Regression Or Progression: Whose Scoring Output Will Increase Or Decrease?

Photo: NHL Via Getty Images

Thus far into the 2019-20 season, the Washington Capitals are the highest scoring team in the NHL. While this is a great place to be for the first place team in the NHL, is this something we can expect to continue going forward? Or will the Capitals level-off and need to rely on their defense more? We’ll take a look at the Capitals’ team and individual statistics to see how the team will fare going forward, as well as what players we can expect to regress back to their career averages, or who will perform better as the season goes on. 

Team Statistics:

As mentioned earlier on, the Capitals are the highest scoring team in the NHL with 81 ‘goals for’ in all strengths this season. Let’s take a look at some key performance indicators, and see how the Caps are faring compared to this point last season (statistics thanks to Natural Stat Trick):team scoring stats

Overall, looking at the expected ‘goals for’ and ‘high danger chances for’, you can see that the Caps’ offense is generating higher quality scoring chances, resulting in higher expectations for goals.

The Caps’ increase in expected ‘goals for’ is a good sign, meaning they’re generating more scoring opportunities and ultimately, more ‘goals for’. The interesting thing here is that the Caps have generated three less scoring chances for, but have scored 12 more goals in scoring chance scenarios.

The Caps went from an 8.36% conversion rate of scoring chance opportunities in 2018-19 to a 10.37% conversion rate. As for high danger chances, the Caps have definitely had an impressive increase in overall chances and ‘goals for’. The Caps went from a high danger chance conversion rate of 13.48% to a 17.57% conversion rate.

An important metric to pay attention to is PDO. PDO measures even strength shooting percentage plus even strength save percentage. Teams that have really high PDO figures (over 100) typically are more “lucky” than good. According to Kent Wilson at the Calgary Herald, players and teams that are around a PDO of 102 are typically not as good as you think they are.

Think of the 2017-18 Colorado Avalanche and New York Islanders, who had a PDO of 103.59 and 103.28 respectively. Those teams didn’t make much noise that season: the Isles ended up with 80 points, 17 points out from a playoff spot, and the Avs ended up with 95 points, a wild card finish, and a first round defeat at the hands of the Nashville Predators in 6 games.

When we look at the Caps’ PDO numbers from last season and this season, they’re essentially solid numbers. They’re not over-performing and riding a PDO bender to the playoffs, where they’ll see an immediate drop in quality of play. Ultimately, PDO is a good way of determining if a team is “for real” or not. Teams closer to 100 and have a strong position in the standings are more likely than not good teams, and the Caps are first in the NHL right now, so you might say that they’re “for real”.

There’s a few things at play here. It’s clear the Capitals have tweaked their system to use their speed a bit more. The Capitals have nine players that are in the double digits for individual high danger chances for, led by captain Alex Ovechkin with 29. Impressively enough, the Capitals are generating high danger chances throughout the forward ranks. The best example of this is the fourth liner Brendan Leipsic clocking in with 15 high danger chances for. When your entire forward group can generate scoring chances, let alone high danger scoring chances, your team will score more. The difference between this year’s iteration of the Caps versus last year’s is that the forward group has more skill depth, rather than more defensively focused players. The less you have to rely on your top level talent, the more success you’ll have since defensive match-ups opposing teams will deploy won’t be nearly as suffocating.

Individual Statistics:


Now, let’s take a look at the individual performances, and make some judgments on who can be candidates for regression, and who could break out in the forward ranks (thanks to Hockey-Reference and Natural Stat Trick):

forward scoring stats

Progression Candidates:

Overall, there’s three players shooting at a lower percentage than their career rates: Backstrom, Hathaway, and Stephenson (by a hair). The biggest progression candidate here is likely Backstrom. He has a sub-10% shooting percentage so far this season for the first time since the 2013-14 season, where he had a 9.2% shooting percentage. Since then, he’s had shooting percentages of 11.8%, 15.5%, 14.2%, 12.7%, and 13% from 2014-15 to the 2018-19 seasons. Backstrom is currently trailing his career points per game pace as well. .722 points per game is certainly nothing to complain about, but there’s definitely more opportunity for Backstrom to get back to his career average there. This is especially the case since Backstrom’s last two seasons’ point per game pace was .925 and .876 respectively. 

Backstrom is also trailing his expected goals for rate by a goal and some change. This isn’t anything to fret about for sure, since good teams usually outperform their expected goals rate. For example, of the top ten scoring teams in the NHL currently, all ten currently have more goals for than expected goals for. This is mostly due to expected goals is typically calculated by “assigning goal expectancy to shots” (Corsica Hockey). The reason there’s such a large variance is that not all shots are good ones, especially ones that end up in the back of the net. For example, a goal that is scored from the neutral zone via a dump in on net won’t have a high value for goal expectancy. 

Overall, there are two players that are due a progression and will likely end up scoring more goals by the end of the season. These two players are Nicklas Backstrom and Brendan Leipsic. Backstrom is the top choice based off of the numbers above, but the dark horse candidate for an increase in scoring is Leipsic.

Leipsic passes the eye test. He’s probably one of the most skilled players that the Capitals’ have had on the fourth line in a long time. He gets time on the second power play unit as well, which should help him get more scoring opportunities when he gets a chance. Like mentioned above, he helps generate high danger chances for, and ultimately, shooting percentages rise when you have high danger chances offensively. Also, he’s currently behind his career point per game pace, which shows that there’s a bit more scoring upside to his game.

Regression Candidates:

Let’s just look at shooting percentages for a second here. Last season, through this point, the league average for team shooting percentage was 9.60%. This season, the league average is 9.67%. There’s only a slight increase in shooting percentage around the league, so seeing huge increases in individual shooting percentages shouldn’t be very common. 

Ultimately, among the Caps’ forward ranks, there’s seven players who are outpacing their career average in shooting percentage. There’s only a few players that are significantly outpacing their career averages in shooting percentage. These players are Dowd (low shot volume), Oshie, Kuznetsov, and Wilson.

For Dowd, it’s just sheer lack of shot volume due to limited ice time. Oshie, Kuznetsov, and Wilson are top six players who get opportunities on the power play. Kuznetsov, Oshie, and Wilson are three of the top five players who generate the most high danger ‘chances for’ on the team among the forward ranks. Those types of chances are always going to result in higher chances of scoring, so shooting percentages can be carried there as well. So long as the Caps keep generating high danger chances, the shooting percentages will be less important.

At the end of the day, the most likely “regression” candidate is Oshie. That’s not saying he isn’t an effective or good player at all; it’s more that he’s 12 years into his NHL career and is more of a known quantity in terms of production. He’s scoring at a .45 goals per game rate so far this season, compared to his .36 goals per game mark he set last season. Look for his numbers to get back to that pace, where Oshie scored 25 goals in 69 games last season. That’s still more than effective, and is just under a 30 goal pace for an entire 82 game season. Since Oshie has 10 goals already, it doesn’t seem like it’s a huge stretch for Oshie to get to the 30 goal mark for the second time in his career this season.


Now, let’s take a look at defensemen and see what their production looks like currently, and where we could possibly see progression or regression. These stats are thanks to Natural Stat Trick and Hockey-Reference:

defensemen scoring stats

Progression Candidates:

There’s two real progression candidates from the scoring perspective on the back-end based on the numbers above: Gudas and Jensen. They’re both scoring below their career points per game pace, and both have zero goals on the season. In fact, Jensen has yet to register a goal as a Capital in 42 regular season games. Based on Jensen’s career stats, there’s not much to expect from him goal-scoring wise, but should certainly be dishing out more assists based on his career rate. His lowest scoring total in a full season was 15 points, which has his first full season in Detroit back in the 2017-18 season. If he doesn’t up the scoring rate a bit, at least close to his career average, you’d have to start to wonder about Jensen’s fit in the Caps’ scheme.

Gudas has shown that he’s good for three goals on average a season, so expect him to pot a few this season. He’s certainly had his chances, and is tied for second in shots by a defenseman at 29 with Dmitry Orlov. Gudas is also coming off of a 20 point season in 77 games with Philadelphia last season, and is basically getting the same amount of average time on ice in Washington as his was last season with Philadelphia. We can likely see more assists from Gudas this season with his skill-set for outlet passes through the neutral zone. He’s connected on a few that have led to scoring chances, it’s just up to the forward at that point to make it count.

Regression Candidates:

Realistically, you’d have to put Michal Kempny down as a regression candidate. He’s come out of the gates blisteringly hot after returning from his injury, trailing only Ovechkin, Kuznetsov, and Carlson for points per game this season. Based on what we’ve seen from Kempny, and his career stats, this isn’t really the player that Kempny is. The question is, does getting opportunities with Norris Trophy front-runner John Carlson help him add to his point totals, and that’s what’s driving it? Maybe. Has Kempny turned into a prolific offensive defenseman? Probably not. It’s hard to say if we’ll see a quick regression, but we’ll likely see his scoring totals trail off. That being said, we’ve seen scoring totals spike for a player in a given season, but flames out after that, so is it possible that Kempny eclipses the 50 point mark? Sure, if he keeps up this pace. But it’s more likely than not that he’ll slow down a bit to closer to his career points per game average rather than being an extremely late blooming offensive threat at 29.

The other regression candidate that you might deduce from the chart above is John Carlson. He’s basically doubled his career points per game figure this season, which is extremely impressive. The reason we won’t put him down as a regression candidate is that Carlson’s career stats year over year have seen a real increase in scoring.

If you go back to before the 2017-18 season, Carlson’s career high in points was 55, and he set that mark in the 2014-15 season. Since the start of the 2017-18 season, Carlson has racked up 170 points in 184 games, good for a .923 points per game mark. Even if you want to exclude this year’s performance, the past two full seasons, Carlson scored 138 points in 162 games, good for a .851 points per game pace. So, realistically, Carlson is performing just a touch ahead of his scoring pace per game for the last three seasons.

Carlson’s game has really evolved to a point where he is in the conversation for being a top defenseman in the league. He’s getting a ton of ice time this season so far, averaging 25:25 in average time on ice. That’s good for fourth most in the NHL, behind only San Jose’s Brent Burns, Edmonton’s Oscar Klefbom, and Ottawa’s Thomas Chabot. That’s pretty good company there. 

By Justin Trudel

About Justin Trudel

Justin is a lifelong Caps fan, with some of his first memories of the sport watching the team in the USAir Arena and the 1998 Stanley Cup appearance. Now a resident of St. Augustine, FL, Justin watches the Caps from afar. Justin graduated with a Bachelor's of Science in Political Science from Towson University, and a Master's of Science in Applied Information Technology from Towson University. Justin is currently a product manager. Justin enjoys geeking out over advanced analytics, roster construction, and cap management.
This entry was posted in Data and Analytics, News, Washington Capitals and tagged , , , , , , , , , . Bookmark the permalink.

2 Responses to Regression Or Progression: Whose Scoring Output Will Increase Or Decrease?

  1. Anonymous says:

    Off to a great start both home and away so far this season! I wouldn’t be to concern about who’s up and or who’s down. Team hockey sixty minute end to end is exactly what the doctor ordered. The stats will generally average out when it’s all said and done.

  2. Pingback: Boyd Moves Up To Third Line; Hagelin, Dowd Still Out: Notes From Morning Skate | NoVa Caps

Leave a Reply

Your email address will not be published.