4.375 Million is Not a Reasonable Price For Brandon Sutter

Yesterday
on TSN 1040, midday show co-host Blake Price, whose opinion I am usually
inclined to agree with, delivered an editorial on Brandon Sutter. The gist of
the argument was as follows: Brandon Sutter’s $4.375 million salary is not an
overpayment, by virtue of his goal total, his faceoff prowess, his penalty
killing, and, as the icing on the cake, his willingness to play through injury.

As
respectfully as I could possibly put this, I don’t believe that I could
possibly be more opposed to this argument. While Sutter’s goal total is indeed
worthy of distinction, I would quibble with the value of the rest of the
evidence provided. Indeed, I wouldn’t accept it outright that Sutter is even
benefitting the team.

And so
my theory is going to be the logical opposite of Price’s: No, $4.375 million is
not a reasonable price tag for Brandon Sutter, and I’ll explain why.

I’ll predicate what is going to be a very lengthy essay on Brandon Sutter by saying that I’m not trying to go after Blake Price specifically. After all, as I mentioned, I tend to concur with the majority of his opinions, plus he was nice enough to allow me on his radio show earlier this week (his show, his show, his show). Rather, I just have a special interest in Brandon Sutter – namely that he’s overpaid and his excessive deployment is pumping up what should otherwise be pedestrian counting stats, while in the meantime causing more problems than he solves on the ice.

Nevertheless, Price has afforded me a list of traits that many others have referenced in defence of Brandon Sutter, so they have become convenient points around which to structure my argument. Let’s get into it.

Brandon
Sutter has two offensive assets going for him: his speed, and his shot, both of
which are above average at the NHL level. He’s been termed a “one-shot
scorer” because he “doesn’t need many chances to get one to go
in”. He possesses a sneaky wrist shot that’s shown plenty of success off
of the rush, particularly from the right side, and he’s had his share of tips,
deflections, and garbage goals to pad his total, as most goal scorers do.

All
told, purely based on games played and remaining, Sutter is on pace for exactly
20 goals this season, which is the expectation that management has generated
for him since acquiring him. He has consistently been referred to as a
“20-goal scorer” because he has reached the 20-goal plateau. He has
done so exactly two times in his eight-year career, both times peaking at 21
goals. So I’d find Blake Price’s presumption of “22-24 goals”, as
noted in the editorial, a bit premature. The low end of that range, meant to
indicate something that he can accomplish with consistency, is one higher than
he has ever managed in his career.

This
isn’t to say that goals are bad, or that scoring 20 goals isn’t impressive in
today’s NHL – we all know the Canucks have precious few players capable of
doing it. But it isn’t the total number of goals that he’s piling up, but how
he’s doing it that is concerning.

Brandon
Sutter is third among Canucks forwards in all situations ice time, behind only
Henrik Sedin and Loui Eriksson. He’d have Eriksson beat if he hadn’t missed
Tuesday’s game again Pittsburgh, and is just three seconds behind Henrik
Sedin in average ice time per game. Even with that missed game, Sutter ranks
22nd in the NHL in total ice time (the Canucks inexplicably have four players
in the top 30 for ice time), sandwiched directly in between Jamie Benn and Kyle
Okposo, and not far off of players like Tyler Seguin, Mikael Granlund, and
Wayne Simmonds. Pretty impressive company. In such company, his 15 goals look
rather modest. His 0.83 all situations goals per 60 minutes (G60) puts him
behind 144 forwards with at least 100 minutes played.

Perhaps
more alarming is his 5-on-4 ice time. His 157 minutes with a single-man
advantage ranks 39th among NHL forwards, within just a few minutes of Patrice
Bergeron, Leon Draisaitl and Max Pacioretty. In this company, his four power
play goals are pretty underwhelming. His 1.52 5-on-4 goals per 60 minutes puts
him behind 98 forwards with at least 50 power play minutes.

The
overarching point here is there Sutter’s goal totals have been, in a way,
manufactured by Willie Desjardins insistence on playing him so much. Those goal
totals then reinforce Willie’s notion that Sutter is a strong offensive player,
and Jim Benning’s notion that Sutter is worth north of $4 million. The massive
amount of ice time is obscuring the fact that Sutter is scoring at the rate of
a middle six forward, then run out on to the ice over and over until he becomes
a 20-goal scorer by sheer stubbornness.

That by
itself is not a crime. However, it takes ice time away from teammates like Bo
Horvat, who is a more efficient scorer in any given situation. What’s worse is
that Sutter’s underlying metrics indicate that the more time he spends on the
ice, the further the team slips into the red. But we’ll get to that later on.

Faceoffs
are, to a great extent, overrated by almost everyone in hockey that hasn’t
spent the time digging into or reading up on the subject from a statistical
perspective. Be it a lifetime analyst, a current professional player, or a
brand new fan, faceoffs are seen as vitally important, because their value
appears to be intrinsic. Winning a faceoff means gaining possession of the
puck, and we are told by both the old and new schools that puck possession is
important.

I have
an article in progress on this very subject, so I won’t get too in depth with
it at this point, but the underlying message is this: faceoffs are worth far
less than you think they are. The data indicating this has been around for
some time. Take for instance Michael Schuckers’ assertion that it requires, onaverage, 76.5 faceoff wins for a single goal. Faceoffs have been analyzed in
excruciating detail since then, but you need look
no further than Jeff Veillette’s article earlier this week, in which he
assigned goal value’s to each of Toronto’s centres based purely on their
faceoff numbers.

While a
similar exercise for the entire Canucks roster may be of interest, for now I’ll
give you the goal differential Brandon Sutter has generated due to his faceoff
percentages:

Faceoff data via Puckbase.com

While 9.92 is a decent number of goals over a large portion of the season (it would be worth approximately three standings points, according to Schuckers), we have to remember that if it weren’t Sutter taking these faceoffs, it would be someone else. The far right column of the graphs indicates Goals Against Replacement – a very simple version that fits Sutter’s number against a replacement level centre, given a faceoff percentage of 47.4% (borrowed from Matt Cane’s article on the subject) across the board. This gives Sutter’s faceoff prowess a Goals Above Replacement of 1.53 – just over a goal and a half, or one half of a standings point, roughly three quarters of the way through the season.

If you
find that to be entirely underwhelming, it’s because faceoffs, and faceoff
prowess, are entirely underwhelming. Certainly not something that hockey
executives should be paying extra money to obtain.

There’s
a saying that suggests that all one needs to do to be labeled a “good
penalty killer” is to play a lot on the penalty kill. If the NHL coach
keeps sending the player out there shorthanded time and time again, it must indicate that the player is effective in that role, since coaches want to win.

This is
of course ignores the fact that coaches do a whole lot of things in the name of
winning that don’t necessarily lead to winning.

In this
situation, the perception isn’t entirely wrong. Brandon Sutter actually has
some of the team’s best penalty killing numbers. Although this doesn’t
completely make him good at it, so much as just better than the rest of the
team, which has been largely abysmal when down a man.

Sutter’s
only real competition among centres on the penalty kill is Bo Horvat, seeing as
neither Henrik Sedin, Brendan Gaunce or Michael Chaput has seen significant
time shorthanded: each has each that 30 minutes of 4-on-5 time (with Gaunce and
Chaput closer to 10 minutes), compared to Sutter and Horvat’s 107 and 99
minutes respectively.

As for
Horvat, it pains me to say that he’s been terrible on the penalty kill. Of all
Canucks players with at least 20 minutes of 4-on-5 time on ice, Horvat has the
highest (read: worst) rate of unblocked shots against, expected goals against,
and actual goals against. Sutter, meanwhile, is bested only by Loui Eriksson in
unblocked shots against per 60 minutes (FA60), and by a handful of players in
expected and actual goals against per 60 minutes.

Of
course, that’s only relative to the rest of the Canucks. Against the backdrop
of a team with the seventh worst penalty kill in the NHL (78.6%), it isn’t hard
to look impressive. Against the rest of the NHL, Sutter looks decidedly more
average.

There
are 65 NHL forwards with at least 50 minutes of 4-on-5 time this season that
have allowed a lower rate of unblocked shots per 60 minutes, out of a sample of
170. There are 83 such players allowing a lower rate of goals against, and 136
such players allowing a lower rate of scoring chances against.

Compared
to the rest of the NHL’s penalty killing forwards, Sutter is performing just
slightly above average in some areas, like suppressing unblocked shots and
shots-on-net against, but slightly below average where it counts, like
suppressing scoring chances and expected goals against.

All in
all, Sutter appears to deserve the distinction of “adequate penalty
killer”, which in and of itself is not a bad thing. But as with faceoffs,
it’s not something that you should be paying through the nose for. After all,
the vast majority of those players that are suppressing power play offence more
effectively than Sutter are doing so for much less money.

This may
be far and away the toughest trait to quantify, for a variety of reasons. While
we have ways of assessing the how a team is disadvantaged (and in some cases,
aided) when players are out of the lineup thanks to websites like Man Games Lost, we have a much tougher
time analyzing players who are playing hurt. This is because our main source
of determining injury in large sample is when players miss games due to
injury. If the player is playing the game, we don’t know that they’re injured
unless we’re following closely – easy enough to do for single cases, but
impossible to do for every player, every game, every year. That doesn’t make
for very reliable testing under the scientific method.

This
doesn’t even begin to cope with the fact that injuries can be perceived on a
continuum, and some are obviously worse than others. Playing through a bone
bruise is surely not the same as playing through cracked ribs, which in itself
is very different than playing through a broken foot or hand – and yet each
injury most certainly effects a player’s performance.

One of
the only indicators we have of how much pain a player is playing through is by
self-report, in addition to their general pained expressions. Of course, each
of these is entirely subjective and directly tied to a player’s pain tolerance,
which we have no method of measuring.

In the
end, the only thing that we are capable of measuring is a player’s performance
on an individual basis. Thursday night in St. Louis, Sutter scored a goal and
won 50% of his faceoffs, which certainly makes it seem like a success. Granted,
the goal was a deflection (deft as it was) that had no business beating a
goalie like Jake Allen through the five-hole, and we’ve already gone over the
true value of faceoffs. Even then, Brendan Gaunce, who was in the press box
last night, is 51.0% in the faceoff circle this year. Additionally, he also deflected a goal into his own net, which should probably make his goal a wash.

Outside
of the actual results, Sutter was having an off night, even by his standards.
An injury in his right wrist made it difficult to put any velocity on his shot
(as pointed out by John Garrett in the broadcast), and likely resulted in him
taking fewer faceoffs than normal. He was replaced Jayson Megna for the most
part on the top unit power play, and while he was on the ice when the Canucks
pulled their goalie for an extra attacker in the waning minutes, trailing by a
goal, his most noticeable contribution was an errant pass that fled the
attacking zone and killed the play.

Hockey
culture tends to deify players that “gut it out” and play through
various injuries and ailments, without really ever asking this simple question:
at which point does the player’s injury make him less valuable than the next
available player on the roster? Even if the question is asked, a satisfying and
scientific answer is never attempted.

It’s
just one more thing that I am hesitant – bordering on resistant – to put a
dollar amount on in contract negotiations.

So we’ve
gone through each one of the individual traits and discussed why I would not
be placing a premium on them. So what is Brandon Sutter’s value?

Before
answering that, we’ll go back to the fundamentals of hockey analytics – namely,
the first two of the Ten Laws of Hockey Analytics.

The
first rule is that nothing is more important than wins. They are why you play
the game (outside of enjoying it of course), and they are the most necessary
ingredient in climbing the standings, making the playoffs, and succeeding in
the playoffs.

Wins, of
course, only come about because of goals. That’s why the second rule is that
goals are the most important factors in wins. Sure, everything else in the game
of hockey is a component of goals, be it shots, saves, passes, turnovers,
takeaways and so on. Even intangibles like compete level and leadership only
have value because somewhere done the line, they eventually lead to goals, and
subsequently wins.

With
that in mind, one of the most important things for a player is to be on the ice
for more goals for than goals against. This is extraordinarily simple logic.
The logic that makes it obvious that scoring more goals than an opponent means
winning. It’s the same logic that prompts old school hockey people to pour
faith into plus-minus. And it’s the same logic that keeps risky, high event
players like Jordan Subban in the minors until they can sufficiently dull their
game to the point where their goals against are clearly and consistently fewer
than their goals for.

Apply
this logic to Brandon Sutter, and you will be left wanting. Among 436 NHL
forwards with at least 100 minutes played at 5-on-5, Sutter’s 40% goals for
percentage ranks 357th. On the Canucks, he ranks 12th out of 14, with Jayson
Megna, Michael Chaput, and Jack Skille all netting better results – and each
for less than $1 million.

Even
outside of comparing him against other players, that abysmal ratio means that
for every two goals he is on the ice for, he’s on the ice for three against.
When you consider that this player averages the fourth most 5-on-5 time on ice
per game on his team, it’s no wonder that the Canucks struggle to keep their
heads above water. One of their most used players is an anchor dragging them
under.

Even if
you were to control for save percentages (which forwards have little to no
influence on), Sutter doesn’t show well by any method of shot shares.

Score and venue adjusted data via Corsica.hockey

While
some may still scoff at analysis based on these numbers (though that group is
growing ever smaller), the fact of the matter is that the statistics are highly
predictive of future success (and failure) – expected goals in particular.

Another measure with a high degree of predictability is DTMAboutHeart’s attempt at a single-number metric, WAR. Due to his series of injuries last season, you won’t find a WAR value for Sutter’s 2015-16 season. Nor will you find one for his 2014-15 season with the Penguins. You can however find a value for each of the six prior seasons, ranging from 2008-09 to 2013-14. The results are not appealing.

Data via DTMAboutHeart and Hockey-Graphs.com

If you aren’t familiar with the concept of WAR, it’s originally a baseball statistic that stands for wins above replacement. Attempts have often been made to convert it to hockey, and DTMAboutHeart’s version is the latest and most prevalent iteration. The basic idea is that each component is compared against a pre-determined replacement level (in rough theory, the best player outside of all 30 NHL lineups), and then converted in to Wins. The overall total represents the total wins above (or below) what a replacement level player with be worth under the exact same circumstances.

From left to right, the components are even strength offence, even strength defence, power play offence, drawing penalties, taking penalties, and faceoffs. As you can see from the final column, Brandon Sutter has been below replacement level for most of his career. And it’s not like he was trending upward with age.

It’s likely that he would have been a positive player in WAR in 2014-15, scoring 21 goals that year (much like in 2009-10, the only other year he was above replacement level. It’s also possible that he will be above replacement level this season, given his goal trends. That’s the nice thing about pumping out offence: it covers the warts. But it doesn’t eliminate them. Even in season above replacement level, he is only just so. The 2015-16 Canucks, as bad as they were, finishing 28th in the league, had 13 players with a higher WAR than Sutter accrued in 2009-10.

The long and short of it is that Brandon Sutter does not add much value beyond a replacement level player, and as such he shouldn’t be compensated much more than one. Of course, the fact that given the opportunity, Sutter can produce offence provides him with some intrinsic value. The opposing fact that he damages the team by consistently getting scored on, thus making it more difficult for them to win the more they play him should put a serious dent in his value. Though that doesn’t appear to be the way that current NHL executive’s think.

Actual player value is incredibly difficult to determine, because value is largely dictated by the rest of the market. And the rest of the market is consistently willing to shell out large sums of real money for bad reasons.

When the market assesses Sutter’s value, the first thing it looks at is his goal and point totals. Sutter has scored 20 goals in a season before (twice), and thus he is considered a 20-goal scorer even if he’s only done it twice and peaked at 21. The market does not care how he got to 20 goals, nor does it care if more goals are routinely scored in the other direction.

Last season, the average cap hit for forwards that scored between 19 and 23 goals (Sutter’s career high plus-or-minus two goals) was $3,866,563. Of course, that average includes salaries players like Joe Thornton, Nick Backstrom, and Ryan O’Reilly, players in the $6-8.5 million range that scored 82, 70, 67, and 60 points respectfully.

Meanwhile, Brandon Sutter has topped out at 40 points, and is on pace for 38 points this season, despite his inflated ice time. Reducing the list to players that scored at least 20 goals but no more than 40 points, the average salary falls to $3,270,833, more than a million dollars less than Sutter’s average salary. This list also includes players that are reaching these goal and point totals with second or third line minutes, rather than having their ice time in the top 30 of the NHL as Sutter’s is.

To compensate for that excessive ice time, we move to rate stats. Corsica’s Similarity Calculator allows us to find Sutter’s closest comparable players based on individual shot generation (iCF60), expected shooting percentage (ixFSh%), expected on-ice goal data relative to team, and actual production (deployment variables were removed from the formula, but it otherwise remained standard). I’ve added the cap hits of each of the top closest 20 matches (ELC players not included, since their salaries are capped at an outrageously affordable number).

The average salary of these players is $1,976,458, roughly $2.4 million less than Sutter’s salary.

If there is anything that makes Sutter’s salary reasonable, I have yet to find it.

After considering a great deal of evidence, I am left with the following conclusions regarding the original four points in Sutter’s favour:

None of these factors should lead to a player being paid $4.375 million. Sutter’s statistically assessed value places him closer to replacement level than core player. And finally, players that do what he does – even by raw numbers – can be found around the NHL doing so for much cheaper.

The final conclusion is this: Brandon Sutter is absolutely overpaid at his current wage. The goals, the faceoffs, the all situations play, and the gutting it out form a mirage that influences people into thinking that he is worth far more to his hockey team. They are in fact, to borrow a phrase from Blake Price, just lipstick on a pig.

Faceoff data from Puckbase.com. Salary data from CapFriendly.com. All other data from Corsica.hockey.