Canada’s Very Questionable Decision

“As the game gets tighter and tighter, it makes it even more important, especially for right-handers, because of the fact that right-handers, we don’t have enough in the NHL, so they tend to always play on their forehand side. There’s way more left-handed guys playing on their off side who have spent their careers doing it.”

That’s what Mike Babcock had to say about the more-than-ever relevant discussion, defenseman and the side they play on. For as long as one can remember, the norm is that lefties play on the left side and the same with righties. Easier to collect the puck off the boards, push players to the outside, etc. It’s one of Babcock’s biggest beliefs, it seems.

Recently, a few in the stats community made an attempt to quantify exactly how much it matters. All of them found, as expected, playing on your own side does in fact help. That’s not what we’re looking for. What we are curious to know is how much better does a defenseman playing his off-side have to be than his opposite-handed counterpart to exceed his results.

Domenic Jr. Galamini (@mimicohero) wrote a great piece you can read here which touched very well on the debate. These were his findings:

 

“It turns out that an unsuitably handed defenseman must have a CorsiRel that is greater than or equal to 6.2* Corsi events / 60** better than a suitably handed alternative in order to be the better option to pair with a partner-less defenseman on the roster.”

 

*Was later changed to 6.83

**CorsiRel

What does this mean? Well, if you have a left spot open, unless a righty is 6.83 RelCD/60 better than the left-handed option, you’re better off with the left handed-defenseman. To put that into context, say you have a left spot open. You can have Aaron Ekblad, who won the Calder Cup a year ago, or Dion Phaneuf. The logical answer is, to the surprise of most, Phaneuf. Phaneuf’s RelCD/60 was 0.98, 3.11 less than Ekblad. Since 3.11 < 6.83, Phaneuf is expected to have better possession.

Shifting focus, Canada’s defense was a big controversy this World Cup. Subban, Letang and Giordano were among the names left off. The idea we get is that Subban and Letang were stuck behind what Team Canada thinks is a great right side. We have no clue about Giordano.

The tension heightened when Keith was pulled off the team thanks to injury and needed to be replaced. Perfect time to fix our mistake and put Subban in right? That’s a big fat no. Instead, Jay Bouwmeester was slotted in to play the Chicago defender’s spot. Here we’re gonna analyze if it was really worth it to put him in over Subban and company. First off, let’s check out Bouwmeester.

Jay Bouwmeester

The Blues defender’s biggest asset seems to be his experience. The 32 year old has played in two olympic games, one world cup, three world championships and three world juniors. His play, however, is far from eye-catching. It is incredibly surprising to see a defender who hasn’t put up more than 20 points in the past two seasons playing on Team Canada. He doesn’t play a ‘gritty’ game, which adds to the question marks surrounding his addition. Add his xGF% relative, a disappointing -3.56% , and you’re left as confused I am. He shouldn’t be on this team. Who should have replaced him?

P.K. Subban

The former Canadiens star is just one year removed from being on the NHL’s first all-star team. At 27, he’s solidified his role as one of the most dynamic and exciting players in the league, for better or for worse. Being fourth in 5v5 points since 2014 out of defenseman will do that for you. He was average in terms of possession, at -0.11 xGF% relative. If we check though, his Corsi differential relative was 13.22 corsi events per 60 better than Bouwmeester, so even playing a left-handed spot, he would have probably had better possession. In simple terms, Subban would’ve been better than Bouwmeester in about every way.

Kris Letang

Letang has been the cornerstone of the recently-crowned Stanley Cup champions’ defense for years. At 29, he’s at the tail  end of his prime, but still going strong. He’s often been shadowed by stars Crosby and Malkin, but you should know him as one of the best defenders in the game. Letang had a +7.9 corsi differential relative per 60, even better than Subban, and 17.74 corsi events per 60 better than Bouwmeester. He produces more than a point per 60 minutes of 5v5 play, so he’ll rack up points too. Letang also wins against Bouwmeester.

 

Mark Giordano

Finally, if you’re not convinced they can do well in the left position, how about Giordano? Him being snubbed makes no sense – he’s better than Bouwmeester in every possible way. He puts up points more than twice as much and is levels above him in terms of possession. If you’re crying about experience, they’re the same age, too. To put his superiority into context, Giordano was third out of all defenders in CF% Rel and 12th in points per 60. Crazy.

 

Conclusion

All in all, Bouwmeester should not be in this team. Letang and Subban would’ve done better on the opposite side than Bouwmeester does his own, and probably produced more offense, too. If you don’t believe me, than you can sleep knowing Giordano is miles better, too, and he is left-handed. To be fair, a depth defenseman probably doesn’t mean much. However, Subban and company can be game breakers, and you can always use just one more goal.

All stats are 2014-16 and are 5v5 score, zone and venue adjusted from corsica.hockey

Advertisements

Asset Management and Alex Tanguay

Cheap, low-risk contracts. Everyone loves them. So much so, that teams like Toronto have abused the left-overs of the market in the past years. Players like Daniel Winnik, Mike Santorelli and Shawn Matthias have landed them picks and prospects like it’s NHL 16. While they probably won’t do that this year, they’ve presented a model of asset management that teams should consider doing themselves.

Enter unrestricted free agent Alex Tanguay. The 36 year old split time with the Colorado Avalanche and Arizona Coyotes last season, notching a respectable 35 points in 70 games. He was part of the shipment Colorado sent to Arizona for Boedker, but was likely mostly just a way to make the money work. However, even as a relatively old player who had to start again with a young, struggling team, he didn’t do poorly. 13 of his 35 points came in the last 18 games, which he played with Arizona. Anyways, I really like him, and here’s why.

Screen Shot 2016-08-08 at 1.10.02 AM.png

I want you to look at his production (that’s his goal scoring, playmaking and individual production on that chart). He racks up points at about the rate of a first-liner. Yeah, I’m surprised too. You may not understand how 35 points in 70 games makes a top player, but that’s not the whole story. Tanguay did not get many minutes – over the course of the season he played about 12.5 5v5 minutes a game. When converting his production into a per-60 rate, you see his numbers stand out a bit.

Tanguay put up an impressive 1.7 points per 60 minutes of 5v5 play. To put that into context, that’s ahead of players Stamkos (1.69), Oshie (1.68), Logan Couture (1.61) and Nathan Mackinnon (1.63). That kind of offensive addition can help a lot of teams in different situations. He could be a low-cost power-play addition to a contending team to give that boost, or a minute-logger on a basement team looking to ship him for assets at the deadline.

If we take a specific look at how Tanguay did after he was traded to Arizona, it gets even more impressive. Before I start giving numbers, I want to stress that this is a very small sample, just something to give an idea on how he might do in the right system. In the time he spent with Arizona, he racked 3 goals and 4 assists at 5v5 – a 2.57 points per 60. A season at that pace would place him tied for sixth in the league. Pair that with a 76.5 GF% and you have some insane numbers. Again, he’s not that good, it’s just worth noting.

In terms of possession, Tanguay is very average. He played on two of the worst possession teams last season, Colorado and Arizona, and his corsi stats were almost identical to his team. He had a +0.4 CF% RelTM (relative to his teammates without him), so he wasn’t bad, and he wasn’t great. In terms of goals, however, he stacked a 58.0 GF%, +11.2% relative to his teammates. Remember, however, goals are often too small a sample, so don’t put much stock into that.

One thing I noticed about Alex Tanguay is his insanely high shooting percentage. He has 520 shots in 556 games at 5v5, less than 1 shot at even strength play a game. Pair these stats as well as fellow writer Chris’ (@CorsiGuy) observations, he only shoots when he has a great opportunity. Whether this will aggravate fans and coaches when he doesn’t capitalize on good chances, or maybe make the players around him better by passing so much is unknown. In any case, it’s just something you should know – he had the single highest shooting percentage of the whole NHL since 2007.

To be clear – I don’t think Tanguay is a first-liner, or anything along those lines. If anything, he’ll probably take a decline next year. What he still is is a cheap forward who will in all likelihood put up points, to either squeak a team into the playoffs or fill in spots on bare rosters. Best case scenario – he produces and either helps a contender or nets you a couple of picks/prospects. Worst case he comes off your books in a year. In the end, Tanguay is a great UFA if your heart desires someone who can do that. He made 3.5 million for the past 5 years, but you can expect a huge pay cut for him, especially considering he still isn’t signed. In all likelihood, he’ll go to a training camp or two and hopefully find a home.

HERO Chart courtesy of Own The Puck

 

The Predators and Forward Deployment

Peter Laviolette, the coach of the Nashville Predators, did something a bit peculiar last season. He had an idea on how to use his lines, and he used them very specifically. What I mean by this is he had lines meant only for offense and the same for defense. Let’s take a look at some of the lines he deployed last season.

Screen Shot 2016-08-04 at 11.26.08 PM.pngChart courtesy of www.leftwinglock.com

This chart, if you couldn’t tell, shows us the most frequented combinations by the Preds. Now clearly this doesn’t mean too much because Laviolette, as any coach would, changed his lines constantly. What we can still do is look at some of his individual lines. His most commonly played line this season was Forsberg, Smith and Ribeiro (keep in mind this is probably just because all three players were there all season, unlike someone like Ryan Johansen). Where they were played stands out a bit. Thanks to some great stats from Puckalytics.com, we can get a full grasp on it.

When all three were together (that was for 469 minutes) they were the ‘forget defence’ line. They had an 89.6 OZone%, meaning 89.6% of their faceoffs were on the offensive half of the ice. Yeah, that’s crazy. Only 6.4% of their faceoffs were in the full-on defensive zone (the 2 dots closest to their own net). It had its obvious effects, as you might imagine. Separated, they all had 53-54 CF%, but ensemble they had a whopping 59.7%. In terms of GF%, they were all the way up at 64.7%. But hey, that’s one side of the story.

On the opposite end of the spectrum, you have Watson and Gaustad. I’m going to look individually at the two, because the duty of their third partner was essentially split in half by Salomaki and Watson. It won’t skew the numbers too much, because they were played pretty equally regardless of the complementary player. Anyways, they started a depressing 7.4% of their faceoffs in actual offensive zone. The rest were either in the neutral zone (31.2%) or the defensive zone (61.5%). Basically, these two were used to get the puck out of the defensive zone, or at very least out of their net. A lot of this is probably due to Gaustad’s face-off percentage, a very respectable 55.3%. The tough minutes hurt them – they had a less than ideal (and by that I mean terrible) GF% of 23.1.

However, like any coach, Laviolette had his lines he’d play everywhere. His second most used line (which in reality is his most, because of the Johansen situation) was Neal, Jarnkrok and Johansen. This trio was leaning more towards offense (58.4 OZone%), but they played their fair share of defensive responsibility. They took 26% of their faceoffs in the defensive end, compared to 36.5% on offense. This push forward didn’t really help them, though. With a CF% of 50.5, it seems less than ideal, but not devastating.

The Preds are pretty unusual in their deployment, yes. The three highest OZFO%’s in the league all belong to Nashville: Ribeiro, Forsberg and Smith. How it worked out? Pretty decently.  A 50.9 GF% isn’t too outstanding, but a 52.5 CF% landed them 4th in the league. We’ll see if Laviolette continues with this strategy next season, and if he does, we can analyze it all over again.