Yesterday, my fellow Detroit Red Wings contributor Tom Mitsos discussed how Detroit’s fourth line has been the an integral part of the Red Wings’ success thus far. However, I’ve carried an opposite opinion for most of the season and decided that it would be an opportune time to illustrate why I believe that Detroit’s fourth line may ultimately be their undoing.
The Role of the Fourth Line
When talking about Detroit’s fourth line, I’m focusing on center Luke Glendening and wingers Joakim Andersson and Drew Miller. This season, Mike Babcock has utilized the fourth line in a variety of ways, but most prominently, he has utilized them as a defensive unit tasked with stopping the opposition’s top scoring line. Theoretically, there are pros to this strategy, most notably the fact that the Wings are able to shelter superstars Pavel Datsyuk and Henrik Zetterberg by providing them more ice time against a lower quality of competition. On the surface this strategy has worked, as Glendening’s line has faced the highest quality of competition, but is yielding only 1.77 goals against/60 minutes at 5-on-5. You also see Glendening’s faceoff percentage at 55.4% and your opinion is swayed to believe that this fourth line is effective.
The Fourth Line Reality
While everyone focuses on the goals against for the fourth line, very few are talking about the other side of the puck and that’s where the issue exists. For the season, Glendening has been on the ice for just 17 5v5 goals despite playing more than 500 minutes at 5v5. If he was able to play the entire 60 minutes, the Wings would average a paltry 2.01 goals at 5v5. However the problem runs even deeper than just the goals scored versus goals allowed – the Wings become one of the worst puck possession teams as defined by Corsi when the fourth line receives too much ice time.
I decided to identify the total time the Red Wings have spent at 5v5 this season, followed by the total time Luke Glendening has spent on the ice at 5v5. I then calculated the percentage of total even strength minutes available that he plays. I then stratified his data into games where he played >25% of the total even strength minutes against games where he played <25% of the available even strength minutes. I then decided to take a look and see how the Red Wings overall stats were affected by the amount of ice time Glendening received. The Corsi and TOI numbers reported are for the team and not Glendening’s on-ice Corsi and TOI numbers.
Games where Glendening played >25% of available 5v5 minutes
| Games w/ >25% TOI | Total 5v5 TOI in those games | Wings 5v5 CF/60 | Wings 5v5 CA/60 | 5v5 GF/60 | 5v5 GA/60 |
|---|---|---|---|---|---|
| 10 | 468.68 | 41.8 | 63.2 | 2.69 | 1.79 |
Games where Glendening played <25% of available 5v5 minutes
| Games w/ <25% TOI | Total 5v5 TOI in those games | Wings 5v5 CF/60 | Wings 5v5 CA/60 | 5v5 GF/60 | 5v5 GA/60 |
|---|---|---|---|---|---|
| 38 | 1769.63 | 54.5 | 38.4 | 2.03 | 1.90 |
Obviously we are looking at very small sample sizes here, but the Corsi numbers are extremely troubling. In games where Glendening plays <25% of available 5v5 minutes, the Wings average 16 more shot attempts per 60 minutes of 5v5 play than their opposition. Once Glendening crosses that 25% TOI threshold, all of a sudden the Wings offense dries up. They average 13 fewer shot attempts and opponents average a whopping 63.2 attempts per 60 minutes. The 63.2 attempts against would rank the Wings 2nd to last in the NHL, with only the Buffalo Sabres allowing more attempts.
Essentially, Glendening’s line is a bend-but-don’t-break line. Babcock throws them out there, they get dominated, but they don’t give up goals. Glendening’s 5v5 PDO right now, a metric that measures on-ice save percentage + on-ice shot percentage, is sitting at 102.8, buoyed by a fantastic on-ice save percentage of 94.35%. If that save percentage in any way regresses, all of a sudden you’ll start to see the goals end up in the back of the net and the domination of the fourth line will become more apparent.
Think about it logically. You are throwing your worst puck possession line out on the ice against the best line of the opposition, and asking them to bend-but-not-break. You are allowing the best players on the other team to have the puck a majority of the time and dominate the offensive chances. Right now the issues haven’t manifested because the goaltending has been superb when the fourth line is on the ice, but it’s only a matter of time before this regresses.
My Suggestion for the Fourth Line
Kyle from Winging It In Motown addressed this topic a couple of months back and suggested that the Wings try to shelter Glendening and his linemates from the top competition a bit more and play strength-on-strength. I’ve been a staunch advocate of this idea, specifically using Detroit’s “Kid Line” of Riley Sheahan, Tomas Tatar, and Gustav Nyquist against the best opposition line. For me, there’s the theoretical benefit of that line possessing so much speed that they will be able to forecheck and backcheck aggressively, taking away time and space from the opponents. Statistically speaking, the “Kid Line” has been Detroit’s best line in terms of puck possession over the last two seasons. Granted, those numbers have come against a weaker competition, but they have at least earned the opportunity to be challenged with stiffer competition.
In the playoffs, every mistake is heightened. The checking intensifies, the goaltending rises to another level, and scoring chances are at a premium. If Detroit wants to be successful, they would be wise to not utilize their current bend-but-don’t-break strategy with the fourth line. Detroit’s at its best when the fourth line plays “fourth-line minutes” and the Wings give more time to their top three lines. Babcock would be smart to adopt this strategy, otherwise the Wings may be faced with another early playoff exit.
Data from War-On-Ice
The corsi ratio is a little misleading , specifically w/t respect to 5>5 x 25% threshold . It doesn’t appear that this model is being used w/t other factors w/t consideration to the 2&3 or 4 lines ,the snapshot is of teams 1st lines and not a complete picture of the combined percentage of all 4, lines as opopposed to teams number one line .
Despite all the advanced stats used, you are failing to look at this in its most simplistic way. The fourth line for the Red Wings is typically matched up against the top line of the opposition. Those games where Glendening plays over 25% are likely teams with a ‘superstar’ on their team who the fourth line matches with all night. Those guys when matching up against Crosby, Parise, Stamkos, Ovechkin etc. will never even try to match their insane shot output. Glendening and company are not out there to shoot the puck as much as the other team’s top line, they are out their to hold down the fort against offensive competition. This gives our top line a matchup they can dominate against. If they eat up all the toughest minutes without any damage, then they did their job and our skill guys can have a better chance to score against lines 2-4 for the other team.
First off, why are you only using Glendening’s TOI to represent the entire fourth line? I think that if you were to use the TOI for the entire fourth line you would get very different results. And what makes you think that Glendening is the reason for the lower stats for the entire team? Maybe the opposite is true, and the reason Glendening has more TOI in games where the team is struggling is because his defensive abilities are more useful when the team has less puck possession. You also don’t mention that in games where he has more than 25% TOI the 5v5 GF and GA stats are both better than when he has less than 25% TOI.
All in all, I get what you’re saying, but I also think that you can’t use stats alone in order to assess the effectiveness of a line meant to eat up time and maintain the score. You have to look at the way Babcock is utilizing this line and if they’re doing what the coach expects of them.
Your argument is built upon the faulty assumption that Mike Babcock would fail to make adjustments if his 4th line’s GAA regresses to the point where the opposition’s 1st lines are regularly lighting them up.
Babcock’s current use of the Red Wings 4th line will only be their undoing if he allows it to be their undoing. Do you think he’d just obstinately roll the Glendening line out against their opponents top lines if they’re being torched by them? I don’t. I don’t see him allowing that for more than a couple games at most, and certainly not long enough for it to become the team’s “undoing.” I see this as a strategy to keep Zetterberg and Datsyuk fresh for increased playoff minutes.
Whatever the reason, it’s kind of hard to argue with its success to this point.
Hi Chuck thanks for commenting. I think you too my post one step further than I was intending. I was merely trying to show that eventually this massive shot attempt discrepancy will come back and bite the Wings IF no adjustments are made. Obviously Babcock is a supreme coach and would make the adjustment once he begins to see the ill effects but my concern is that he should be looking to make the change before Detroit realizes regression. Thanks again for your input
I suppose I’m a bit confused by the stats you’ve shown. While the Corsi values aren’t as good as when Glendening plays less than 25% of the 5v5 minutes, we’re scoring .66 more GF/60, and giving up .11 less GA/60. Seems like the Wings are significantly more efficient when he’s out there for longer. In addition, is response you Alex, you point out that the few goals against isn’t due to their outstanding play, and while opposing teams may be getting more shots off, that doesn’t mean they’re better shots. The fourth line may be limiting their shooters by not allowing off good shots, and as a result, could the opposing shooters are trying to force the issue?
Hey Tom thanks for commenting. I suppose what I’m missing here is the graph that shows where the shot attempts have occurred. Unfortunately the site that provides that information, War On Ice, has been down for the or 36 hours or so. I have seen the shot distribution and it is no different from what the rest of the team allows. In fact it’s very similar. The issue then is not quality, but rather quantity. Eventually the exceedingly high quantity will catch up with them
It’s an interesting view that the 4th line isn’t as big a net positive as it looks right now.
I’d even likely agree with that.
But, on the other hand… I still think it is great when the 4th line can soak up a lot of minutes. Especially for the way the rest of the team is constructed. Pavs, Z, and the rest of the offensive production players are getting older. It is very possible that they are actually more effective with 15 minutes a game than they would be with 23+. Also, limiting their minutes limits their injury exposure. Staying healthy has got to be a key consideration for the team.
So, the 4th line might not add that much. But they may still be a lot less costly than the alternatives. Detroit has used this strategy before, with reasonable success. Draper and Maltby soaked up a lot of minutes to protect Stevie and Shanny, too.
And there is nothing saying you’d want to ride that time distribution and match-up far into the playoffs.
I just don’t see the downside to the 4th line, at this point.
Can’t say I agree at all with using Nyquist defensively. Although I cannot backup with stats, I watch every game and he is not a good positional defender.
Hi there, thanks for commenting. My rationale for using Nyquist’s line doesn’t have anything to do with his defensive capabilities. Rather, I think because Nyquist’s line is so good offensively, they may be able to reduce the amount of time the opposition has the puck simply because they possess it. It’s a different way of thinking about defense but think about it as using your best offense as your defense.
Are you able to get visibility into where the shots are coming from? Just using save percentage can be a bit deceptive at time – especially if Glendening, Miller, and Andersson are doing a good job of clogging passing lanes and limiting the opposition to shots from the point with limited traffic in front.
If that is the case (and I’m not saying that it is), then it’s not necessarily a given that the save percentage would regress in the future.
Hey Dave thanks for reading. Yesterday when I was putting this together, War-On-Ice was down, but today I can provide that info for you. When looking at the shot placement, when Glendening is on the ice, the Wings allow a higher relative rate of shots in the slot and from the points when compared to the minutes when he’s off the ice. His line is not forcing “bad” shots so I feel confident that they will regress.
Line changes go both ways. Need to consider home vs. away games. Would also like to see opponents line stats that face the kid line because our fourth line occupied their top line.
Hey GK, I’ll gather this for you and post in a comment later. Thanks for reading!
Agree, with one caveat. Are there games you can point to as examples that show this phenomenon? It seems to me that Glendening getting more than 25% of ice time at 5 vs 5 would be because of a perceived necessity rather than because Babcock thought that was a good idea at the outset of the game. In other words, were the Wings forced to give the fourth more time because they were playing behind the 8-ball somehow?
Hey Geoff, thanks for reading! Im at a different computer for the next couple of hours, but I will return and list the 10 games I pulled the sample from. Ill also list the score differential to see whether or not the Wings were getting blown out or had a significant lead.
The 4th line isn’t out there to score goals, they are out there to stop the other team from scoring hence the horrible goals for stat. If you can grind down an opposing teams top line all night and keep them to the outside for bad shots (what should be easy saves) then who cares if they don’t score. You pay guys like Pav, Hank, Gus, Tats, and Franzen (for about 10 games then he needs a rest for 20). No team in the league has 4 lines that are all offense, the game is about mismatches and this 4th line is what gives the Wings so many mismatches.
Hey Alex thanks for commenting. I think the point I was trying to illustrate is that right now everything looks good because they are limiting the goals against. However, the reason they have so few goals against isn’t due to their outstanding play – rather its due to an unsustainably high SV%. They are allowing an incredible amount of shot attempts against and the offense dries up. My point isn’t that this line shouldn’t exist – rather we should not be matching them against the opposition’s best line because they are not doing what you suggested.