BSJ Analytics: Why the Bruins should trade for Rangers defenseman Ryan McDonagh

BostonSportsJournal.com has always been about covering the sports and teams you love in non-traditional ways to look at things in a different light. Writing with analytics has long been something we've wanted to bring to BSJ, and this seems to be a good time to get that rolling. If you're interested in contributing to a weekly analytics column on each of the four major sports, send us samples of your writing and ideas for future columns to greg@bostonsportsjournal.com.

Of all the names the Bruins have been connected to with the trade deadline coming up, Rangers defenseman Ryan McDonagh is the biggest. This is the big boy. The book on McDonagh is he's a top-pairing guy. The eye test will tell you that. Anyone who's really watched him will tell you that. But I'm a little bit of a nerd, so a guy's reputation isn't solid enough evidence. I want some hard numbers if I'm going to make a franchise-altering trade.

So let's dive into the numbers on McDonagh, which show why he'd be a worthy investment for the Bruins.

How good is McDonagh?

I've spent too many hours analyzing the numbers behind every possible trade deadline target for the Bruins, but here is the skinny on McDonagh. The first thing you need to know is that in his time with the Rangers, he was paired with some stationary traffic cones. In the past two seasons, he's spent over 80 percent of his total ice time (TOI) at five-vs-five (5v5) with Dan Girardi, Nick Holden (newly acquired by the Bruins), and Marc Staal. Woof. Here are his stats from being on the ice with the three of them.

[table id=14 /]

Quick explanation for these numbers. TOI is just time on ice. Corsi is shot attempts. If your team has a 40 percent "Corsi For", that means that only 40 percent of the shot attempts taken were by your team. It’s the most predictive stat for future success, due to the giant sample size and low variance, and is often used as a proxy for possession. xG is similar to Corsi, but every shot is weighted by its likelihood to score a goal. Because you’re adding in more measurements, it’s less of a predictive stat and more of a descriptive stat. Finally, the “rel” prefix indicates how much of a difference your Corsi/xG was from your teammates.

McDonagh's Corsi (shots on goal + missed shots + blocked shots) is negative with all of those guys. The Rangers have been dominated in the possession game when he’s on the ice, especially those numbers with Girardi. Truly elite guys don’t get caved in like that, no matter how hard the minutes are. That -5.38 percent is really alarming. It means the Rangers were more than one standard deviation worse with their top pairing on the ice than when they played their other defensemen.

He does much better when you look at stats that take shot quality into account, managing to get above even with Staal and Holden. No saving Girardi, though. A giant change between CF% and xGF% is slightly concerning because influencing shot quality isn’t nearly as repeatable as influencing possession. But, man, has he been tanked by bad defensive partners.

Stats that specifically try to isolate singular players from their team show him in an even better light. Here’s how he stacks up in two different advanced WAR models, one by Manny Elk and the other by Dawson Sprigings, in charts made by Sean Tierney.





WAR and GAR are wins above replacement and goals above replacement.

These are from last year’s stats, but these numbers tell a pretty stunning story. Dawson’s model has McDonagh at ninth among defensemen last year, and Manny’s model has him at third. That’s a pretty stunning revelation considering his Corsi stats. In both models, his contributions to shot rates are considered positive. That means his partners and deployment were really heavily dragging him down. The other interesting nugget to take from his WAR stats is his impact on penalty differential.

Both models have him taking less and drawing more than he’d be expected to. That could be very useful on a Bruins team that’s historically struggled when it comes to penalty differential. Moral of the story is that McDonagh definitely is as good as his reputation would suggest. But you have to dig through the numbers. That's not usually the case with truly dominant players. Elite guys are good no matter how you slice the numbers. McDonagh isn't that, and that's fine. He can still be a game changer, but don’t expect Erik Karlsson.

Where does he fit in now?

First off, let’s get some numbers to work with. Here are some charts from the 3 zone project, which tracks individual plays made by defensemen. This information is useful because we can see where certain players struggle or excel. It’s always a good idea to try to create balanced defensive pairings, where players make up for each other’s weaknesses.







McDonagh's bread and butter, at 5 on 5, is his transition game. He's more of a puck mover than a traditional offensive defenseman, very similar in playing style to Bruins' rookie Charlie McAvoy. They both excel in the neutral zone, and that’s reflected in the numbers. They can enter the other teams’ end, exit their own end, and defend against entries with the best of them. But neither are incredibly successful in the offensive zone.

Look, a lot of people are projecting McDonagh as a potential partner for McAvoy if he’s added. I get it. Play your best right-handed defenseman with your best left-handed defenseman. I get it, I really do. But man, I don’t like it. Playing them together feels like it would be such a waste. They can't both make that first pass out of their end, and neither are super great at cycling the puck and generating offense from the point. I don’t think I’d play them together.

But McAvoy and Chara together? They’re money in the bank. The two of them make up for the weaknesses of the other. McAvoy has trouble generating offense from the point, but Chara's shot is still fierce and he's still very good at cycling the puck. And while McAvoy's no slouch in the defensive zone, Chara's still one of the scariest players in the league around his own net and on the boards. Chara's trouble lately has been making those key passes and skating plays out of his own zone and into the other team's zone, and that's where McAvoy's been at his best.

If you play McDonagh with McAvoy, are either player's best strengths being utilized? Will you start to see cracks in either of their games? Who do you play with Chara? The easy answer is to play Chara with Carlo again, but I'm betting Carlo is gone in any prospective McDonagh trade. And without him, the Bruins really don't have anyone with a good transition game through the neutral zone who can play on the right side. The next best option for Chara is Kevan Miller, and that’s not too enticing. McAvoy's really the only right-handed handed defender they have who can let Chara shine.

For my money, I bet that if McDonagh's added that McAvoy and Chara stay together as a pair and you end up with a 1A/1B Top Four. If Carlo's still in the mix at that point, I'd like to see him as McDonagh's partner. If not, I think it'd be very interesting to see McDonagh play on his off side next to Krug. Those two could be dominant if given predatory starts. That way you also get to preserve the criminally unheralded Matt Grzelcyk/Miller pairing, which has been dynamite for most of the year.

But what about long-term?

The real big elephant in the room, after any sort of McDonagh trade, is re-signing him. He'll be 28 when his deal is up after next season. He'll want something like eights years and more than $8 million annual average value, and he'll likely get it on the open market, if not more. He'll be in his prime as a defenseman. The worst-case scenario is he starts to seriously decline around 32-34 and by that time, you'd only get 2-4 years of serious overpayment. Top pairing defensemen also have a tendency to age in goofy ways. Guys like Brian Campbell and Chara had their peaks closer to 36 than 26. It's really hard to imagine re-signing McDonagh going badly. If you give up a really good package for him, re-signing McDonagh will probably end up as a victory condition for winning the trade.

The problem is affording it. McAvoy will be up at the same time, and he'll want a very similar deal. You'd need to clear out a bunch of cap space to fit both deals under in the same offseason. Both players combined will probably need $14-16 million of cap space. Then you've got the other players who are up. Danton Heinen, Austin Czarnik, and others could play their way into really hurting your cap. I'd try to get $20 million-plus in cap space ready for the summer of 2019.

The Bruins will need to cut some fat in the next two years to afford these deals. Chara and Ryan Spooner will probably be gone and net you a nice $7 million. But that's not enough. Guys like David Backes, Matt Beleskey, and probably David Krejci will need their deals moved. You're going to have to really remake your team to figure out the cap situation when McDonagh's up.

What does it all mean?

Probably what you thought it meant. McDonagh's a game changer, in every way. He's a legit No. 1 defenseman who gives you an insane upgrade on talent. He'll be a giant rock in your blue line for the next two-to-10 years. And he'll probably destroy your cap. But, it's hard to imagine it not being all worth it in the end.

All stats are from corsica.hockey, charts by Sean Tierney and Cory Sznajder, and WAR models from Dawson Sprigings and Manny Elk