10/06 Power-Play Changes: Bennett, Toffoli, Lindholm Onto PP1
Teams have played 2 to 4 games through October 5, so this is a small sample and it is labelled as one. What a few games can already show is who the coaches send out first on the power play, and that decision moves before the points do. This article runs every two weeks early in the season, then weekly once there is enough to compare.
The first unit here is the five skaters who played the most power-play time together in a game, read from the NHL's shift charts. It is not the five with the most power-play minutes each: a long 5-on-3 or a second unit that stays out can hand a PP2 skater more minutes on a given night without him ever joining the top group.
The fantasy angle in one line: skaters on their team's first unit in most of their games are scoring 0.29 power-play points a game so far (151 skaters), against 0.10 for second-unit skaters who have not been on PP1 at all (145 skaters).
TL;DR
New on PP1
- Sam Bennett (FLA, C, 81% owned) took Aleksander Barkov's spot on Florida's top unit Saturday against Anaheim and played 6:12 on the power play. Our chart has him staying there unless they want to rotate in Anton Lundell.
- Tyler Toffoli (SJS, RW, 51%) moved up with Macklin Celebrini a late scratch Sunday, 2:27 on the advantage, and projects to hold PP1 as long as Celebrini is out.
- Paul Cotter (VAN, LW, 41%) has had the top unit since Jake DeBrusk went down, 3:00 Saturday against Vegas, one power-play point in four. DeBrusk has missed two with an illness and gets the seat back when he's healthy - treat Cotter as a rental.
- Elias Lindholm (BOS, C, 37%) and Travis Konecny (PHI, RW, 82%) both ran PP1 on Sunday, Lindholm in for JJ Peterka and Konecny in for Porter Martone.
- Mason Marchment (SJS, LW, 18%) got 2:33 on the top unit in place of Igor Chernyshov. We're not sure where SJ is going with those 2.
Dropped off PP1
- JJ Peterka (BOS, RW, 54%) and Porter Martone (PHI, RW, 48%) each lost the top unit last game after holding it in two of three.
- Anton Frondell (CHI, C, 17%) was off Chicago's top unit Friday against Buffalo with just 0:46 of power-play time, after starting the season on it.
League leaders
Utah has the hottest power play at six goals on nine chances, a 66.7% rate nobody sustains. Dallas is generating the better looks - 2.66 expected goals on eight chances, worth about 33% - and is the one more likely to still be there in a month.
The biggest changes to PP1s last game
Each team's first unit in its last game against the five it usually sends out. A skater counts as new when he was on the top unit in fewer than half of the earlier games; one counts as gone when he was on it in more than half and was not last game. Where a beat report has already moved someone and the game has not shown it yet, the move is marked projected.
San Jose Sharks changed 2 of five on Oct 5 at DAL: Tyler Toffoli and Mason Marchment in, Macklin Celebrini (did not play) out. That five played 1:24 of power play together.
Boston Bruins changed 1 of five on Oct 5 vs OTT: Elias Lindholm in, JJ Peterka out. That five played 2:17 of power play together.
Chicago Blackhawks changed 1 of five on Oct 3 at BUF: Artyom Levshunov in, Bowen Byram (did not play) and Anton Frondell out. That five played 1:01 of power play together.
Every team's power-play units, game by game
Pick a team. A dark cell means the player was in that night's first unit (the five with the most power-play time together), an outlined cell the second unit. The number is his own power-play time that night. The last column is the current projected unit from our lines, which includes the most recent practice and morning-skate reports.
San Jose Sharks power play, game by game| Player | Oct 1 vs FLA 5:38 PP PP1 3:37 | Oct 3 vs LAK 3:54 PP PP1 2:21 | Oct 5 @ DAL 4:32 PP PP1 1:24 | Now |
|---|---|---|---|---|
 L. CagnoniD | 5:21 | 3:07 | 2:39 | PP1 |
 A. WennbergC | 4:52 | 3:06 | 2:30 | PP1 |
 W. SmithRW | 5:21 | 2:23 | 2:29 | PP1 |
 I. ChernyshovLW | 3:37 | 2:25 | 1:46 | PP2 |
| M. CelebriniC | 5:21 | 3:06 | - | - |
| T. ToffoliRW ▲ PP1 | 0:32 | 1:30 | 2:27 | PP1 |
| M. MarchmentLW ▲ PP1 | 0:38 | 0:48 | 2:33 | PP2 |
 I. StenbergLW | 0:35 | 1:31 | 1:50 | PP1 |
 M. MisaC | 1:07 | 0:47 | 2:00 | PP2 |
 D. OrlovD | 0:38 | 0:47 | 1:54 | - |
Moving up
Sam Bennett, FLA C. On the top unit Oct 4 vs ANA, after PP1 in 0 of 2 earlier games, replacing Aleksander Barkov, 6:12 of power play that night, projected PP1 now, 0 PPP in 3 GP, 81% owned on ESPN. Seen: in last game and projected.
Tyler Toffoli, SJS RW. On the top unit Oct 5 vs DAL, after PP1 in 0 of 2 earlier games, replacing Macklin Celebrini, 2:27 of power play that night, projected PP1 now, 1 PPP in 3 GP, 51% owned on ESPN. Seen: in last game and projected.
Elias Lindholm, BOS C. On the top unit Oct 5 vs OTT, after PP1 in 1 of 3 earlier games, replacing JJ Peterka, 3:33 of power play that night, projected PP2 now, 0 PPP in 4 GP, 37% owned on ESPN. Seen: in last game.
Travis Konecny, PHI RW. On the top unit Oct 5 vs TBL, after PP1 in 1 of 3 earlier games, replacing Porter Martone, 3:22 of power play that night, projected PP2 now, 0 PPP in 4 GP, 82% owned on ESPN. Seen: in last game.
Paul Cotter, VAN LW. On the top unit Oct 4 vs VGK, after PP1 in 0 of 3 earlier games, replacing Jake DeBrusk, 3:00 of power play that night, 1 PPP in 4 GP, 41% owned on ESPN. Seen: in last game.
Mason Marchment, SJS LW. On the top unit Oct 5 vs DAL, after PP1 in 0 of 2 earlier games, replacing Igor Chernyshov, 2:33 of power play that night, projected PP2 now, 0 PPP in 3 GP, 18% owned on ESPN. Seen: in last game.
Moving down
Anton Frondell, CHI C. PP1 in 2 of 2 earlier games, not on the top unit Oct 3 vs BUF, 0:46 of power play that night, projected PP1 now, 0 PPP in 3 GP, 17% owned on ESPN. Seen: in last game.
JJ Peterka, BOS RW. PP1 in 2 of 3 earlier games, not on the top unit Oct 5 vs OTT, replaced by Elias Lindholm, 2:20 of power play that night, projected PP1 now, 0 PPP in 4 GP, 54% owned on ESPN. Seen: in last game.
Porter Martone, PHI RW. PP1 in 2 of 3 earlier games, not on the top unit Oct 5 vs TBL, replaced by Travis Konecny, 1:44 of power play that night, projected PP1 now, 0 PPP in 4 GP, 48% owned on ESPN. Seen: in last game.
Opportunities and production by team
Power-play opportunities are the chances a team gets; PP% is goals per opportunity. Expected PP% is the same thing with expected goals (shot quality) in place of goals, so it says how good the chances have been regardless of whether they went in. Goals minus xG above zero means the finishing has run ahead of the chances.
Florida Panthers have drawn the most chances, 5.7 a game, and Calgary Flames the fewest, 1.7. Utah Mammoth have scored 6 power-play goals on 2.09 expected, the most finishing above the chances (+3.9). Florida Panthers have 1 on 3.76 expected (-2.8), the power play most likely to start scoring if the chances keep coming.
The shooters: shots, iPP and where they shoot from
Pick a team for its power-play shot map and every skater who plays on one of its units: shots on goal, goals, shooting percentage, iPP and where his attempts come from, with his own shot map. iPP is the share of the power-play goals scored with him on the ice that he had a point on: a high one is a player the puck goes through.
San Jose Sharks power-play attempts18 attempts, 2 goals, 2.38 xGMost power-play shots on goal so far: Leon Draisaitl (EDM) 7 shots, 1 goal, iPP 100%; Andrei Svechnikov (CAR) 7 shots, 0 goals, iPP 67%; Dylan Guenther (UTA) 5 shots, 3 goals, iPP 83%; Sebastian Aho (CAR) 5 shots, 2 goals, iPP 100%; Nick Schmaltz (UTA) 5 shots, 2 goals, iPP 40%.
Seattle Kraken take the most from the middle of the ice, 82% of attempts from the net front or slot; Columbus Blue Jackets lean on the point the most, 46% of attempts.
Power-play production by player
First-unit skaters, and second-unit skaters with a power-play point (a unit is PP1 or PP2 in at least half of a player's games). Power-play points, shots on goal, shooting percentage and iPP, with the share of the team's power-play time each one is on the ice for.
UTA RW
WPG LW
CAR C
VGK C
EDM RW
ANA LW
COL C
LAK RW
PIT DLeaders so far: Dylan Guenther (UTA) 5 PPP in 3 games, 2:28 a game; Kyle Connor (WPG) 4 PPP in 3 games, 2:40 a game; Cole Perfetti (WPG) 4 PPP in 3 games, 2:38 a game; Clayton Keller (UTA) 4 PPP in 3 games, 2:22 a game; Sebastian Aho (CAR) 3 PPP in 3 games, 6:02 a game.
Rostered in under 30% of ESPN leagues, on a first unit and with a power-play point already: Tommy Novak (PIT C, 16% owned) 2 PPP, 55% of the power play; Joel Farabee (CGY LW, 5% owned) 2 PPP, 68% of the power play; Andrew Copp (DET C, 2% owned) 1 PPP, 64% of the power play; Viktor Arvidsson (DET LW, 10% owned) 1 PPP, 62% of the power play; Maxim Shabanov (MIN RW, 2% owned) 1 PPP, 83% of the power play.
How it works
A move onto the first unit is worth acting on when it shows up in a game and the projected lines agree, and the player is still on waivers. A move marked projected comes from a beat report and has not been seen in a game yet: check that it holds tonight. Team opportunities and expected PP% say which power plays will produce the most chances for whoever is on them.
How it is measured: 2 to 4 games per team, so treat every number as early. Each game's power-play units come from the NHL shift charts: every stretch where a team had five skaters against three or four is credited to the five on the ice, and the five with the most time together is that night's PP1; the most-used five sharing no more than two of them is PP2. A five with under 30 seconds together is no unit. When the second unit played within 85% of the first unit's time, both count as a top unit that night, so a 1A/1B split is not a move. Opportunities come from the NHL; expected goals from the site's shot model; ownership from ESPN. Once the season has a few weeks of games this article switches to the changepoint model, which waits for a role to hold before calling it a change.