View from the Press Box – an experiment in analytic game stories

Having spent countless hours as a visiting media member at AHL and OHL rinks, on Saturday, October 25, 2014, I had my first opportunity to take on the same role at an NHL sized version. The game in question featured a suddenly Zdeno Chara-less Boston Bruins travelling to Toronto to take on the Maple Leafs. For the hometown squad, while they were not missing any roster regulars – none that they haven’t been missing since preseason, anyway – this is their first game since their mid-week matchup in Ottawa against the Senators was postponed due to an act of terror, as a lone gunman murdered a Canadian reservist soldier serving as an honor guard at a war memorial.

The first question to be answered in the game would be how Bruins coach Claude Julien adjusted his lineup in the absence of the towering minute-eater Chara. Through their first nine games, Chara was half of the Bruins’ shut-down pair, alongside the still-young Dougie Hamilton (he will continue to be young as long as he goes by “Dougie”, even though that is the name on his birth certificate.) Dennis Seidenberg has also played a very defensive role in the early goings, with a team low offensive zone start ratio of 35.1% entering play, although without having played against the level of competition that was Chara’s standard. Seidenberg had been typically paired up with Adam McQuaid, while the third pairing had featured Torey Krug and Kevan Miller when healthy, or since Miller has also been injured, Krug would play with Seidenberg and McQuaid would be paired with Matthew Bartkowski.

Analytics would suggest that Seidenberg play with McQuaid against the Leafs top line of Kessel, James van Riemsdyk and Tyler Bozak, while they reserve Krug and Hamilton for offensive roles more suitable to their talents. 24 year old Zach Trotman, a former 7th round pick with two games of NHL experience should then join the third pairing alongside Bartkowski. Does Julien agree? Not at game-time, anyway, as the opening faceoff saw McQuaid and Hamilton facing the Leafs’ third line of Mike Santorelli between Leo Komarov and David Clarkson. Trotman did see his first action skating with Bartkowski, though.

The Bruins drew first blood, a power play marker from Carl Soderberg, who gathered a loose puck hard by the Toronto crease after Bernier stopped, but failed to control, a previous shot, and tapped it in for an early 1-0 edge. By the mid-point of the first period, the defensive pairings with Boston began to resemble my pre-game ideas, as Krug paired with McQuaid, while Hamilton played with Seidenberg. Not my exact idea, but close enough.

The first period was relatively uneventful outside of the Soderberg powerplay marker. While the shot clock showed a slight edge to Boston of 11-9, the official numbers were unduly kind to Toronto, as at least two of their “shots” were merely dump-ins which hit Tuukka Rask in the Boston crease. Per, the Corsi events favored Boston by a count of 24-15, a far truer representation of the run of play through 20 minutes. Even without Chara, Boston played a style more conducive to sustained pressure, as its players actually take the time to think before moving the puck onwards. The Maple Leafs have tended to shoot first, ask questions later, generally resulting in the puck quickly ending up back on the stick of a Boston player and very quickly back in the Toronto zone. Also of note, the Maple Leafs very seldom used the five forwards not in the top six, placing the bulk of the burden on the shoulders of the above-mentioned first line, as well as Nazem Kadri, Joffrey Lupul and Daniel Winnik.

To get a better grasp of how both teams play, for the second period, I tracked all zone exits, tabulating carries, long passes, short passes (passes past the center ice line deemed long) and non-pass clears (both open-ice and icing). I also tracked zone entries, splitting those between dumps, passes over the blueline and carries/carry attempts. For each of the latter group, I also noted how many resulted in sustained possession in the offensive zone. By the way, zone exit/entry tracking in real-time is shockingly difficult to do, and I must have missed a handful of either simply while marking a previous one. If you hadn’t noticed, hockey is a very fast-paced game. The difficulty would be compounded ten-fold if I was also trying to track which individual players did each.

For the second period, each team accrued 19 Corsi events. The Leafs dumped the puck in 13 times, and only two of those led to some measured of sustained offensive zone time. Out of 15 tracked carry-ins, 10 led to sustained possession in the Boston end. One of three pass-ins were successful, with the other two pass attempts leading to offside calls. On the other side the Bruins dumped the puck in the Toronto end nine times, none of which led to sustained possession. They had one pass-in which went nowhere. Of 11 carry-ins, seven led to drawn out zone time, including one beauty of a play by David Krejci, as he crossed the blueline along the boards, curled towards the net as he out-raced Phil Kessel, who was chasing him – note that he had no offensive support during this rush – and put a puck on Bernier in the Toronto crease. Bernier stopped the initial shot but, much like with the Bruins’ first period goal, the rebound was nice and juicy, and Krejci had no problem picking it up in the course of his forward momentum through the crease and was able to slide the puck past the red line for a 2-0 cushion for the Bruins.

Even though the actual numbers should be taken with some grains of salt as I could not double check, the severe trend that saw carry-ins lead to sustained offensive zone possession, while dump-ins more often than not ended up returning from whence they come in short order, was unmistakable. This is not news to the analytics community, but it cannot be reminded often enough. The best way to establish possession is to not give it away. When the puck is held, the opposition needs to actively get it from the carrier. When the puck is put to a race, it becomes a 50-50 battle with an unreliable outcome – even more so when the opposition has two defenders back and both cognizant of the actions of the initial puck carrier.

The third period saw the pace of play pick up a notch or two and the Bruins padded their lead by two in a little over four minutes of play. The first addition came during a penalty kill, as a forechecking Daniel Paille played keepaway with Jake Gardiner by the half-wall before pivoting and sending a pass to the slot where Gregory Campbell had gained some space from Dion Phaneuf. Campbell got his stick on the puck and redirected it behind Bernier in the Toronto crease. A few minutes later, a puck battle in the Boston end was won the by a backchecking Patrice Bergeron, who pushed the disc towards the center of his own blueline, where Dougie Hamilton picked it up with pace. He burst up the ice, showing high end speed, beating Jake Gardiner (a common refrain on the evening) and avoiding an angling Stephane Robidas before firing a hard wrist shot from just inside the “Homeplate” area in front of the Toronto net. That goal spelled the end of the line for Bernier, as Toronto made the switch to James Reimer to see the game through.

Toronto finally got on board with under six minutes to go, as they managed to keep possession on a delayed penalty call in the offensive end. With six attackers on the ice, a ferocious goalmouth scramble saw waiver pickup Richard Panik score his first for Toronto, meagre consolation for a crowd that had been either silent or cantankerous ever since the first puck dropped.


The Bruins cycling with the puck in the Maple Leafs zone, a common sight (Photo courtesy of the author)


–          After the final whistle, as his charges prepared to return home for a date with the Minnesota Wild, Bruins head coach Claude Julien, was asked how he planned on dealing with the absence of Zdeno Chara. After reminding the questioner of the long-term absence of do-it-all center Bergeron due to a concussion shortly after he was hired by the Bruins, he mentioned that “instead of leaning on one guy, we’ll lean on each other”, suggesting that his defense corps would not simply be shifted up a man, but that the burden taken by Chara, who averaged 24:39 TOI in 2013-14 and nearly 24 minutes per game this year, would be shared by the entire lineup. On this night, he was partially accurate, in that his top four of Hamilton, Seidenberg, McQuaid and Krug carried the bulk of the load, each playing over 20 minutes, although he still sheltered his bottom pairing of Bartkowski and Trotman. The latter two mostly started shifts in the offensive zone and generally after the Leafs top line of Kessel, van Riemsdyk and Bozak had recently left the ice, so as to minimize the risk that Carlyle would create a mismatch for his offensive talents. Hamilton and Seidenberg received most of the own zone starts against the Leafs’ top six, while McQuaid and Krug were given more offensive zone time. While many lamented the trade of Johnny Boychuk after Chara was hurt, based on this small sample, it looks as if the Bruins can withstand his absence with intelligent playing time distribution of the top four blueliners on staff. In particular, Hamilton, despite the incredibly tough assignments on the night, not only scored a wonderful goal showcasing immense individual talents, but had an even strength Corsi For% of 65%, tops among Boston blueliners and indicative of a very special player.

–          A common issue in Toronto is the lack of ice time given to the team’s fourth forward line. On three occasions already this season, the team has gone so far as to dress only 11 forwards and seven defensemen, a tacit admission of lack of faith in the available players. Your intrepid reporter has long held the opinion that the over-reliance on the first line has led to fatigue issues that hit the team late in periods and in games. I had the chance to address the issue with one of the players directly impacted by this approach in James van Riemsdyk. His reply began with a denial, but morphed into a plausible reason that may ultimately have positive and negative effects.

“It’s tough to say. I know how I train in the summer and I’m ready whenever the coach calls me, I’m going to go out there. I feel fine, I feel like I have legs…it’s a double edged sword. There are times when you say you split it up good, but you get tired when you don’t get out. There’s a fine line between feeling like you’re into the game and splitting it up. I haven’t felt winded.”

–          In contrast to JVR’s statement, the other Maple Leafs who were available to the press after the game (Lupul, Kadri and Robidas) all mentioned a lack of compete by their squad in this game. When Randy Carlyle faced the press, he placed the blame on the phenomena of “squeezing the stick” a common occurrence by a trailing team. He also indicated that the Leafs should play more like the Bruins, at least in as much as needing to hold on to the puck for longer stretches. While Carlyle has often been derided for being a very old-school coach and lacking even basic understanding of the lessons of advanced metrics, his comments seem to show that some insight has seeped through. On the other hand, he seems to be at a complete loss for how to implement the basic concept of puck possession. Whatever the solution ultimately is, the current approach isn’t it.

Leave a Reply

Your email address will not be published. Required fields are marked *