Dragging Since December: The Analytics Behind the Washington Capitals’ Prolonged Rut

December 8, 2019. The Washington Capitals are 22-4-5, good for 49 standings points, and sit three points clear of the Boston Bruins as the best team in the NHL. The Capitals have a nine-point lead in the Metropolitan Division. Many pundits and analysts across the hockey-sphere are praising the Capitals and their dominant start to the season, crowning them bona fide Stanley Cup contenders. Defenseman John Carlson has cemented himself as a Norris Trophy favorite, racking up 43 points in 31 games played. Captain Alex Ovechkin racks up another 20-plus goal season. The Caps have three players operating at a point-per-game or higher pace. In summary, the team looks as strong as they appeared to be on paper.

Now, sitting in early March, reality has set in.  As of the time of writing, the Capitals are now tied atop the Metropolitan Division with the Philadelphia Flyers, after forward Mika Zibanejad torched the team for five goals, including the overtime game-winner on March 5 in a tilt against the New York Rangers. The game in the Big Apple marked the second time in the past month that the Caps have either lost, or tied, for the division lead. The Capitals have been slumping for just about three months, and the question is starting to boil: is this the real iteration of the 2019-20 Capitals? Was the torrid start in the first 31 games just a flash in the pan? In this piece, NoVa Caps takes a look at the analytics over the course of the prolonged slump.

Possession Metrics

Taking a look at the differences between the two portions of the season in which one sees the real divergence in the only statistic that really matters at the end of the day: wins and losses:

possession and xgf stats

Now, without context, one might look at the Corsi For and the Fenwick For during the more recent months of play, and say that’s an improvement. The reality of possession statistics are that teams that are playing from behind more often, tend to have higher possession numbers since they’re pushing to tie the game up or reduce the deficit they’re facing. To establish that context, one has to see how the two versions of the Capitals stacked up in different situations at five-on-five play:

possession while leadingpossession while tiedpossession metrics while trailing

To add some context, take a look at the time on ice at five-on-five play by date range for each of these scenarios:

  • From 10/3 to 12/8 (31 games played):
    • Leading: 646:52 (best in the NHL)
    • Tied: 493:32 (20th in the NHL)
    • Trailing: 321:26 (second-best in the NHL)
  • From 12/9 to present (36 games played):
    • Leading: 444:21 (28th in the NHL)
    • Tied: 502:01 (30th in the NHL)
    • Trailing: 691:49 (27th in the NHL)

Realistically, one would be looking at a relatively similar amount of games played in these two date ranges. The differences are stark. The Capitals had just over 200 minutes more in five-on-five ice time while playing with the lead in the first 31 games versus the last 36. They’ve been trailing in games by over 370 minutes since 12/9 compared to the first 31 games played.

The difficulty here is, for instance, the Caps are more likely to give up a goal than score one with the lead. That means that teams are more likely to tie a game or close a deficit than the Capitals are to extend their lead. Additionally, their Expected Goals For dip below their Expected Goals Against while tied, meaning that they’re more likely to start trailing in a game than they are in taking the lead.

The biggest story here is the difference in Expected Goals For and Expected Goals Against holistically. To add insult to injury, since 2/1, the Caps are actually negative in this category. They have an Expected Goals For at five-on-five play at 13.59 (sixth in the NHL), but have an Expected Goals Against at 13.85 (third-worst in the NHL). The five best teams in Expected Goals Against are clear contenders in the playoff race: Vegas, St. Louis, Philadelphia, Tampa Bay, and Boston.

The picture here is clear: teams that have low Expected Goals Against figures fair much better than teams that do not. The Capitals do not, therefore, they’re not really playing like a playoff team at the moment. This is confirmed by the eye test as well. The Caps are having considerable issues defensively that can’t entirely be pinned on the team’s defensemen. There’s been a lot of lazy back-checking (see Ivan Provorov’s goal against the Capitals when they faced off with Philadelphia on Wednesday as an example). Systematically, things might be fine, but there’s a definite effort concern that has a simple consequence: pucks in the back of the Capitals’ net.

Scoring Chance Metrics

Now, one must dive into scoring chance metrics and compare the two date ranges:

scoring chance generation

The quick takeaway here is that while the Capitals didn’t own quite the percentage of scoring chances for earlier this season, the finish was much better. That would be the easiest assumption to make here, since the High Danger chances didn’t go completely the other way in either date range. This seems to reign true, since the Caps weren’t necessarily defensively stout earlier this season, but had a higher percentage of finishing a scoring chance with a goal on the board.

The most notable thing to mention too is, teams are more willing to take risks to generate a high danger scoring chance when they’re trailing in a game, which likely accounts for the rise in High Danger Chances For and High Danger Goals For percentages, since the team has been trailing in games at the highest proportional rate.

Conclusions

Ultimately, there are a lot of gaps in the Capitals’ play since 12/9. The underlying metrics, without added context, make it look as though the Caps have improved in some areas, with the truth being that the team has bloated possession and scoring chance generation metrics because they are more likely to be trailing in a game than they are tied or leading in a game.

There’s a lot of on-ice reasons for this, including the incredibly high amount of penalties this team has taken. It’s a solid point, but it’s important to add that penalties were an issue in the first 31 games of this season too (188:55 in time on ice for the penalty kill). The likely explanation for the Caps’ woes is probably simpler than most would expect. It’s not solely on the on-ice systems, the personnel, or the coaching staff. Watch most of the Capitals’ Goals Against, and one will see see a trend: forwards losing containment on their assignments in the defensive zone. Once players have defenders shifting their positioning to compensate for the missed assignment, they have mismatches and openings that are difficult to account for. This is a group responsibility, and it’s a group blame. The players on the ice, for the most part, haven’t performed at the level many are accustomed to seeing them perform at. There’s obviously exceptions to this, but at the end of the day, it’s a 200-foot game, and there’s far too many players on the team that are too focused on the offensive half.

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 at a non-profit in Jacksonville, FL. Justin enjoys geeking out over roster construction and cap management.
This entry was posted in Data and Analytics, News, NHL, Players, Teams, Washington Capitals and tagged , , , , , . Bookmark the permalink.

Leave a Reply