The place to start here is the basic strategy employed by the BSens: the umbrella. The BSens’ strategy with that deployment involves moving the puck to the top of the umbrella, distributing left or right for a shot, and then trying to take advantage of the rebound or loose puck. The down low stuff play is almost never utilized and the tendency has been trying to get a shot from the left side by a right-handed shooter (although both Chabot and Harpur, as lefties, tend to shoot from the middle of the umbrella into the crowd).
How is the coaching staff deploying the team’s roster in this situation versus their production? I crunched numbers that I’ve tracked throughout the season and the easiest way to start with is with literal deployment: who is playing the most with the man advantage, keeping in mind that smaller sample sizes tend to skew results (positive or negative):
Powerplay Frequency/Shifts per game (min 5 shifts on the season)
Forwards
DiDomenico 5.0 (10/50) (21/0/29)
White 4.09 (21/86) (0/50/36)
Rodewald 4.0 (20/80) (26/22/32)
Paul 3.88 (17/66) (30/18/18)
Chlapik 3.66 (27/99) (36/40/23)
Sexton 3.5 (8/28) (15/13/0)
McCormick 3.35 (28/94) (25/37/32)
O’Brien 2.28 (28/64) (5/28/31)
Reinhart 2.28 (28/64) (17/31/16)
Gagne 2.27 (29/66) (6/39/21)
Perron 2.17 (28/61) (10/32/19)
Blunden 2.0 (9/18) (13/0/5)
Werek 1.63 (22/36) (19/12/5)
Randell 0.48 (27/13) (2/0/11)
Defense
Chabot 5.15 (13/67) (31/36/0)
Jaros 3.89 (19/74) (36/19/19)
Murray 3.47 (23/80) (12/30/38)
Lajoie 3.47 (19/66) (34/12/20)
Harpur 2.42 (7/17) (4/4/9)
Burgdoerfer 1.76 (30/53) (28/24/1)
The above does not differentiate between first and second unit play (see below), which can mean quite large variances in actual TOI with the advantage. From the above I think it’s worth noting that if you remove October from O’Brien and Gagne’s averages they remain in the same relative position, but far closer to the other top forwards (3.1 and 3.0 respectively). Also significantly impacted by varying usage are Lajoie (who drops to 3.2 after October–not a huge change, but some) and Burgdoerfer (2.36 prior to December).
If the above represents preference (debatable, but let’s follow the thread), based on frequency and a healthy roster we’d see DiDomenico-White-Rodewald with Chabot-Jaros on the first unit and Paul-Chlapik-Sexton with Lajoie-Murray on the second. While this is not ideal (see below), it does remove the anchors that are McCormick and Reinhart, albeit Rodewald’s and Murray’s mindnumbing inability to produce on the PP (see below) are cause for concern.
Unit Use (first/second unit/pairing)
Forwards
DiDomenico 100% (50/0)
Sexton 100% (28/0)
White 98.8% (85/1 – 49/1 36/0)
O’Brien 75% (48/16 – 0/5 25/3 23/8)
Paul 59.1% (39/27 – 15/15 12/6 12/6)
McCormick 58.5% (55/39 – 12/13 37/0 6/26)
Reinhart 53.1% (34/30 – 3/14 21/10 10/6)
Perron 50.8% (31/30 – 7/3 8/24 16/3)
Werek 50% (18/18 – 13/6 0/12 5/0)
Rodewald 48.7% (39/41 – 15/11 6/16 18/14)
Chlapik 42.4% (42/57 – 24/12 8/32 10/13)
Blunden 27.7% (5/13 – 0/13 5/0)
Gagne 10.6% (7/59 – 0/6 7/32 0/21)
Randell 0.00% (0/13 – 0/2 0/0 0/11)
Chabot 100% (67/0)
Harpur 58.8% (10/7 – 4/0 4/0 2/7)
Jaros 39.1% (29/45 – 26/12 6/13 0/19)
Murray 31.2% (25/55 – 9/3 0/30 16/22)
Burgdoerfer 28.3% (15/38 – 12/16 3/21 0/1)
Lajoie 16.6% (11/55 – 0/34 3/9 8/12)
Something that’s immediately apparent is that defensemen are not used 100% of the time as Colin White, Chris DiDomenico, Jim O’Brien, Filip Chlapik, Franics Perron, and Max McCormick have all spent varying time on the point (White the most). It’s amazing that Gagne has almost exclusively been stuck on the second unit and that Chlapik’s first line appearances have been (relative to production) so limited. Keep these numbers in mind as we get to production below. As for the lines these preferences suggest, it would yield DiDomencio-White-Sexton with Chabot-Harpur as the first unit, with McCormick-Paul-O’Brien with Murray-Jaros as the second.
Powerplay on-ice for Goals
Forwards
Sexton 0.14 4/28 (3/1/0)
O’Brien 0.093 6/64 (1/3/2)
Werek 0.083 3/36 (2/0/1)
White 0.081 7/86 (0/3/4)
DiDomenico 0.08 (4/50) (2/0/2)
Rodewald 0.075 6/80 (2/2/2)
Paul 0.075 5/66 (3/1/1)
Chlapik 0.07 7/99 (3/3/1)
Perron 0.065 4/61 (1/1/2)
Blunden 0.055 1/18 (0/0/1)
Reinhart 0.046 3/64 (1/1/1)
Gagne 0.045 3/66 (0/2/1)
McCormick 0.042 4/94 (0/3/1)
Randell 0.00 0/13 (0/0/0)
Defense
Burgdoerfer 0.094 5/53 (3/2/0)
Lajoie 0.075 5/66 (3/2/0)
Chabot 0.074 5/67 (3/2/0)
Murray 0.062 5/80 (0/2/3)
Harpur 0.058 1/17 (0/0/1)
Jaros 0.054 4/74 (3/0/1)
How much of an indicator is being on-ice for goals? Given how meaningless the plus/minus stat is you want to take it with a grain of salt. Sexton, Werek, Harpur, and Blunden are benefiting somewhat from their limited sample size.
If we create a lineup based on these numbers we’d wind up with slightly different lines compared to raw usage: Werek-O’Brien-Sexton with Lajoie-Burgdoerfer as the first, and DiDomenico-White-Rodewald with Chabot-Murray as the second. There are positive elements here as well, but as we’ll see below, they present problems as well.
Powerplay Points Per Shift
Forwards
DiDomenico 0.08 (4/50)
Chlapik 0.06 (6/99)
Werek 0.055 (2/36)
Blunden 0.055 (1/18)
Perron 0.049 (3/61)
O’Brien 0.046 (3/64)
Gagne 0.045 (3/66)
White 0.034 (3/86)
McCormick 0.032 (3/94)
Reinhart 0.031 (2/64)
Paul 0.03 (2/66)
Rodewald 0.00 (0/80)
Sexton 0.00 (0/28)
Randell 0.00 (0/13)
Defense
Chabot 0.074 (5/67)
Lajoie 0.06 (4/66)
Burgdoerfer 0.056 (3/53)
Jaros 0.04 (3/74)
Murray 0.012 (1/80)
Harpur 0.00 (0/17)
The most meaningful element of all, particularly for a team that struggles to score–actual production (and we can ignore Blunden’s position due to the small sample size). There are some extraordinary numbers here, with Rodewald and Murray standing out in ineptitude (and with far too many reps for this not to indicate some underlying problems). McCormick’s numbers overall are appalling on the powerplay–despite playing the second most shifts on the team (behind Chlapik) his on-ice and his production totals are absolute garbage (this also applies to Reinhart, who hasn’t played quite as much).
Our lines from this set of data change quite a bit: Werek-Chlapik-DiDomenico with Chabot-Lajoie, and Gagne-Perron-O’Brien with Burgdoerfer-Jaros. The only change I’d make to this is inserting White in the second unit (for O’Brien) and remove Burgdoerfer (for Harpur if he’s available, or O’Brien on the point), but this is far closer to what I would do with the man advantage.
Conclusions/Moving Forward
I think it’s worthwhile comparing usage to production directly before wrapping this up–rather than list the raw numbers I’ll stick to how they rank above and we’ll ignore players with less than 10 games (so excising Sexton, Blunden, Randell, and Harpur from consideration):
Forwards (usage/production)
DiDomenico 1st/1st
White 2nd/7th
Rodewald 3rd/no points
Paul 4th/11th
Chlapik 5th/2nd
McCormick 6th/9th
O’Brien 7th/6th
Reinhart 8th/10th
Gagne 9th/6th
Perron 10th/5th
Werek 11th/3rd
Defense
Chabot 1st/1st
Jaros 2nd/4th
Murray t-3rd/5th
Lajoie t-3rd/2nd
Burgdoerfer 5th/3rd
The two top prospects (White and Chabot) play a ton, as does the DiDomenico, but after that we get into coaching decisions that are at best a mixed bag. You look at the bottom-three set of forwards or where Lajoie sits and you can’t help but shake your head. As much as the org and coaching staff want to blame their problems on injuries or call-ups, clearly there are mistakes being made in how talent is being deployed.
This article is written by Peter Levi (@eyeonthesens)
eyeonthesens 