Does the NHL Trade Deadline Matter?
And if it does, by how much?
Today’s NHL trade deadline — at 3 p.m. ET — marks the last meaningful moment for teams to make big improvements before the playoffs, and we’ve already seen a few blockbusters involving the best names available this week. (I wrote about why the defending champ Vegas Golden Knights had to go for it at the deadline yesterday.)
But how much does the deadline really matter? Can it truly make or break a team’s future chances as much as we tend to think?
To answer that question, I went back and looked at players who were acquired in the middle of a season and played fewer than 45% of a team’s games.1 For each team, I calculated their net full-season adjusted Goals Above Replacement (GAR) from players added minus subtracted, as a proxy for total talent added at the deadline, and I also calculated their Elo power rating as of the deadline.
The first component of our study will be to examine how a big deadline haul affects a team’s performance down the stretch. Here’s a plot of net GAR talent added versus how a team’s Elo changed between the deadline and the end of the regular season, going back to the first NHL trade deadline in 1980:
As you can see, there isn’t a whole lot in the way of a predictive relationship there. Some teams, such as the 1996 Mighty Ducks (who went and got future franchise icon Teemu Selänne at the deadline from Winnipeg), loaded up and it helped them play much better down the stretch — even if, in Anaheim’s case, they still missed the playoffs. But just as many others made additions that looked good on paper, only to backslide in their performance.
(Still others, like the 2006 Oilers, took a big deadline swing, played poorly down the stretch, then caught fire in the playoffs anyway. Hockey is a very weird game like that sometimes.)
Another way we can measure the impact of a strong deadline is to look at the relative value of deadline talent added in predicting postseason success, conditional on making the playoffs. For that purpose, I ran a regression against total playoff wins,2 using a team’s Elo rating on Deadline Day and its net deadline GAR added as the independent variables.
In our regression, it would take more than 20 net GAR of talent added at the deadline to improve your expected playoff performance by a single win. (For reference’s sake, only 13 of the 688 playoff teams in our sample had that kind of deadline haul.) According to the same analysis, a team’s pre-deadline Elo form was about 15.5 times as important as its deadline talent added in determining how well it did in the playoffs.
Based on that, it seems like the trade deadline might be a bit overhyped. And it makes sense: Hockey is difficult to forecast over the largest of samples, and at the deadline we’re asking ourselves to predict how additional players will coalesce within a new team over the final 25% of the regular season and a sequence of best-of-seven playoff series. When you think about it that way, of course there is a weak relationship between talent added at the deadline and future performance.
However, if you’re looking for meaning in today’s moves, I do have some good news. In the same data, the deadline does appear to be gaining relative importance during the salary cap era. Since 2005-06, a team’s pre-deadline Elo is only 7.4 times as important as its deadline GAR added in terms of predicting playoff wins. That still means it’s tough to say whether a great deadline will manifest in a Stanley Cup run or not — but it seems to be more likely now than it was before the salary cap came into play.
Filed under: NHL
This is based on the fact that the deadline usually happens with 25-30% of the schedule remaining, but I also gave an extra grace period before then to account for deals that happen a few weeks earlier than Deadline Day (think Elias Lindholm to Vancouver this season). One other note — for this calculation, I also accounted for starting goalies not playing every game, so their threshold to be included was raised — lest we include netminders who were acquired far before the deadline, but only started every other game for their new team.
Considering only the Round of 16 and onward in the 2020 bubble playoffs.