Analysis · By · October 29, 2026

Detroit, the Red Wings have a finishing problem

When I think of finishing, I think about a team that plays extremely hard for 60 minutes and puts themselves in position to win the game. When they get their opportunities, they bury enough of them to take the lead or to extend it. When the game comes to a close, they don’t turtle but play high IQ hockey so that they don’t give up high-danger chances to the other team.

I also think of Monday night’s game in Edmonton as an example of good finishing. They kept their foot on the gas for the entirety of the game. When they fell down a goal, they immediately responded and kept possession in the offensive zone for long stretches. Even when they’d give up an opportunity or a goal, they continued to win the majority of their shifts.

All this still led to being down a goal with under two minutes left, but the process wasn’t broken even if the results up until that point were discouraging. They finished at expectation for the first time since Vancouver, 3 goals on 3.1 expected, and when Detroit does that this season they are 4-0-2. When they don’t, 1-4-1. And then…

Larkin went and TOTALLY redeemed himself. Well, maybe not totally. But it’s a nice parting gift now that he’s waiting on the sidelines for a trade to get finalized.

The rest is Postava history. The rookie goalie who got thrown into the Connor McDavid/Leon Draisaitl fire acquitted himself better than could have been expected. He’s the third Red Wings goalie in 25 years to win his first NHL start, after Jimmy Howard in 2005 and Petr Mrazek in 2013. Not only did he keep the Red Wings in the game, he stuffed McDavid, Draisaitl, and Ryan Nugent-Hopkins in order during the shootout to snatch victory from the jaws of defeat.

The vibes are high in Hockeytown. But I hate to say that some of the underlying numbers point to some deficiencies that could hold the team back from ultimate success. And it’s not just the Larkin and Edvinsson-shaped cloud hanging around the organization.

Seven dimensions of winning, Detroit against the top three, Oct. 29, 2026Radar of seven league percentiles for Detroit against the top-three average: Finishing 6 vs 52; Goaltending 42 vs 52; Attacking 52 vs 63; Defending 90 vs 82; Possession 77 vs 68; Power play 84 vs 86; Penalty kill 26 vs 68.Finishing6Goaltending42Attacking52Defending90Possession77Power play84Penalty kill26DetroitTop three, average · Carolina, Minnesota, EdmontonLeague percentiles · 2026-27 through Oct. 28 · the team page as of Oct. 29

What stayed the same in the Edmonton game

Medium danger finishing

How do I put this lightly? This is THE problem anchoring the Red Wings to the bottom of the Atlantic (pun intended…may as well be the ocean).

Detroit was 2 of 11 on medium danger shots, their best since Ottawa, but that includes Larkin’s game-tying goal, which just so happened to be right on the border of high-danger/medium-danger. And some zone-based models have it as high-danger. Based on the play itself and all that led up to it, I would also consider it high-danger, dropping that number down to 1 of 10.

So what? Let me explain.

The Red Wings are finishing medium-danger chances at 6.6%, 30th in the league against a league rate of 11.5%. The only teams below them are Los Angeles and Tampa Bay, and Tampa is the one worth a look. The Lightning sit on the same 9-goal medium-danger hole Detroit does and are a .682 team anyway, because they finish everywhere else. Detroit finishes nowhere else.

Medium-danger conversion by teamBar chart of medium-danger shooting percentage for all 32 teams; Detroit is 30th at 6.6% against a league rate of 11.5%.Medium-danger conversion, all 32 teams, through Oct. 28Goals divided by shots carrying .08 to .20 expected goals (MoneyPuck). Detroit in red.5%10%15%20%league 11.5%SEADETDetroit, 11 of 167, 6.6%bestworst30th of 32. Only Los Angeles (5.8%) and Tampa Bay (6.2%) are lower. Detroit, 11 goals on 167 shots.

If the Red Wings scored even league average on medium-danger chances, that would amount to 8 goals over the course of just 12 games. Here are the games those goals would have swung:

  • October 4 vs. Winnipeg, L 2-3 (0-16 on medium-danger shots)
  • October 10 at Montreal, L 0-1 SO (0-20 on medium-danger shots)
  • October 20 at Seattle, L 3-4 SO (1-12 on medium-danger shots)

There were other rough outings in there, too, but the result was either already a win (vs PHI, 1-17), or the result was far enough out of reach that it would have still been a loss (vs SEA, 0-15). That’s 4 points right there that could have the Red Wings sitting in 2nd place in the Atlantic if everything fell correctly (17 points, 8-3-1).

Obviously hockey doesn’t work like that. But when you’re sitting in the 6th percentile for finishing as a whole, with the ninth-worst medium-danger start of the 160 twelve-game openings in the last five seasons, these are the chances that decide where you sit in the standings, especially if you’re a fringe team like the Red Wings.

The player having the worst finishing season for the Red Wings is Lucas Raymond, who is 0-24 with 2.8 xG on those shots. Arvidsson, who has five goals overall, is still just 1-21 with 2.7 xG. DeBrincat is 2-25 with 3.1 xG. That’s 3 goals on 70 shots where the model expected 8.5.

Medium-danger goals against expected, Detroit shootersPaired bars for eight Red Wings: expected goals on medium-danger shots against actual goals; Raymond is 0 on 2.8, DeBrincat 2 on 3.1, Arvidsson 1 on 2.7, Copp 1 on 1.9.Detroit's medium-danger shooters: goals against what the model expectedHollow bar = expected goals on those shots (MoneyPuck). Solid = goals. Regular season through Oct. 28.Lucas Raymond2.8 xG on 24 shots0 goalsAlex DeBrincat3.2 xG on 25 shots2 goalsViktor Arvidsson2.6 xG on 21 shots1 goalDylan Larkin2.0 xG on 16 shots1 goalAndrew Copp1.8 xG on 15 shots1 goalMichael Brandsegg-Nygård1.8 xG on 15 shots1 goalEmmitt Finnie1.4 xG on 12 shots2 goalsMichael Rasmussen1.3 xG on 12 shots1 goal0123Raymond, DeBrincat and Arvidsson together: 3 goals on 70 shots where the model expected 8.6.

But I’m feeling optimistic, and it’s not just because their xG is high. It’s because all three have finished above expected for years: Raymond and DeBrincat at about 1.17 goals per expected goal over the past two seasons, Arvidsson at 1.02. This is in stark contrast to a player like Copp who is 1-16 with 1.93 xG. His entire career looks like this. So reverting to league average for Raymond, DeBrincat, and Arvidsson is taking the pessimistic view.

The Red Wings will never be a top-tier finishing team with this roster, but they could be a top-20 one instead of 30th, and that alone changes their place in an Atlantic where 16 points is the middle of the pack.

Penalty kill

You may say the penalty kill looked better than it typically does, and Edmonton going 1 for 3 on the power play is below expectation.

And to you, I would say you’re right…on the surface. But the PK bled chances to the tune of 12 expected goals allowed per 60, worse than their season average of 8.7, which is itself 24th in the league.

There was one difference that made the PK look respectable: Michal Postava. He has an .875 save percentage on the PK this season compared to John Gibson’s .633. That works out to 0.4 goals saved above expected for Postava and 6 goals below expected for Gibson, on 30 shots.

John Gibson has been the worst penalty killer on the Red Wings this year. It’s too simple to say that the supporting cast has been leaving him out to dry. He’s been in a slump to begin the year. His .633 is 58th of 61 goalies who have faced eight or more power-play shots.

And yet, I’m optimistic again. This penalty kill is not broken. They give up the 5th fewest high-danger shots in the league on the kill, so teams aren’t getting to the crease. The trade is that Detroit allows the 4th most medium-danger shots, so the shooting is coming from the second layer, and a league-average goalie handles that. More successful kills follow if Gibson just saves the number of goals expected.

Even more positive news: penalty-kill results are about the least sticky thing in hockey. Goals against minus expected on the kill has a year-to-year correlation of -.07 across the league, which is to say none, so there is no evidence Gibson’s start tells you anything about his next 30 shots. For context, he ran a league-average .839 on the kill through his first eight games last year and finished at .846, and by expected goals he was 2.7 goals better than average.

Gibson's penalty-kill save percentage, running totalTwo lines: Gibson's cumulative shorthanded save percentage through 2025-26, finishing at .846 after a league-average start, and through eight games of 2026-27 at .633.John Gibson on the penalty kill: save percentage, running total by gameShots on goal at 5-on-4 against him. Last season in black, this season in red..600.700.800.900.100league .856 last season2025-26: .846 (30 GA on 195 shots, 57 GP)2026-27: .633 (11 GA on 30 shots, 8 GP)GP 10GP 20GP 30GP 40GP 50Lines start at eight shots faced. Last season: .833 through 8 GP, .830 through 12, .846 at the end.

The other reason for optimism is that so far this season the Red Wings have spent the 12th most minutes on the PK, 5.2 a game against 4.0 last year, so some of the goals they’ve given up are sheer volume. Last season, they were shorthanded the third-fewest minutes in the league, and the league as a whole is only up from 4.5 to 4.9 minutes a night. This is partially due to small sample size and more playing time for younger players. The Red Wings have been more disciplined in recent seasons.

When you add up the 8 goals below expected from finishing and the 4 goals below expected on the penalty kill, that’s a 12-goal swing if you just plug in league averages. I don’t think the Red Wings are league average in either of these categories, but I do think they are better than they’ve shown. So we may not be looking at 2nd in the Atlantic, but you might be able to squint hard enough to see that 8 turn into a 6.

What was different in the Edmonton game

High-danger shots against

I keep talking about how optimistic I am in the face of negative data. Let me tell you how optimistic I am in the face of positive data.

Do you know how many high-danger shots the Red Wings gave up at 5v5? Zero. That’s right. Zero. Against a team with McJesus and his right-hand man Leon Draisaitl, Detroit held Edmonton to 1.06 expected goals at 5v5 and took 65% of the expected goals in the game, their fourth-best share of the season against the best offense they’ve faced.

Do you know why that was possible? Execution. This was the most dedicated to a game plan I’ve seen the Red Wings all season, and honestly maybe even last season. Every line was dedicated to executing the game plan, and it resulted in zero head-scratcher goals. You got a deflection off a skate for a goal, a changeup one-timer on the power play, and a barely legal redirect.

Sometimes the bounces don’t go your way. If this happened in March 2026, this team would probably have folded. But the battle-tested, resilient Red Wings that are fighting every single day to drown out the peripheral noise are stronger than that team.

After every setback, they responded. It’s this attitude that gives me the most optimism. I really shouldn’t call this section “High-danger shots against.” I should call it “togetherness.”

At the beginning of the season, I was afraid that this would be a lost season and a setback for the trajectory of Mo Seider and Lucas Raymond. I was afraid DeBrincat would leave in free agency. I was afraid none of the young guys would assert themselves.

Instead, what I’m seeing is a willingness to fight, and to fight together. Dylan Larkin’s departure will be sad in the context of being the former captain of this team and spending his entire career here. But it won’t be sad in the sense that now the locker room can move on and band even closer together.

What the Red Wings will take into the Chicago game tonight

I’m not going to say anything groundbreaking here. I’m looking for just two things. One is for the team’s execution to continue to improve. This team needs to take that positive trend from the Oilers game and apply it to a team they should beat. This, in turn, will lead to fewer minutes on the PK.

The other thing that I’m looking for is for Lucas Raymond to bury a medium-danger chance to break his season-long slump. He’s a much sharper shooter than he’s shown thus far, and the Blackhawks are a perfect team to turn it around against.

Beyond tonight, I’m looking for this team to stay competitive. I predicted at the beginning of the season that they would get just 75 points, but this team is making me think that a .500+ point percentage is possible.

Finishing, the shooting kind, is a problem for Detroit right now, and I think they end the year closer to 20th than 30th. The other kind of finishing, the 60-minute kind they showed in Edmonton, is the one I’d bet on.

Method: season numbers are MoneyPuck's team, skater and goalie tables and shot-level data through Oct. 28, 2026 (regular season, shootouts excluded), with ranks against all 32 teams; the Edmonton game is MoneyPuck's shot log joined to the NHL play-by-play; penalty-kill rates are 4-on-5; the goalie save percentages on the kill are shots on goal while shorthanded; Gibson's 2025-26 kill numbers are rebuilt game by game from last season's shot file. Danger bands are MoneyPuck's (an expected goal under .08 is low, .08 to .20 medium, .20 and up high). The 160 twelve-game openings are every team's first 12 games of 2021-22 through 2026-27, compared as goals per expected goal against that season's league. Year-to-year correlations are 2024-25 to 2025-26 full seasons, 32 teams. The record splits and the three-game swing use the NHL's standings as of Oct. 29.