Heading into San Jose on February 10th, the Oilers knew a few things:
- After two losses in Disneywood, the Oilers needed a win.
- Given these are the teams the Oilers are chasing for an unlikely post-season spot, they had to salvage at least a few points to keep the California trip from being an utter disaster.
- After giving up a goal to Los Angeles on the first shot, and then a goal to Anaheim just two-and-a-half minutes into the game, the Oilers needed to keep San Jose off the scoreboard early and get on the scoreboard first.
So what happened? Naturally, the Sharks scored on their first shot on net. It was just barely more than a minute into the game. Though the rest of the game had more ups and downs than the Matterhorn, the reality is that the Oilers once again spent a large proportion of yet another devastating loss chasing the game from behind.
These two characteristics (giving up early goals, and playing from behind) are basically now canon for this lost Oilers season. I plan to look at the pathology of “playing from behind” in a later post, but today the focus is to dig a little deeper on the “early goals” situation.
For this analysis, I pulled the shots and goals given up by the Oilers this season (as of the San Jose game) from the database that drives the PuckIQ website*.
* As such, since both the database and the analysis of the data are mine, any errors you find are, sadly, also mine. Unless they’re the NHL’s.
Please note! Looking at this data is not for the faint of heart or weak of stomach. You have been warned.
The first thing I do when I pull a new dataset is look for interesting patterns. Here are a few things of note:
- It’s well known by now that the San Jose game was the tenth time this season the Oilers have given up a goal on the very first shot against. That’s almost 20% of their games this season!
- Somehow, though, the Oilers have managed to do this in back to back games. Twice! (CAR/CHI and DAL/STL).
If we extend the pattern from ‘first shot goals’ to ‘early goals’, those given up on (say) the first 6 shots of the game, it turns out the Oilers have some really neat streaks of this kind of thing:
- After that brilliant CGY game to open the season, the Oilers subsequently managed to give up goals on the 2nd (VAN), 4th (WPG), 3rd (OTT), 1st (CAR), 1st (CHI), and 2nd (PHI) shots. Whatever is causing this early goals problem, it showed up from the beginning.
- The Oilers followed that with a streak of giving up goals on shots 1, 1, 4 (DAL, STL, DET), a one-game respite against BUF, then 5, 5, 1 (BOS, ARI, TOR).
- Then it was 4, 6, 6, 1 (DAL, CHI, NSH, ARI).
- And now the latest streak that has refocused attention on the issue, where it began with a 1st shot goal (CGY), a temporary reprieve with COL, then 6 (TB), 1 (LA), 4 (ANA), and 1 (SJS).
It’s an admirable quality that if someone decides to do something, they really commit to it wholeheartedly. The Oilers clearly have.
Not that admirable in this case I guess.
The Extended Breakdown
Of course, by extended breakdown I *could* mean the Oilers season, or the psyche of Oilers fans.
But in this case, it means delving into the Oilers ‘early goal’ data a little deeper.
First, it helps to define what an ‘early goal’ might be. Since on average teams give up a shade fewer than three goals per game, and on average give up a c-hair more than 30 shots per game, that equates to about a goal every 11 shots. (note: shade and c-hair are formal statistical terms, I’m pretty sure)
For that reason, I would argue that a reasonable line in the sand for an “early” goal is one that occurs in six or fewer shots, which is why I used that as the line in the previous section.
If we look at the Oilers season, the Oilers have given up an early goal 28 times so far, an average of just over five minutes in. The earliest was 20 seconds in.
The PK has gotten lots of criticism, and a quarter (7) of those 28 goals were given up while shorthanded. (Not entirely clear if that percentage is good or bad, but with this team, safe to assume bad).
If we count how many occurrences of each of these shots-until-goal-given up games and chart them (called a histogram) we get this:
It will please the completists among you to know that the Oilers have successfully managed to hit for the cycle, and give up a goal in at least one game on the 1st, 2nd, 3rd, 4th, 5th, 6th, 7th, 8th, 9th, and 10th shots of the game.
One of the flaws I find in a lot of hockey stats reasoning is that an analyst will look at data for a player or team, but not set the league-wide context for that data. As in, something can seem unusual on its own, but in context you often discover it isn’t.
Well, bub, not in this case.
The Oilers have given up a goal on the first shot in 10 games this season. The next nearest are VAN and NYR with seven games. The league average is just 3.74 games.
The league (population) standard deviation is 1.99 games. At 10 games, the Oilers sit more than 3 standard deviations above the mean. (if this paragraph is statistical gobbledygook to you, it’s all good – feel free to ignore, it doesn’t change the story)
On the other hand, you may be oddly pleased to discover that every single team in the league has given up a goal on the first shot against at least once!
As we did earlier, we can extend this beyond goals-on-first-shot and instead look at early goals.
With this extended definition, the Oilers once again lead the league with 28 games giving up that early goal. The average is 20.
The standard deviation is 3.37, so this time the Oilers are a little less extreme – only 2.3 SD away! yay! – but still well outside normal bounds.
Yeah, the data confirm what you knew anyway – it’s not just bizarre, it’s unusually bizarre.
NHL play by play data lets us isolate who the players are on the ice when these early goals go in. It’s kind of a big list, but I’ve tabulated here the goalies, plus the defenders and forwards with (somewhat arbitrarily) at least four early goals against.
Because the penalty kill is a different beast, I’ve stuck to five-on-five for this particular dataset.
|Goalie||Games with Early Goal|
|Defenseman||Games with Early Goal|
|Gryba, Benning, Sekera||4|
|Forward||Games with Early Goal|
|Strome, Lucic, Letestu, Nuge||5|
* Yes, I’m also disappointed that it wasn’t Maroon 5
Certainly, giving up the early uncontested goal has been a significant part of this season’s struggles. It’s easy to look at certain players, or the goalies, or the coaches, and decide that’s who is to blame. I caution you against this.
First of all, I disagree with the idea that this is on the coaches. Whether we’re dealing with raw rookies or experienced vets, the Oilers players are all good enough to make the pros, and they’ve been playing the game for a loooong time.
They shouldn’t (ever) need a coach to tell them they need to be prepared to play hard from the first whistle.
For sure, goaltending has been subpar this season, and it’s as big a factor in this season’s underperformance* as brilliant goaltending was in last year’s overperformance. It is interesting to note, though, that every one of the Oilers three goalies this season have been victimized by an even strength first-shot goal at least once.
* Yes, I will go to my grave swearing to you this season is an underperformance, just like I was adamant throughout last season that it was an overperformance relative to the ‘true talent’ of the roster.
As for the actual players on the ice … well, certainly there are patterns there. The thing that I notice most as far as the forwards go is the sheer number of goals given up by the third and fourth liners.
Slepyshev at the top of the list may be one reason why he’s played as little as he has, despite generally being one of the more dangerous wingers many games.
As for the defenders, I won’t call out anyone specific, but the pattern points a little to players who’ve been defensively struggling much of the season – and they’d probably be the first to tell you that.
The one thing we know for sure is that we are waaaay out in La La Land as far as giving up first-shot goals and early goals this season.
If the players or the coaches or the goalies knew what the issue was, I’m sure they’d do something about it.
If I knew what it was, I’d fly up to Edmonton right now and bang on the doors of OEG screaming that I know how to fix the problem. (I would then subsequently direct my lawyer to contact OEG after she bailed me out).
As it is, the only suggestion I can make is maybe the Oilers should force feed every player — especially the bottom-six — two cups of coffee ten minutes before the game.
It can’t hurt, right? Because we’re already in next-season country.