The Shifting Depths of Frederik Andersen
Frederik Andersen was brought in this summer to extinguish the Toronto Maple Leafs Annual Goalie Panic. Instead, he stoked that fire to almost unprecedented heights over the month of October. His numbers were awful, he looked lost, and as time went on, media-proposed excuses about injury and missed training camp became less and less convincing. Something was wrong with Frederik Andersen, and Martin Biron, a former NHL goalie and current analyst for TSN, was clear about what it was:
Here’s my report on Andersen. Way too far outside of his crease being way too aggressive. Relying only on size and not on save skill. https://t.co/YWCrY8yPor
— Martin Biron (@martybiron43) October 20, 2016
Finding evidence for Biron’s claim was easy. Andersen’s depth was so extreme that it was hard to miss:
Andersen’s first game of the season saw him taking up residence in the white paint all night long. In this sequence from the three-on-three overtime, he begins by challenging the puck carrier outside the white. As a result, he arrives late to the pass receiver, and even later to the next pass, leaving him off-angle when the game-winning goal is finally scored.
His most egregious foray into the wilds of 80s goalie depth came against Winnipeg in his third start of the season:
The Jets have a six-on-four power play, and when a pass moves laterally from the boards into the high slot, Andersen races out to challenge the shooter. One major problem is that he never actually stops and sets. As the puck carrier finishes his fake shot and makes the pass, Andersen is still moving toward him.
Even if the shot had come, this isn’t good form.
Stopping and getting set for shots (especially when you easily have time to take ample depth beforehand) is a very basic priority for any goaltender. If you’re still in motion when the shot comes, your balance and ability to read the shot are compromised. Further, rebounds become much harder to control because the puck hitting a stationary goalie comes off differently than a puck hitting a goalie in motion.
Finally, the closer you get to the shot, the less time you have to react, making anything other than a pure blocking save almost impossible. Of course, the expected shot doesn’t come from the high slot. Andersen is so far out that he has to step around his defenceman to get behind him before moving across. He has an enormous amount of ice to cover in very little time, and comes up far short.
Andersen wasn’t caught out that badly again, but persisted in charging out to challenge shots from the centre of the ice, especially the point and high slot, and especially on the penalty kill. This save against Shea Weber is typical:
As the puck moves low to high, then laterally, Andersen’s initial push takes him near the top of his crease. He then does an additional c-cut to move well beyond the blue. In this case, he is able to (mostly) stop and get set for the shot, making an impressive save. Of course, if Weber had faked the shot, and instead passed to the Canadiens player standing behind Andersen to his right, the net would have been empty.
Moments later in the same game, a very similar play unfolds, with very different results:
The play is almost the same as the previous one, except that the initial pass receiver is closer to the passer, and the eventual shooter (Weber again) keeps the middle of the ice better. As a result, the play develops more quickly, and Andersen doesn’t have time to make his push for extra depth and still stop to get set. He’s beaten cleanly.
Throughout October, Andersen was so committed to challenging beyond the parameters of his crease that it sometimes meant he would wreck against the screen established in front:
The most baffling aspect of Andersen’s commitment to great depth was its utter lack of precedent. Last season with the Ducks, he played the same situations very differently:
Shorthanded, facing a low-high pass to an open shooter in the high slot early last season, Andersen is at the top of the blue, set and ready to make the save.
Next, Andersen has space and time to take more ice as the puck comes back to the point, but he decides to wait patiently in his crease:
Instead of crashing into the screen, Andersen’s tendency last season was to settle a comfortable distance away from it, giving himself room to move and adjust freely. Here, we see him so far back it’s sub-optimal in the other direction:
The reason for Andersen’s sudden, significant, and unsuccessful shift to a far more aggressive challenging style remains a mystery. Well, officially, at least.
In late October, when asked whether the coaching staff had influenced Andersen’s radical style change, head coach Mike Babcock denied it: “He’s obviously had a way he’s done things and it’s important, until we get to know him, that we don’t change anything.” At the same press conference, Andersen himself refused to comment on the issue, stating simply, “I don’t want to go too much into that.”
Looking at the historical evidence, however, we see that Babcock has a history of asking his goaltenders to play at a more aggressive depth within his system. In 2011, Jimmy Howard discussed Babcock’s depth preferences in an interview with InGoal: “I’ve got a great defensive crew with me and I know they are going to take care of that backdoor. Coach [Mike] Babcock stresses that, so you know the goalie has the shooter and you pick up everything backdoor, and that allows not only myself but also Chris [Osgood] to be more aggressive.”
An article about Andersen at NHL.com noted how James Reimer was far more aggressively positioned under Babcock last season than ever before, and how he moved back after being traded to San Jose. The article also predicted that, given Babcock’s history with goaltenders, Andersen would be asked to play further out than usual: it just didn’t expect the difference to be so extreme.
And then, as suddenly and inexplicably as it began, Andersen’s newfound extreme-depth style was no more. The change happened in the span of a single game. After a 2-1 loss to Montreal on October 29 (the game featured above where Andersen was out in the white challenging Weber), Andersen sat out a 5-1 Leafs loss to the Islanders, and then started against Edmonton on November 1.
Though he maintained slightly more aggressive positioning than he had in Anaheim, he wandered into the white only on select rush shots. Gone was the extra push to propel him out of the crease to challenge shooters in the high slot. Gone was the tendency to mash himself into screens to achieve as much depth as possible. It was as though the whole previous month had been some kind of strange dream, or a failed scientific experiment unceremoniously cancelled without notice.
In this game from November 3, Buffalo sets up just the kind of power play chance in the high slot that Andersen had been challenging so aggressively. Instead of racing out, he stops, sets, and waits:
The shift back to reasonable depth seems to have been a permanent one. In this November 26 game, the Washington shooter has the puck in the high slot on the power play. There is no close screen and no clear passing option. This is exactly the situation that October Andersen would have raced out to challenge against, but he simply moves to the top of the blue, sets, and waits patiently:
The silence around Andersen’s shift back to his accustomed depth equals the silence around his initial move into uncharted white territory: no one is saying why, or even acknowledging that such shifts took place. The results, however, speak volumes.
Under the October depth experiment, Andersen’s raw save percentage was an awful 87.6, while his save percentage 4-on-5 was 85.7. Since his move back to the blue, his raw save percentage is an excellent 93.1, while his 4-on-5 save percentage is a dramatically improved 92.7 (all stats via Corsica).
It’s too early to say whether Andersen will continue to have November’s success, but it’s safe to say that he’s finally found his game in Toronto. It was right there all along, of course, where he had temporarily, and seemingly intentionally, left it at the end of last season.