clock menu more-arrow no yes mobile

Filed under:

Reading Between The Lines: A Look at Barry Trotz's Forward Combinations

Evaluating how the Capitals have handled their forward combos through the season's first 20 games

Rob Carr/Getty Images

The Washington Capitals are a better team this year than they were last year, full stop.

Through 20 games last season the Capitals controlled 48.9% of even-strength shot attempts after adjusting for score effects; this year that percentage is all the way up to 53.1%. But despite the dramatic improvement in puck possession, the Capitals have two fewer points through 20 games (21) this season than they did last. Nevertheless, score adjusted puck possession metrics are currently considered the best predictor of a team's future point percentage, so there's reason for some optimism that the points will come to the Caps as long as the team continues to control the game.

The overwhelming majority of the team's improvement can be attributed to the change(s) they made behind the bench, of course, but that doesn't make the coaches' choices immune from criticism. For the first month of the season the team was unable or unwilling to settle on line combinations, electing to shuffle the deck every couple of games. It's logical for a new coach to want to give different lines and players a chance to show their merit, see if there's instant chemistry and so on, and they gave plenty of lines a chance - 21 different line combinations have played at least ten minutes together this season at five-a-side.

But here we are at Thanksgiving and, at long last, line combinations seem to be settling in. Unfortunately they're not necessarily optimal or even good ones. Rather than optimize their lines in order to play better puck possession hockey, as Ken Hitchcock has done in St. Louis, the Capitals have elected to create their forward lines using more... traditional methods. Here is a look at the numbers behind Barry Trotz's most recent (pre-injury) line combinations:

Forward 1 Forward 2 Forward 3 Time On Ice (Minutes) Corsi For % Corsi For % Rel. Zone Start % Rel.
Ovechkin Backstrom Wilson* 97.2 57.8% 6.6% 10.1%
Chimera Laich Ward 35.3 42.1% -10.1% -8.6%
Johansson Burakovsky Brouwer 133.3 55% 3.5% 21.8%
O'Brien Latta Beagle 17 56.5% 4.7% -32.0%

Three of those lines have possession numbers north of 50%. Pretty good right? The line of Alex Ovechkin, Nicklas Backstrom, and Tom Wilson looks particularly successful, but dig a little deeper and that optimism needs to be tempered a bit - about 30% of Wilson's time with the big guns has been spent in the last two games, against Buffalo and Colorado, which are currently the worst- and third-worst puck possession teams in the League, respectively. It's not often that two games (even this early in the season) are so lopsided that they heavily influence a player's numbers, but heavily influence Wilson's numbers those two games did. Here are the Vollman usage charts (generated by War-On-Ice) both before and after the games in question:


Wilson was a negative (relative) puck possession player prior to the two games in question. If you remove those two games from the equation you get a better look at how those three (but more specifically Ovechkin and Wilson) have played when together:

Tom Wilson hasn't been as good on the top line as his underlying numbers would lead you to believe, and the box cars aren't there either, as the trio has only been on the ice for three five-on-five goals-for. Even if you include the Buffalo and Colorado games, Ovechkin and Backstrom have had more trouble scoring with Wilson than they had with either Troy Brouwer or Eric Fehr. Despite the lack of production, however, Trotz has committed to that line (at least in the short term).

The goal for your scoring lines is, well, to score, and having Wilson on the top line has hampered the goal output from Ovechkin and Backstrom. The Capitals offensive woes have been compounded by a second scoring line of Brouwer, Marcus Johansson, and Andre Burakovsky that hasn't registered a single five-on-five goal since the team's win over Columbus over two weeks ago (despite very favorable deployments). How long can these trios remain together without producing?

So... How About Eric Fehr on the top line?

Given the strong possession numbers of the second scoring line above, we'll leave it alone for the moment and look to the other "good 1RW option." If the Capitals believe that puck possession wins hockey games (and championships) Eric Fehr should, not only be in the lineup, but be getting top-six minutes. Prior to Ovechkin's injury, Fehr was slated to watch Wednesday's match-up with the Islanders (one of the best puck possession team's in hockey) from the press box, a scratch that would have most onlookers scratching their heads. Fehr has been one of the Capitals' best puck possession forwards over the last few seasons, and that trend has continued into this year.

Forward 1 Forward 2 Forward 3 Time On Ice (Minutes) Corsi-For % Corsi-For % Rel. Zone Start % Rel.
Chimera Fehr Ward 80.6 47.6% -4.6% -22.2%
Ovechkin Backstrom Fehr 70.2 61.9% 10.8% -9.4%

Note: Fehr is one of only three Capitals' players to play at least 50 minutes with two different lines (the other two players are Backstrom and Ovechkin).

While together this season, the line of Fehr, Jason Chimera, and Joel Ward has not been putting up astounding puck possession numbers... but those numbers are good enough given their heavy dosage of defensive deployments (and far and away the best numbers Chimera and Ward have with any center with whom they've skated more than a handful of shifts). In fact, those numbers are significantly better than those of the Brooks Laich centered version of the 'green line' (currently at a CF% of 42.1% and a Relative CF% of -10.1%, in more favorable zone starts).

We advocated for Eric Fehr to be the Capitals top line right wing prior to the start of the campaign, and the possession numbers for that line were impressive during their short time together. Their success together is not limited to just this year, though - since 2007, Fehr and Ovechkin have played over 400 minutes together and the results have been exceptional:

Fehr and Ovi

The goals-for percentages fall in line similarly.

Prior to this season Ovechkin has a higher CF% with Eric Fehr than he has with any other forward with whom he has played at least 150 minutes. But despite strong play while together (both historically and in the present) Eric Fehr seems more likely to be in the press box than to see more playing time alongside Ovechkin. And that's unfortunate.

Like Mike

Trotz appears to have a blueprint for his two scoring lines... and it's a blueprint that the Capitals have seen some success with in the past. During the peak years of the Ovechkin Era, Mike Knuble was a regular on the top line. Knuble used his large frame to maneuver himself into prime scoring real estate, and his hard work often paid off. Here, via SportingCharts, is a look at the heat maps for Knuble and the Caps' current trio of first-line right wing contenders over the course of their respective careers (for which we have data):

Heat Maps

Going to the net is a time-tested method of scoring goals (and something the Caps believe they need to be doing more of), so it's easy to see why Barry Trotz wants each of his two scoring lines to consist of two skilled players (Ovechkin/Backstrom - Johansson/Burakovsky) and one "net presence" winger (Brouwer - Wilson, presently). If this is indeed the case, perhaps it adds some context to Fehr's exclusion from the top-six. Fehr has the physical gifts to be the heavy-hitting-puck-retriever that Barry Trotz wants (perhaps that is why he started the year on the top-line), but has never shown a particular aptitude for body checking. Fehr is very adept at obtaining the puck, but he does it by utilizing his reach. The result is the same, and there's nothing in the heat map above that would seem to indicate a particular aversion to high-traffic areas of the ice. If the belief of the coaching staff is that each of the scoring lines needs a net-front presence in order to score... where are the goals?

There is value in establishing an organizational identity, and perhaps Fehr's perceived inability (or reluctance) to modify his playing style has resulted in him being relegated to the press box. But without more information, it's difficult to see why he's been unable to stick in the lineup, much less on the top line.

Right Method, Wrong Results

While Fehr has put up solid puck possession numbers and results while apparently playing the wrong way, Jason Chimera has been one of the Capitals' worst puck possession forwards, despite presumably playing the right way. Chimera seems like a good leader and a hard worker, but there is strong evidence that he has become less effective over the last few seasons, whether as a result of father time catching up to him or systems that are less conducive to his style or whatever. He can still contribute to the team on a nightly basis, but the team would best be served utilizing him less than they are currently. Ditto Jay Beagle.

Forecasting the Forwards

Pulling it all together a bit, here's a look at the how the Capitals' forwards have fared when playing with one another (minimum 30 minutes of five-on-five ice time, ordered by per-game even-strength ice time):

Couples' skate

Puck possession is not the goal of any hockey team, winning is the goal. It just happens that the best and most predictable way to win is by possessing the puck. Based on external assessment it appears that the Capitals are not acting upon possession based metrics to either form or assess their forward lines. When used properly analytics can be a powerful tool in the decision making process, but the numbers don't encompass everything. Through the data it's possible to objectively view what has or hasn't worked in the past, but things are always in flux. That said, the line combinations below have the potential to produce strong puck possession numbers, but a large part of that potential is based around visual assessments of the skill-sets of the player's that make up these proposed lines.

Line Left Wing Center Right Wing
1 Ovechkin Backstrom Brouwer
2 Johansson Burakovsky Fehr
3 Kuznetsov Laich Wilson
4 Chimera Latta Ward

First Line - After all that advocating for Fehr, we're going to throw Brouwer on the top line? Yup. Brouwer was the second player (after Fehr) to get a chance to play with Ovechkin and Backstrom and stuck around for just over 40 minutes of five-on-five play. The line came out of the gate well (Ovechkin scored a goal against the Devils during their first shift together) and saw over 58% of the shot attempts directed towards their opposition's net. Troy Brouwer is likely not a huge upgrade over Tom Wilson, but moving Brouwer should improve the puck possession numbers for Ovechkin and Backstrom while freeing up a spot on the second line.

Second Line - Johansson and Burakovsky have played together for the majority of the season, and for good reason. While together the pairing is putting up very strong puck possession numbers. Part of their success can be traced back to the relatively light assignments that the two have been facing, but even their zone-start-adjusted numbers are solid. Adding Fehr (who has been relied upon to face tougher competition) to the line would increase the versatility of the second line and allow for a more even distribution of offensive zone face-offs throughout the four forward lines. Granted, this line would still likely see a heavy dose of offensive-zone starts.

Third Line - Is it time to move Evgeny Kuznetsov to the wing and put him with some quality linemates? Maybe (and if not, swap him and Laich on this line) and yes, respectively. His puck possession numbers have been improving, but his game is obviously not at the level where Trotz wants it to be. Wilson and Laich can provide Kuznetsov with exactly what he needs to be successful - time and space to utilize his skill, and some defensive conscience. Don't forget that we have seen Kuznetsov and Wilson put together some solid transitional plays together in the past...


Fourth Line - As mentioned above, Chimera has been given a lot of slack by the coaching staff and a large part of that can be attributed to how hard he works... but it's time to reduce his ice time. Michael Latta and Ward are both decent possession players and together this line could potentially match the second scoring line of most opposing teams.

It is nigh impossible to forecast the future of lines that have not played together with much (or any) frequency, but you can identify which lines have struggled when together by examining the numbers. Once those areas are identified you can isolate and correct the cause of the struggle through the use of video and other methods. And all of this isn't to say that the line combos are the biggest issue with the Caps right now or that "fixing" them is a silver bullet - the Caps aren't using awful line combinations (it's not like Jay Beagle is playing with Ovechkin and Backstrom). But with just a few tweaks the team could likely produce better results.