When Andrew MacDonald was acquired from the New York Islanders for two draft picks, there was a large outcry from the advanced stat community. MacD's possession metrics were horrible on the Island. Many wondered if this was yet another medicore possession d-man being added to a mediocre defense in Philly. The counter argument goes something like this.
MacDonald was over-exposed on a bad team. He'll be much better in a reduced role against easier match-ups on the Flyers.
Its been 10 games since MacD's acquisition, which is a reasonable amount of time to start judging his performance. Charlie O'Conner of Broad Street Hockey has confirmed that Berube has given MacDonald much softer minutes. But how do his numbers look in those minutes? To put it bluntly, his numbers have been atrocious.
MacDonald's shot metrics haven't improved. In fact, they've gotten worse.
Andrew MacDonald was -9% CF% rel with NYI. He's -11% in 8 games with Flyers. http://t.co/LdYjdpxsda— Extra Skater (@ExtraSkater) March 23, 2014
Lets take closer look and see how MacD is affecting his teammates.
Obligatory small sample size warning!!!!
WOWY CF% Forwards
Looking at forwards who've spent +25 minutes TOI with MacD show's that he negatively impacts almost everyone, with the notable exception of Sean Couturier.
WOWY CF% Defense
Looking at his most common defense partners, we see a similar pattern. He has a dramatically negative impact on Coburn. He hasn't improved Luke Schenn. In fact, he's made him slightly worse. So far the best partner for MacDonald has been Mark Streit.
In his first 10 games as a Flyer, MacD's softer usage has been demonstrably ineffective. He's still giving up more shots against relative to his team. And he's still making many of his teammates worse. Now lets look at scoring chances.
Is he giving up quality shots?
Many might argue that its possible MacDonald is giving up shots, but they aren't good quality scoring chances. As it happens, I've been tracking scoring chances for the Flyers this season. To answer this criticism, we can look at how the defense has performed in the last 10 games with my chance data.
The Flyers third pair has been bleeding quality chances against. Opposition teams have controlled a staggering 65% of the scoring chances with MacDonald and Luke Schenn on the ice. As you can see, Timonen and Coburn have done far better while getting substantially tougher assignments.
A familiar refrain of the anti-stat crowd is that proponents need to "watch the games". Setting aside the insulting implication that analysts don't watch games, lets watch some film and hone in on what makes MacDonald a mediocre third pairing defender.
Checked off by Patrick Kane
Blown Coverage against Bickell
Beaten Wide by Bodie
Turnstiled by JvR
Beaten by Berglund
Deked by Schwartz
Egregious PP turn over leads to shorty against.
Beaten to the net by Moore
Disrobed by Nash
Crushed in the Corner
MacDonald is easily pushed off the puck by strong forechecking and physical forwards. He turns the puck over when pressured and can make clumsy decisions even when he isn't being directly pressured. Both Luke and MacD are very vulnerable on the rush. I've shown you multiple examples of MacD getting beaten wide with speed. And to top it off he's prone to losing his assignments from time to time.
With all of this said, he's not the worst defender I've ever seen play at an NHL level. His first pass is as good as advertised, and he is certainly a plus skater. Unfortunately, these positives are often outweighed by his negative attributes.
But couldn't you do that with any defender???
Sure. If you go back and scout the last 10 games you can probably find similar mistakes being made by every Flyers d-man. This is why eye test scouting is often biased and subjective. People see what they want to see with a player, no matter how effective he might be otherwise. Case in point, Flyers fans would often lament about Jeff "High and Wide" Carter and Matt "POS" Carle. Its easy to develop tunnel vision after you've seen someone make a few glaring mistakes or conversely a great scoring play. This is why looking at aggregate statistical scouting has proven valuable time and time again. Every defender has made mistakes over the past 10 games. But MacDonald and Luke Schenn have made significantly more.
Ten games is still a relatively small sample and its possible that MacDonald improves a bit with time as a Flyer. However, its clear that his poor metrics are not because of his extremely difficult usage in New York. There is some substance behind the numbers.
The Real Problem
My biggest issues with the MacDonald acquisition remain the benching of Erik Gustafsson, the loss of draft assets, and the issue of money. Expending assets to acquire yet another passable 5,6,7 d-man while benching the best performing 3rd pairing guy on the roster is inexcusable. And if reports are correct that MacDonald turned down a multi-year contract worth 4 million AAV, its probable that MacDonald is looking for an even better deal from the Flyers. Signing MacDonald to a long term contract with Top-4 money is invariably a mistake. His on-ice results simply do not justify a contract extension worth 4+ AAV.
MacDonald hasn't improved the third pairing since his acquisition. In terms of on-ice results he is a clear downgrade from Erik Gustafsson. In the 59 games before he was scratched, Gustafsson was a +16 in scoring chances, putting up substantially better possession numbers, and was far more effective at preventing chances against. Even more impressive, Gus's scoring chance numbers were good enough for 2nd on the Flyers defense, behind only Kimmo Timonen. Why are the Flyers sitting their most effective third pairing defender in favor of Andy MacDonald?
Thanks for reading, and #FREEGUS!
EDIT: I've made a few changes. The WOWY graphs are now using 5v5 zone start adjusted Corsi instead of straight up 5v5 Corsi. Thanks to OrangeNBlack for pointing out the error.
I've also made the small sample size warning more noticeable. Seriously....I warned you.