Earlier this season I was sitting in the hockey offices at Yost waiting to do an interview with a player when I heard Mel Pearson's voice from around the corner. He was sitting down with the excellent beat writers from The Michigan Daily, and the one question I could hear was about goaltending. One of the writers asked about Strauss Mann pushing Hayden Lavigne and I remember thinking in that gap between question and answer that I better catalog this information in long-term memory, as it would give a good starting point to refer back to when we finally saw Mann in extended time in a couple years. Lavigne may have some ocassional issues but he backstopped Michigan to a Frozen Four, after all. Fast forward three months and our database has more data on Mann than Lavigne this season. Daily writers are really good, man.

It's worth noting, though, that it looked like Michigan was entering the season with something approximating an every-night starter for the first time since Shawn Hunwick started 37 of 39 games in 2011-12. Lavigne took the reigns and was steady in the second half of the 2017-18 season, finishing the regular season with a .908 save percentage and a 2.81 GAA. His numbers dropped a bit in the postseason (.883 save percentage, 3.00 GAA), but some of that is explained by the defense and the potent offenses Michigan faced en route to the Frozen Four. The one goal that you're most likely to remember--and one that I'm not inserting a GIF of because I don't know if the broken jump issues has been fixed and I don't want to dissuade you from continuing to read--Lavigne still got a small piece of as he tried to push across the crease.

Enter Strauss Mann. A 20-year-old freshman, Mann is a product of the Connecticut high school circuit who spent one season in the USHL. It was a heck of a season, though, as Mann posted a 1.86 GAA and .932 save percentage in 34 regular season games and a 2.09 GAA and .932 save percentage in the playoffs, a run that ended in a Clark Cup championship and USHL third-team All-Star selection.

It took two games for Michigan to turn to him and see what he could do at this level, and the situation the staff threw him into says quite a bit about both what they think of his talent and what they think of the goaltending situation. Lavigne gave up five goals against Vermont and five goals against Western Michigan in the first two games of the season (at Yost, no less); Mann got his first start in the second game of the Western Michigan series, which was in Kalamazoo against a ranked opponent.

Lavigne and Mann have bounced in and out of the starting net from there. No one has started more than three games in a row, with Lavigne's longest streak coming at the end of November and early December with 1-1 and 2-2 ties against Wisconsin and a 4-3 loss in East Lansing. Mann started the next three games, 1-1 and 2-2 ties against Michigan State and Minnesota and a 4-3 loss to Minnesota.

​

[JD Scott]

As I've mentioned in previous posts, our shot level is meticulously curated by David, but it's missing a few games. With the Lake State series out (which, meh) as well as a Wisconsin and MSU game, we're missing a five-goals-allowed performance from Lavigne up north, a four-goals-allowed performance in East Lansing, and a game in which he allowed just one against Wisconsin. We're also missing a game in which Mann gave up three, and all of that is without considering shot attempts. 

Turning to what we do have, Michigan has allowed 33 goals on 572 shot attempts. Of those, 25 have come from zone 1, the area right in front of the goaltender (see last post for the zone chart). Mann has allowed 17 of those 25, and he's only allowed 19 goals in our data. If it's going in on Mann, it's coming from right in front of him. Here we have one of the two goals in our set that he's allowed that wasn't from zone 1, and it's a matter of leaving the five hole open as he tries to seal the post and pull his right leg in.

That certainly seems like a trend, and it's easy to point at something like that and the basic stats (.882 save percentage, 3.20 GAA) and say that Mann's doing a poor job on shots from the most critical area of the ice. It's never that simple, though. In the GIF below, there's no way Khodorenko should regain possession of the puck after getting hit entering the zone yet he does. Then he not only spins away from Van Wyhe to retain possession in the corner but finds the trailer while doing so, threading a perfect pass through a forrest of lumber. The pass is so unexpected it twists a Michigan defender to the ice, leaving Mann one-on-one with a shooter in the low slot. The shot goes wide, but you can see that Mann's in position to stop anything that isn't the snipe of all snipes. 

And here Mann again has to deal with an unchecked shooter in the slot, and this time the degree of difficulty is amped up as he has even less time and space to read the pass and stop the shot, plus he has to brace for contact. Mann does a great job squaring and sealing his pads on the ice.

Mann's been a surprise, but this is still a competition for a reason. Here it can in part be explained away by Slaker falling, which allows Penn State a running start into the offensive zone. They find the trailer, who splits two Michigan defenders. Boka is playing the cross-ice pass and leaving Mann the shooter, and Mann whiffs on the rising puck.

Then there's this, which is another one that's set up by a defensive mistake. Independent of that, it seems like the Very Unfortunate Situation with the lowest degree of difficulty relative to the other Very Unfortunate Situations based on time to track the puck and shot location. Unless, of course, Mann's being screened by his teammate who's floating through the slot trying to get in position to block the shot. Hockey!

Mann's not the only one who's had the occasional issue tracking shots. Here Lavigne's put in a similar situation to all of the ones Mann is in above, and you can see about 1.75 seconds in that he's got a clear read on the puck and think it's going glove side only to be surprised when it hits the net blocker side.

It happened again in the same game, with this square-to-the-shooter-trickle-through reminiscent of some of last season's flubs.

Of course, like Mann, Lavigne balances these with moments of brilliance. Here a failed clearing attempt turns into an excellent stop while heavily screened, and the rebound turns into a failed clearing attempt and another and then a point-blank shot that's kept out by the heel of Lavigne's stick.

Lavigne had an excellent series of stops after Michigan State forechecked Cecconi into oblivion and gained control. Not only does he have to come off the post but he has to track a pass to the high slot, make a stop through a bunch of traffic, then track the rebound. He does well coming out to challenge and sticking with the puck outside the crease.

Taken together, it's easy to see why Michigan's in the situation they are. Both have had moments where they look like they could lock down the starting role, and both have had moments where they've missed on what one would think should be relatively routine saves. It's also clear from the clips above that the defense is prone to breakdowns, and even if it's only happening a few times a game it's an issue when both netminders are in the boom-or-bust mold.

For what it's worth, our numbers give an edge to Mann. Lavigne has allowed a goal on 6.7% of shot attempts he has faced while benefiting from a higher percentage of blocked shots (29.3% to 24.5%) and misses shots (26.9% to 23.6%). 

My amateur scouting assessment is that Lavigne is a big body who blocks pucks and is generally in the right position to stop the first shot but isn't that athletic and sometimes gives up baffling goals when he misjudges puck flight, whereas Mann is more athletic but smaller and less positionally sound (e.g. more prone to push across and make a stop but also more prone to start that push late and give up something five-hole as tries to get across).

Competition isn't a bad thing, of course, as long as fear of making a mistake and losing the job doesn't cause somebody to start gripping their stick a little too tight. One counter to that is Steve Shields, Michigan's goaltending coach, who has a wealth of experience handling this exact situation. All this is to say I have absolutely no idea what happens in the second half. At this point it seems reasonable to expect that the first-half issues that led to a rotation (goalies seldom but nevertheless allowing baffling goals, the defense hanging them out to dry a few times each game, etc.) will remain and that no one player will entrench himself as the starter.