clock menu more-arrow no yes

Filed under:

"What Hurts the Most": When was the Devils Season Over?

New, comments

How long does it take to make a season? How long does it take to break a season? When did we know that the Devils' season had ended? We look at these questions and more in today's article which delves into what week, game, and even single shot hurt the Devils the most.

Joel Auerbach

Premise

The Devils' season is over. Well, actually everyone's season is over (tears). But, the Devil's season was over in the bad way. Well, actually 29 teams ended the bad way. But, the Devil's haven't been playing for a long time now. The reason for this stumbling first few sentences is that for some teams its easy to determine when the season ended. That 4th loss in a playoff series is pretty clearly the most important one because even a 3-0 lead isn't enough. But, for those teams that didn't make the playoffs, it's a bit tougher to find when their season was over and what got it there.

The reason that the same argument (mathematical elimination) doesn't work out for teams who don't make the playoffs is that the probability of the Buffalo Sabres winning 17 games in a row when they were a 19-38-8 team is somewhere around 0.00000008%. Not to mention the East would have then subsequently had to conspired to make it so that the last wild card spot went to a team with only 80 points. I consider that to be a negligible percentage and at that point it was indeed safe to say their season was over. So when was the Devils' season over. What was the defining period of time?

What Period of Time?

Sportsclubstats.com does a really nice job of presenting probability of postseason births. For this particular exercise I will use their graph of the season which can be found as an interactive version here

Sportsclubstatshighlight

This is a graph of the probability the Devil's will make the postseason over the course of the regular season. I am particularly interested in the highlighted region. Those two drops (losses to Florida and Tampa Bay on 3/14 and 3/15) were the largest of the whole season and they came back to back. The loss to Florida came on the heels of a nice win against Philly that had catapulted us into the driver seat so we weren't really dead after that loss. Our percentage likelihood of making it to the postseason took the largest drop of the year (just over 17%) but was still over 40. It was the second loss that was truly crippling as it brought down another 15% to the lowest likelihood since we were 4-7-5. Now that we have a nice tidy 2-game span to investigate let's see if we can get even more specific.

Finding "The Goal"

The guys over at Hockey Prospectus were kind enough to post their raw data of shots here. Along with a plethora of useful information, they include the instantaneous win probability of the team shooting and also the win probability added by each shot. Using this we find that the most game-changing goal of the two games in question was by far the goal by Scottie Upshall in the 3rd period that put Florida up 1 goal with 10 minutes left. This had a crazy 46.5% win probability added. Combine that with the 17% drop that a loss here would incur and that goal single-handedly dropped our playoff hopes by almost 8% which is absurd for a single shot. After that shot, the Panthers had over a 95% win probability and the Devils were pushed to the outside looking in with only a game against the Rangers o really give them that boost back.

Now that is mathematically the most hurtful goal in the season so that is the namesake of this article. However, I wouldn't be shocked if most of this article's readers would be critical of the fact that the Devils still had over 40% chance of making it despite that heartbreaker so the Devils season was not over. Via the WaybackMachine, here is what the standings looked like after that loss. It isn't pretty, but its still surmountable.

Moving on to the NJ versus TB game; the goal that Tampa Bay Lightning, B.J. Crombeen scored in the second period increased the win probablity by 23% to just under 80%. Basically this made it very likely that we would suffer a second loss in a row during the crunch time of the season. Now just about every other team won that day so that hurt us just as much but we still had life.

A possible 3rd nominee was Rick Nash's eventual game-winner. When the Devil's for a second time in a massively pivotal game failed to register a single goal, this proved to be the difference. A win here would have put us back in the conversation but the loss ended up putting us at our lowest percentage yet that season.

Conclusion and Your Opinions

I think Scottie Upshall delivered the biggest dagger to our playoff hopes because the stats say so and because I remember watching it and my heart dropping since we were coming off that encouraging victory versus the Flyers. However, I don't think we were done until the the 2nd goal of that rangers game that was scored when we had the goalie pulled while shorthanded.

When do you think you know if your season is over? Do you think it is ever possible to break it down to a single goal for teams like the Devils who have been out of it since late march? If not, what length of time defines a season? Do you think we dug ourselves in too deep a hole from the very beginning? Do you think there is another game that was more hurtful or that buried us more finally? Perhaps even one played by another team? Was there a transaction/injury that happened that you think ended our season? Please leave your thoughts in the comments.

(Side Note: The "Answer" to the Schneider Vs. Brodeur Article is coming next week as I needed a little more time to organize the data.)