The Tampa Bay Lightning loss to the Carolina Hurricanes on Sunday was awful. No one needs to dig deep into the numbers to know that. The Bolts looked terrible to the eye and seemed to go five minutes at a chunk without getting into the offensive zone let alone doing anything once they got there. The stat that jumped out immediately to everyone watching live was the Lightning only getting two shots on net during a stretch that lasted from two minutes left in the first period until the game ended in overtime. A truly putrid performance.
Because I am a glutton for punishment and feel compelled to know just how bad the game was, I pulled individual game results from Corsica going back to 2007 so that we can see in detail exactly how atrocious the Lightning played on Sunday evening. I should also give a shout out to Twitter user Rafael who asked where this game ranked among the all time franchise stinkers.
This isn’t a complicated analysis. We’re going to look at where this game sits in the distribution of game outcomes going back to 2007, which is the first year we have the data needed to calculate expected goals. We’re using expected goals here because hockey is a game plagued by randomness. Sunday was a great anecdote to demonstrate that as the Canes clearly housed the Bolts but still needed overtime to squeak out a one goal victory.
To get the basics out of the way, here are the Lightning’s expected goal shares from Sunday night:
5v5 Adjusted: 21.3%
5v5 Unadjusted: 21.5%
All Situations Adjusted: 20.1%
All Situations Unadjusted: 19.6%
As a reminder, an expected goal share of 20% means that the Lightning would be expected to score 20% of the total goals scored in the game by both teams. In other words, they’d be expected to lose 4-1, which for this type of model, is an extremely one sided result.
Those are horrendous numbers but we already knew that. To visualize how bad, let’s plot them on a histogram showing how frequently a team achieves an xG Share within a certain range. What the plot below shows is that most games are close with more lopsided outcomes increasingly rare as we move out from the center. The spot where last night’s Lightning game falls is marked with a dashed blue line.
Yep. That looks about right. Sunday evening’s performance by the Lightning was among the very worst in the data set and sits all the way at the bad end of the tail. Few teams have been demolished the way the Bolts were by the Canes.
In fact, the outcome is such an outlier that I couldn’t help but check on the rank. At 5v5, the trouncing ranks as the 118th or 127th most lopsided game using unadjusted and adjusted stats respectively. In all situations, those numbers drop all the way to 34th and 39th out of all 57,274(!) NHL games going back to the 2007 season. That’s truly astounding when you think about it. Those are 2014-2015 Tank for McDavid Buffalo Sabres numbers.
When we refer to adjusted in this context, we’re talking about adjusting for score and venue. The reason the Lightning get a small bump from the adjustment is because they were playing with a lead on the road for much of the game and teams in both of those situations tend to give up more shots.
If we limit our data set to just Lightning games, the results of course get even worse. At 5v5, this was the 5th worst game for the Lightning since 2007 using both adjusted and unadjusted measures. In all situations, this was the 2nd worst game if we use the adjustment. But if we just use all situations unadjusted expected goal share, Sunday night was the worst game the Lightning have played in the modern stats era. Not even the awful 08-09 team managed to play a worse game.
Typically, my response to one bad game in October would be mostly to ignore it and that’s probably the right approach here. But that game was so awful, so atrocious, so spectacularly stinky, that it’s tough to write off. Sure, the team doesn’t need to be at their best right now. But they shouldn’t be THAT bad. Even a team’s worst showing shouldn’t drop them to that level. At least not a good team.
In the past, the team has always taken a “stick to our game plan and we’ll be fine” approach. But I think maybe a little shakeup is warranted after that showing. Whether that shakeup should be something that can be noticed outside the locker room is up for debate. But I would hope that everyone took a least a few moments for introspection to consider how a game like that can happen to a Stanley Cup contender and what needs to be done to make sure it doesn’t happen again.
Everyone except of course, Curtis McElhinney who is the only one to come out of that nightmare without a share of the blame. He better not be paying for a meal for the rest of this road trip.