value of pieces of a stock or a bond until there was a pile of money to be made from the exercise, it never occurred to anyone in the market for baseball players to assign values to the minute components of a baseball player’s performance – until baseball players became breathtakingly expensive.
Bill James’s work had been all about challenging the traditional understanding of the game, by questioning the meaning of its statistics. The financial experts at AVM took this idea even further, by recording the events that occurred on a baseball field without any reference whatsoever to the traditional statistics. It wasn’t just circumstantial statistics such as “RBIs” and “saves” that the AVM model ignored. It ignored all conventional baseball statistics. The system replaced the game seen by the ordinary fan with an abstraction. In AVM’s computers the game became a collection of derivatives, a parallel world in which baseball players could be evaluated more accurately than they were evaluated in the real world.
This idea of the game seen by ordinary fans being replaced by an abstraction is an interesting one. I think you can make an argument that hockey coaches are, in a way, dealing with a massive abstraction as opposed to the game that regular fans like you and I see. John Tortorella touched on this recently in his comments following yet another Canuck loss:
“We can’t let the momentum change in that third period the way it did. And that’s details. I need to take full responsibility for that. Obviously I have not taught that well enough because we continue to make the same mistakes in crucial times of the game. You guys may not see the little things that happen before goals, we do. That’s situational play and we have not been consistent enough with it.”
The stuff that I’m talking about here is the little things that happens before goals. I have some doubt that coaches are perfect in terms of their own understanding of the little stuff before goals. I’ve listened to coaches talk about minute points before and it strikes me that the good ones have thought through every angle on tiny things. I don’t think that they always come to the right decision though because they’re making that decision based on their own experiences and we’re all terrible at coalescing individual experiences into a coherent belief structure, particularly when the connection between an action and the ultimate reward (a goal) occurs as infrequently as it does in hockey. About the best you can hope for is a coach with the right philosophy – I think Edmonton has one – and that, through trial and error, he gets it right.
Let’s talk about applying this concept of valuing minute plays on a baseball field in terms of runs in hockey. Eric Tulsky has done a lot of work on zone entries. In a piece that he published at NHL Numbers, Eric found that carrying or passing the puck into the offensive zone resulted in about 0.56 Fenwicks (shots on goal or shots that miss the net) and dumping the puck in resulted in about 0.26 Fenwicks. We know that Fenwicks lead to goals, so carrying the puck in has a higher expected goal value.
This shouldn’t be taken as a blanket endorsement of always trying to gain the zone with possession. There are times when it’s simply not there and you have to take what the defence is giving you. That being said, on average, when the puck is dumped in, the outcomes aren’t as good as when it’s carried in.
Let’s look at an example from an Edmonton game this year.
Here’s an Oiler breakout late in the second game of the year. Hall, Hemsky and Yakupov are on the ice as forwards.
The puck is headmanned to Yakupov to knocks it back to Hall, who’s approaching with speed. The Canucks have all five skaters behind the puck.
Hall cuts towards the left boards. He’s got four Canucks in front of him. Hemsky is slowing down in case Hall wants to skate it in. The only realistic option he has to preserve the possibility of a carry into Vancouver’s zone is passing it back across to Yakupov.
Instead, Hall dumped the puck in. The Canucks cleared it back out. On average, the cost of that decision would be about 0.34 Fenwicks. In this case, with four Canucks on one side of the ice and Yakupov on the other, I’d guess that the cost was actually higher. If it’s one decision that was made incorrectly, who cares. About 5.53% of Fenwicks become goals, so if Hall made a decision that cost the Oilers 0.34 Fenwicks, we can convert it to 0.019 goals.
The thing is though – there are a lot of moments like this in a hockey game. And things add up. A lot of the really interesting hockey analytics right now is focused on questions like this – trying to convert things that players do (whether through tactical instruction or as result of their own skill/lack of skill) into goals. This decision from Hall that I’ve highlighted and the analysis of it is a kind of coarse derivative of the type discussed by Lewis in Moneyball.
When SportsVU comes to hockey, it will get a lot more sophisticated – there was a fantastic article on Grantland the other day talking about a new approach that’s been developed with this data. That being said, I’m pretty much convinced that there are NHL teams that are intelligently mining data from games and using it to generate an edge for themselves.
With that context for why this type of analysis is important and why zone entries matter, let’s dip into the data. Here are the Oilers’ zone entries with Hall on the ice in 2012-13 and 2013-14:
Well. Isn’t that interesting. If you read the first post in this series, you’ll know that things looked virtually identical on the defensive end, in terms of how the opposition gained the Oilers zone. That’s not at all the case in terms of how the Oilers gain the offensive zone. There’s an 18% decline in terms of the number of possessions that were started by carrying the puck into the offensive zone.
Unsurprisingly, the decrease in pucks carried in to the offensive zone means an increase in pucks dumped in to the offensive zone. Leaving aside dump-ins that were cover for a change, the Oilers have gone from dumping the puck in on 9.4% of their zone entries with Hall on the ice to doing it on 17% of their zone entries. Worse, they’ve become terrible at recovering pucks that are dumped in – in 2012-13, they were successful, in terms of achieving possession of the puck before the opposition cleared it, on 62.5% of their dump-ins. In the sample I’m looking at, that number’s a smooth 23.9%.
Let’s switch gears a little bit and look at how many shot attempts they’re generating on each type of entry.
You will recall, if you read the previous piece, that things were pretty similar year over year in terms of how many shot attempts the opposition was generating off the various types of zone entry. It’s not really the case here but the samples are so small for everything except carries that I wouldn’t read too much into the results.
The results of carries are sort of troubling though. In 2012-13, the Oilers generated 0.66 shot attempts per carry with Hall on the ice to 0.62 for the opposition. This year, the opposition is generating 0.72 shot attempts per carry with Hall on the ice to 0.56 for the Oilers. When I looked at how opponents were doing against Hall, I found that basically, my sample includes a couple of carries where the opposition had multiple shot attempts off a single entry and that this accounts for the increase in shot attempts allowed by the Oilers. It’s possible that there’s a problem there but it could well just be some bounces running against the Oilers.
Let’s look at how many shot attempts the Oilers generated off each carry into the offensive zone with Hall on the ice in our sample in 2012-13 and 2013-14.
The yellow highlight tells you something I think. That’s a 15% increase in carries into the offensive zone in which the Oilers generated no shot attempts over 2012-13. I note as well that it’s about 15% higher than what the Oilers opposition did when they carried the puck into the Oilers’ zone with Taylor Hall on the ice in in 2012-13 and 2013-14. In both years, the opposition was at about 50% in terms of getting zeroed out. This strikes me as being a pretty large and disturbing increase.
I’ve got a theory as to why this might be. I score a play as a “carry” once the blue line is gained with possession. If a player gains the blue line and then chips a puck deep without any intention to play it to a teammate in a better position, it’s effectively the same thing as dumping the puck in. Possession has been abandoned. I’ve got a theory that they’re doing more of it this year, along with something else that I want to look at. It’s an easy enough thing to look into now that I’ve got a time stamped list of “carries” that resulted in no shot attempts – you shouldn’t just take my word for it – and I’ll get into it before the NHL resumes.
Let’s look at the breakdown in terms of getting multiple shot attempts.
You can see that, despite the overall decline in shot attempts with Hall on the ice, the Oilers have actually been marginally more effective at generating multiple shot attempts once they get the first one when they carry the puck into the offensive zone with Hall on the ice. It’s not a problem with generating multiple shot attempts. They’re just fine…provided they generate that first shot attempt.
What does this all tell us? I see three issues. First, the Oilers have become much more likely to dump pucks into the offensive zone with Hall on the ice at 5v5. Second, (in this sample at least but keep in mind that it was picked because it was pretty representative of the whole), they suck at retrieving those pucks. Third, they’ve gotten worse at generating shot attempts when they do carry the puck into the offensive zone.
As far as the dump-ins go, I think it’s pretty safe to suspect that the coaching staff has been preaching that the puck should be dumped in more. There was a quote from Dallas Eakins on January 27 in which he talked about Hall’s struggles and said:
“I think with Taylor, he’s taken on some of our teachings almost a little bit too far, become maybe too safe. We’re trying to give him a little bit more rope so he can get back to even a higher octane game. That says a load to me about the player, when he’s taken the coaching a little bit too far, trying to listen, trying to buy in and those are great, great signs for me.”
What this leads me to wonder is whether the tactics that they’ve been pursuing are necessarily the best use of Hall. It’s undeniable that the GF% has gotten worse with Hall on the ice this year. This is despite his on-ice shooting percentage and on-ice save percentage being above his career norms. Even that can’t overcome the deluge of shot attempts that’s buried him.
I mentioned above that I think the issue about carries with zero shot attempts being generated can be narrowed further with some video work. I’m inclined to think that I can come up with some stronger views about the dump-ins too. There are a couple of plays that appear to be set type plays that have stuck out to me. It’s worth a longer look, I think.
Assuming that all of this analysis is accurate – I’m fairly confident in it – I think it’s fair to ask why this happened. It’s hard for me to link it to anything other than a tactical emphasis from the coaching staff that has had undesired consequences. Is it reasonable that this has kind of festered all year? I’m not sure. As I said above, I like what I’ve heard from Dallas Eakins.
I’ve considered the possibility that he’s just blowing smoke about what he thinks as far as how the game should be played but it seems unlikely to me. Why get into contentious exchanges with reporters, who are going to take their shots at you afterwards, for something you don’t believe in? There are coaches out there – Michel Therrien, Randy Carlyle, I’m looking at you – who don’t seem to care about being a strong possession team. To me, it seems more likely than not that the Oilers have a guy who has the right view of the game.
That’s only half the battle though. The other half is making tactical decisions that serve that end or, if they aren’t serving that end, identifying them and either abandoning the tactic or correcting the flawed implementation that is leading to the tactic not serving the desired end. I’m inclined to think, on the basis of this data, that the Oilers coaching staff isn’t really accomplishing this.
One final caveat: it may well be that the Oilers need to become more of a dump and chase team to have success and that the problem this year is more tied to a lack of appropriate personnel for that. I don’t believe that that’s true but I don’t really have evidence in support of my position. The Oilers have talked about wanting to be a team like the Blackhawks or the Red Wings. I’d be interested to know if Toews/Hossa/Sharp are dumping the puck in as frequently or as ineffectually as the Oilers are when Hall’s on the ice. I’m pretty doubtful that the answer is yes.Email Tyler Dellow at firstname.lastname@example.org