Solving the Jeff Skinner puzzle

Anybody that has been following Buffalo Sabres hockey knows that Jeff Skinner has been the talk of the town as of late. The $9 million man has had a rough start to his 2020/21 campaign, recording 0 goals and 1 assist through 14 games. This was followed up by a 3-game benching where head coach Ralph Krueger preached “accountability” and “earning your keep.”

One quick glance at Skinner’s stat line this season and you may be wondering what happened to the player that is two seasons removed from a 40-goal campaign. In all honesty, not much has changed in his game. He’s still drawing penalties, he still gets the puck to the dangerous areas on the ice, he still uses his shifty skating ability to carve through the defense. The only difference: the puck has not been going in the net.

So, let’s try to solve the Jeff Skinner puzzle or figure out if there is even a puzzle to solve.

Expected Goals vs. Actual Goals
At the end of the day, hockey is a very cruel game. It is a game that is oftentimes chaotic and very luck-driven on any given night. We have seen “puck luck” create hockey careers and we have seen “puck unluck” end hockey careers. A few bounces in one’s favor can create the illusion of a player being better than they actually are while a few bounces not in one’s favor can create the illusion of a player being worse than they actually are. Jeff Skinner is a player who has found himself in both categories at different points in his career, currently (unfortunately) he finds himself in the latter group.

One of the best resources we have to determine a player’s luck or unluck is expected goals. This statistic can be useful in determining the amount of goals a player should score based on their shot quantity and quality (without adjusting for shooting talent).

Here is a fantastic visualization from JFresh Hockey that shows Jeff Skinner’s actual goals per 60 at 5v5 plotted with his expected goals per 60 at 5v5 for every season since 2013/14:

jfresh Solving the Jeff Skinner puzzle

It is very interesting to see just how consistent Jeff Skinner is in his expected goal numbers at 5v5. His 5v5 xG/60 number is always hovering around that 1.0 mark which is a very solid rate. Most NHL seasons only have 60 to 80 forwards producing at a rate equal to or greater than that.

However, his actual goals per 60 rate at 5v5 is just a roller coaster. Some seasons, Jeff Skinner performs way above his expected goal rate (2016/17 and 2018/19). Other seasons, he plays well below his expected goal rate (2014/15 and 2020/21 so far). Then there are some seasons where the two rates are almost equal (2015/16, 2016/17, and 2019/20).

As you can see, some luck and outperformance of expected results can lead to the illusion that a player is much better than they actually are. Just look at Jeff Skinner’s 8-year, $72 million contract that he signed after his puck luck-filled 2018/19 season. Then there is the other side to the coin where some unluck and underperformance of expected results can lead to the illusion that a player is much worse than they actually are. The recent benching of Jeff Skinner is a perfect example of that illusion.

He is generating just as much danger at 5v5 as he has been throughout his career. In fact, his individual expected goal rate per hour is higher this season (1.06) than it was in his All-Star caliber 2018/19 season (0.95). 

The shifty 5v5 scorer that we have always known and love is still there. He is still the same type of threat in the offensive zone even while playing in Ralph Krueger’s system that tends to neuter any 5v5 offense. The only difference is the pucks are not going into the net this time around and the only real cure for that is time.

skinnerflyers Solving the Jeff Skinner puzzle

Limited Usage
Now that we have discussed the quality of Skinner’s play, let’s take a look at the quantity, particularly his ice time. While producing quality chances on the ice is fantastic, it is much more difficult to turn those quality chances into massive counting stat numbers (i.e. 40 goals) on limited ice time.

Jeff Skinner has only seen an average of 13:36 TOI per game this season which ranks 7th among Sabres forwards. That is a sharp drop off from his 18:31 ATOI in his 40-goal 2018/19 season or even his 16:29 ATOI in his first season under Ralph Krueger in 2019/20. He has been in the NHL since he has been 18 years-old and this is his first season seeing less than 16 minutes of ice time per game. Skinner is not only on the fourth line on paper this season, he is legitimately seeing fourth line deployment and usage. He is even seeing heavy defensive zone deployment in some games.

His lack of non-5v5 usage has been another feature of the Ralph Krueger era. He was booted off the Sabres top power play unit in 2019/20 after scoring 8 goals on that unit under Phil Housley in the previous season. He has not scored a power play goal since, his last one came in a game against the New York Rangers on February 15, 2019.

With Skinner getting all of his power play looks on a second unit that is significantly weaker than the top unit, the power play goals that helped him earn some of that $72 million contract are no longer there. It is much easier to hop on rebound and create chaos when Jack Eichel is out there:

skinnerwings Solving the Jeff Skinner puzzle

It is also worth mentioning that Skinner has gotten virtually no 3v3 OT ice time since Krueger has taken over from behind the bench. He has yet to see any OT ice time in 2020/21 and was 5th among Sabres forwards in OT ice time in 2019/20. This is pretty strange considering that he was their most clutch OT performer in 2018/19 (does the San Jose goal ring a bell?) and plays a game that is pretty well-suited for 3v3 hockey.

While a good portion of Jeff Skinner’s rough start to 2021 can be attributed to bad luck, some of it can also be attributed to a lack of opportunity. His role has been diminishing ever since Ralph Krueger took over the team last season. He has gone from 2nd line usage to 4th line usage to a press box seat and it’s not easy to score from up there.

The Chaos Theory
Jeff Skinner is a pretty unique player in the offensive zone because everything about his physical profile and attributes says he should be more of a perimeter player. That could not be further from the truth as he is one of the best in the league at scoring goals in close and in many different ways.

Naturally, when a player is considered a goal-scorer, such as Jeff Skinner, a coach’s first thought is to pair that player with a passer. This first thought tends to be the correct thought in many cases, just take a look at Ovechkin/Backstrom in Washington or Pastrnak/Marchand in Boston. Skinner is different from your typical goal scorer as he is not a player who relies on a high-end shot, he is a player who relies on chaos.

On MoneyPuck, they show what % of a player’s expected goals come from rebound shots and I noticed a bit of a theme with Jeff Skinner’s “good” years compared to his “bad” years. In his 40-goal season in Buffalo, 31.7% of his all-strength expected goals came off of rebound shots. That placed him 13th among all NHL players in that category.

In 2016/17, a season in Carolina that saw Skinner score 37 goals, he had 21.9% of his expected goals come off of rebounds. This number is less extreme (only slightly above average) but still placed him 4th on the Hurricanes in this category. Both his 2016/17 (37-goal season) and 2018/19 (40-goal season) saw him perform well above his expected goal numbers.

Now let’s take a glance at Skinner’s “down” seasons, where he performed around or below his expected goal numbers. In his final season in Carolina in 2017/18, he saw only 15.8% of his expected goals come off of rebounds. His 2019/20 season in Buffalo saw this number dip to 15.4% and it is currently riding around 16.1% for him in 2020/21.

From this data, there appears to be a bit of a theme with Jeff Skinner’s big seasons where he outperforms his expected numbers and that theme is a lot of rebounds and a lot of chaos. How do you generate rebounds? By shooting the puck.

Let’s take a look at some tracking data by Corey Sznajder that shows 5v5 Shots/60 and Primary Shot Assists/60 to help weed out the passers from the shooters. We will start of with the 2018/19 Sabres:

Screen Shot 2021 02 27 at 9.13.00 AM Solving the Jeff Skinner puzzle

During this season, Skinner’s most common 5v5 linemates were Jack Eichel, Sam Reinhart, and Jason Pominville. As you can see, Eichel and Pominville were both above average in terms of Shots/60 while Reinhart was slightly below average. This is a season where Jeff Skinner thrived on rebounds and it is easy to see why with those linemates.

Now let’s look at 2019/20:

Screen Shot 2021 02 27 at 9.16.34 AM Solving the Jeff Skinner puzzle

Skinner’s most common linemate was Marcus Johansson who clearly tracks as more of a passer than a shooter in this data. His second most common linemate was Conor Sheary, who shows up very well in this data, but he struggled to produce many rebounds relative to his shot production. The Skinner/Sheary duo performed well when Johan Larsson was their center (helped create chaos through his style of play rather than his shooting) but that line was short lived. Skinner still managed to lead the Sabres in 5v5 goals per 60 in 2019/20 despite many considering that season to be a “down” year for him.

To sum things up, Jeff Skinner is a player who scores no two goals that are alike. Unlike some other elite goal scorers in the game, he does not rely on a repeatable shooting motion that can consistently beat goalies. He relies on capitalizing on chaos and using his edgework to cut to the dirty areas. Maybe this style of play is what causes Skinner to be more prone to streakiness in production despite his expected goal rates remaining constant. Maybe this style of play is better suited in an offensive system that generates a lot of shots, therefore leading to a lot of rebounds (aka not Ralph Krueger’s Sabres).

Moving Forward
This was a very concerning quote I found from Jeff Skinner regarding the Ralph Krueger situation:

There are clearly some concerns Krueger has regarding Skinner’s play, but this quote makes it seem like it may be nothing more than some petty thing started by the coach. The whole situation appears to be hopeless for Jeff Skinner because he is in the coach’s dog house, despite playing the same game he has throughout his career, and is offered no advice on how to better fit the often cited “principles.”

Casting this whole coach situation aside, I think it is about time we see the reunion of the Skinner-Eichel-Reinhart line that thrived in the 2018/19 season. The biggest gripe about that line was that it hoarded all of the forward talent on one line, but the emergence of a solid 2nd line with Hall-Staal-Cozens and a nice 3rd line with Mittelstadt-Lazar-Sheahan has made the reunion of that line a more realistic option. Now is the time to re-ignite the Jeff Skinner that you paid for and you have the luxury in forward depth to do so.

No, the answer is not trading Skinner for another bad contract. He can still be useful in the proper role unlike some of those other bad contracts. No, the answer is not benching him for “accountability,” a term that does not appear to apply to Kyle Okposo, Tage Thompson, and Cody Eakin. 

The best way to get Jeff Skinner out of this slump is to let him play out of the slump. You can aid him in working out of it by giving him a bigger role with better linemates and usage. At the end of the day, hockey will always be a cruel, chaotic game that can be heavily influenced by luck and the sooner the Sabres’ coaching staff can come to grips with that reality, the sooner they will be able to properly manage a player like Skinner. The puzzle is not complex to solve.

  1. Dump the player plain and simple …..whatever the cost. If he can’t score a goal unless he’s tied to Jack and Sam …..then F-him.


  2. Now do the same analysis for Loui Eriksson and Milan Lucic and see how much they differ from Skinner.

    Skinner is overpaid and a big part of the Sabre’s struggles.


Leave a Reply

%d bloggers like this: