In extending Alexander Steen for three years, did the St. Louis Blues shrewdly lock down an elite sniper, or get themselves stuck with a $5.8 million 25-goal winger? Though currently second to Alexander Ovechkin in league goal scoring, Steen's previous eight-season best is just 24 goals. Will the Blues get value out of this deal?
Answering that question is a lot tougher than it appears. Not only does it require projecting his future performance, but it also requires assigning a dollar value to that prediction. That feat would be difficult enough for one of the league's most consistent players, let alone for a player whose scoring spiked completely out of the blue (no pun intended).
Steen's scoring will obviously regress over the next three years, but not necessarily all the way down to 25 goals per season. Based on similar players throughout history, and even his own past, we can estimate just how much we can expect his scoring to slide.
Even after projecting his scoring, assigning a price tag is a controversial matter, even in the emotionless world of analytics. Even if we set aside bigger picture concerns like grit, chemistry, leadership and all other intangibles, and confine ourselves to only his recordable on-ice contributions, there's still far too much that goes into a player's performance than can be summarized in a single number.
That being said, we do know enough about measuring value to come up with a rough estimate of what Steen will be worth, and it's surprising just how close that comes to the deal they agreed to. But first let's take a step back and project his scoring.
Gazing Into the Crystal Ball
The first step in figuring out if Steen can live up to his extension is projecting how good he will be over the next three seasons. This is no easy task even for a consistent player, and could even prove impossible for someone like Steen. Is he just riding a hot streak, or have his true talents only surfaced recently?
While there's a strong sense among hockey fans that whatever is happening today will last forever, history does tend to repeat itself. Sudden and dramatic increases in scoring don't usually level out; they tend to drop back closer to previous levels. Even a look at Steen's own past demonstrates this fact.
The graph that follows is the rolling 34-game average through Steen's entire career, which flows from early in his career on the left to the current day on the right. Goals are in blue and assists are in red.
It really isn't a question of whether or not Steen's goal scoring will regress, but by how much.
Steen had two previous goal-scoring spikes, the most significant occurring near the end of the 2009-10 season, his first full year in St. Louis. Specifically there was one 34-game sprint where he scored 18 goals.
While that's not quite the same pace as today, it was an equally dramatic increase from his recent past. It was also right before his last contract extension, so give Steen credit for his excellent timing.
After the 2010 spike, his goal scoring didn't remain at that exact same level, but it didn't slide all the way back either. His scoring only regressed about half of the way back to previous levels.
Will Steen's Scoring Regress?
Just because Steen's goal scoring slid halfway down last time doesn't mean it will happen again. It could regress all the way back down, not at all or anywhere in between.
To determine if a player is going to regress or not, context is everything. There are several underlying causes that can lead to dramatically increased scoring. A partial list would include the following:
- New teams and/or linemates,
- Increases in playing time (especially on the power play),
- More advantageous playing conditions, in opposing zones and/or against weaker opponents,
- Improved play and, of course,
- Blind luck.
To predict his future scoring, we need to know which of these applies in Steen's case.
According to the table below, his playing conditions haven't significantly changed recently. He's still playing with David Backes and T.J. Oshie, and for years he has been ranked in the top three among the team's forwards in average ice time. As for his playing conditions, they've actually gotten more difficult.
|ES TOI||12:26 (8th)||15:28 (2nd)||14:55 (3rd)||14:48 (3rd)||15:28 (1st)|
|SH TOI||1:06 (5th)||0:57 (6th)||1:42 (4th)||1:37 (5th)||2:05 (1st)|
|PP TOI||2:43 (3rd)||3:07 (3rd)||2:30 (3rd)||2:33 (1st)||3:11 (1st)|
|Off. Zone||45.7% (11th)||55.5% (5th)||52.6% (8th)||52.8% (5th)||46.8% (10th)|
|Qual. Comp.||0.97 (2nd)||0.43 (7th)||0.64 (5th)||1.56 (2nd)||1.86 (3rd)|
NHL.com and Behind the Net
Steen's scoring can't be explained by better linemates, more ice time or an easier assignment. It has most likely been a combination of a legitimate increase in offensive ability and a hot streak caused by random chance. But in what proportion?
How Legitimate is Steen's Scoring Increase?
One way to tell if a player has experienced a legitimate increase in his offensive production is to look at the number of shots he's taking per minute and the number of shots he's setting up per minute (for playmaking).
It makes sense that a player whose shots and setup passes increase at the same rate as his goals and assists is more likely to keep scoring at the higher rate. There's also ample historical evidence that this is true.
In Steen's case, however, the increase in goal scoring was not accompanied by an increase in shots. The following offensive profile chart, which maps a player's shots per 60 minutes against his (estimated) setup passes per 60 minutes, shows that Steen is taking roughly the same number of shots that he always has. The only real difference is in his playmaking, which isn't on trial here.
Steen's higher rate of goal scoring isn't based on an increase in the number of shots he's taking, but rather on the percentage of them that go in. His shooting percentage is currently 20.7 percent, over double his previous career average of 9.1 percent.
While this is a relatively rare scenario, there are players who once skated in these same tracks, and seeing what ultimately happened to them can provide insight into what will happen to Steen.
Among those who take a comparable number of shots, only three forwards have experienced such a dramatic increase that far into their careers. According to Hockey Reference, Rod Gilbert went from 9.9 percent to 18.1 percent at age 30, Vic Hadfield went from 9.2 percent to 20.7 percent at age 31 and Jacques Richard went from 9.5 percent to 19.9 percent at age 28.
How did these three players do over the next three seasons? Their combined average shooting percentage over the next three seasons was 14.5 percent. That's a regression of about half the way back down to their previous career average, which was exactly what we saw last time Steen's scoring spiked.
Take a look at their career scoring using the Hockey Reference links provided with their names. After their big years, Gilbert's goal scoring went from high 20s to three straight 36-goal seasons, Hadfield went from a 20-25 goal scorer to a 30-goal man and even Richard's scoring increased modestly.
What does all this mean for Steen? If history can be relied upon, both his own and those who are similar, then Steen could become a 35-goal scorer for the duration of his new deal. But is that worth $5.8 million?
What's a Goal Worth?
Having projected Steen's scoring level over the next three seasons, it remains to assign a dollar value to those goals. This can be a highly contentious thing to attempt, even within the analytics community. Even when restricting an analysis to what can be counted on a scoresheet, there are still too many factors to a player's game to summarize in a single number.
Accepting that the question can be answered definitively nor precisely, we can proceed to craft a rough estimate. The starting point is hockey's 3-1-1 rule, which states that every three goals gets you one point in the standings and costs one million dollars.
Of course, that doesn't mean that a 35-goal scorer is worth $11.7 million. A team isn't paying for those 35 goals, it's paying for the portion of those goals that's over and above what an AHL-level replacement would score instead. Especially one who played the same minutes and on the top line alongside Backes and Oshie.
The fact that Steen does more than score goals also needs to be accounted for. He also sets them up, and he prevents opponents from scoring with sound defensive play at both even strength and on the penalty kill.
That's where a statistic like goals versus threshold (GVT) comes in handy. It's a high-level estimate of all of a player's contributions in terms of the number of goals he has helped score or prevent, relative to what would have been achieved by a replacement-level player.
Prior to this season, Steen helped score or prevent around 12 goals per 82-game season more than a replacement-level player, according to Hockey Prospectus. That means he was previously worth about $4.5 million. Given that his cap hit is just under $3.4 million, he was due for a raise no matter what.
As for this year, Steen's current GVT is already around 12.0, highest in the NHL among skaters (by a long shot). Even if he were to regress to that same extent as we suspect, the remaining increase would easily still be worth an extra $1.3 million.
Now that's just a look at what a player like Steen is worth, not what he actually costs in the open market. Certain types of players are undervalued, but others are overpriced, making it better to let them go and to use the cap space differently.
What would it cost to replace Steen? Assuming Steen truly becomes a 35-goal man who continues to play against top competition and kill penalties, then there are only five other players who have matched that description at least once over the past few seasons.
Amazingly the average combined cap hit of Jeff Carter, Milan Michalek, Loui Eriksson, Patrick Marleau and Eric Staal is almost exactly $5.8 million. It really is quite stunning to arrive at the exact same conclusion as Blues management.
Steen will not contend for the Maurice Richard trophy, and everyone knows it. That being said, there are strong indications that he won't fall all the way back to his previous scoring levels. He will most likely become a 35-goal scorer who plays sound defensive hockey against top opponents, kills penalties and has an amazing upside.
Not only is that type of player worth $5.8 million, but that's also what players of that description cost in the open market. Obviously St. Louis is taking somewhat of a risk, but that's something NHL teams need to do in order to succeed. And if the Blues are going to make this kind of investment, then it might as well be on Steen.
All advanced statistics are via writer's own original research unless otherwise noted.