The Devils are the model's lean over the Kings — with PDO regression in play

The model leans toward New Jersey, and the case rests on how the Devils have played more than on what the standings show.

Start with expected goals, a measure that weights each shot by how likely it was to score. At five-on-five, the Devils own 58.9% of that share, against 51.1% for the Kings. Shot attempts tell the same story, and New Jersey holds 59.6% of them. The Kings sit at 49.5%, so expect New Jersey to spend more of the night in the Kings' end.

PDO, the sum of shooting and save percentage, is where the model sees luck. The Devils sit at 0.973, which the model reads as running cold and due for positive regression. If the bounces turn, that is a tailwind for New Jersey. The Kings' 0.981 carries the same flag.

Recent results lean the other way. The Devils are the hotter side by points pace, going 6-4-0 over their last 10. The Kings are underwater, with a goal differential of -9 across eight games.

In goal, D. Kuemper has let in more goals than his workload should have allowed, sitting at -2.8 goals saved above expected over six games. N. Daws has played four games and sits at -1.3 goals saved above expected.

The projected score is LAK 3.0 - 3.3 NJD, a game the model expects to be close. New Jersey's chance is 56%, which makes it a lean rather than a lock.

Data: NHL API, Backcheck expected goals model