The Washington Capitals haven’t had the best start to the season. Going 3-3-0 – and more importantly how they’ve gone 3-3-0 – has got some statistical models feeling a little iffy about the team.

One particularly bearish model at Moneypuck thinks the Caps have just a 38 percent chance of making the postseason right now. Dom Luszczyszyn’s model at the Athletic has them at 51 percent, which is just an inch above tonight’s opponent, the New Jersey Devils.

So, yeah, we’re getting meaningful hockey in October.

Here’s an absurd closeup of Dom’s playoff probability graph; the line indicates how the chances have changed recently.

The Devils are just a fraction behind the Caps, which means tonight’s game will have an outsize impact on their chances going forward. I should mention that Micah McCurdy’s model at HockeyViz appears to like Washington’s long-term chances better than the other two I mentioned above.

One of my favorite features of HockeyViz (hashtag not sponsored) is showing you the implications of any single game’s outcome. Here is that visualization for tonight:

If Washington wins in rego, that’s a ten-point swing in Washington’s fortunes relative to New Jersey. If New Jersey wins in rego, the opposite is true.

This is all a long way of saying that New Jersey could be a much better team that we thought in the preseason, and therefore they could be a much bigger threat. Without making sucker conclusions based on early-season percentages, we can at least acknowledge that New Jersey has totally dominated play during even strength. No team has owned a larger share of shot attempts or expected goals (using Natural Stat Trick’s reckoning) than the Devils have… so far. There’s no way the team can remain at such haughty heights, and for that reason I’m not even including the numbers here. Instead I think we can just acknowledge that they could be good, and maybe even pretty darn good.

If not for their goaltending, at least. The team is saving .863 during five-on-five play, which is 31st in the league. Bad goaltending (mostly from Mackenzie Blackwood, plus one stinker from Vitek Vanecek) and random chance are the big contributors to Jersey’s two regulation losses – and the only reasons why everyone isn’t raving about this upstart team, led by a player the team didn’t seem to even want, playing way better than we thought they would at this point in their cycle.

Tonight’s game should be interesting.

Thanks to CJ Turturo, my fellow Bratt-head, for seeding this idea. This story would not be possible without Natural Stat Trick and HockeyViz. Please consider joining us in supporting those sites.