## Incorrect things you will hear: “In a 48-game season, every game counts”

After 28 games in the '95 season, the Oilers were the only team in a playoff position to drop out.

By now you’ve already seen some sort of variation on what the NHL standings would have looked like after 48 games last season. The way I have it calculated, there would have been no changes in the Eastern Conference as far as teams making the playoffs. There would have been some qualms with seeding, but the same eight teams would have reached. In the Western Conference, the one difference would have been Minnesota replacing Phoenix for a spot.

Again, seedings change, but it doesn’t make too big of a difference anymore. Teams are probably closer together in talent than they were six or seven years ago, and in each of the last two years, the 8th seed in the Western Conference has been touted as “not your traditional 8 seed”.

The Stanley Cup winning team and players won’t also have asterisks engraved next to their names, although there will be some form of debate, because of the shortened season. It’s funny because for the most part, playoff spots are won and lost after about 45-50 games. The remainder is about seeding, and hope. The single point for a loss in overtime or a shootout was carefully crafted to keep playoff races tight and down to the wire.

So, since it will be so close this year, every game matters, right? Well…

Remember, regression happens. Teams who dominated in the first 48 games of last season scaled back slowly in the second half. Teams that struggled in the first 48 games improved. This can be due to a variety of things. Regression, luck, adjustments, you name it, but the point of the matter is that it happens. Check out the cool graph I made. I don’t know how to work Excel all that well, but the y-axis is points over 82 games:

Now, the playoff teams are still better than the non-playoff teams, just not by as much. My working theory here is that you’re going to get a wider range of outcomes after 48 games. Things fit closer into place after 82, but a particularly strong first-half team (think Boston last year) won’t have to face an entire spring run to mask their teams’ weaknesses. The spread between division winners and losers, pro-rated over 82 games, is going to be a much larger gap than last season.

I think an “every game matters” schtick won’t particularly work, and here’s why. This is a topic covered in depth before by Tyler Dellow, but I wanted to bring a real-life example to the table, the 48-game 1995 season:

28 games is to a 48-game season as 48 games is to an 82-game season. We have one recent 48-game season to look, the lockout-shortened 1995 campaign. Again, I downloaded the standings from that season, and went through every teams’ record and tallied up their record in the first 28 games of the season. Hockey Reference has an excellent feature where they put an asterisk next to each team’s name. Here are, or were, the standings:

(My, it was a fantastic season for the Print Area ordiques, except for the move)

You’ll notice just a single change, a single team pulled itself up from the depths for the last 20 games to make the post-season. That was the San Jose Sharks, who replaced the Edmonton Oilers. In fact, the first five seeds in the East from this particular year are identical: Quebec, Philadelphia, Pittsburgh, Boston and New Jersey. The Flyers take the 2nd seed by virtue of being ahead in the newly-renamed Atlantic Division. Pittsburgh was still in the Northeast at this time.

In the West, it’s something similar. In actuality, Chicago and St. Louis flip-flopped, while Vancouver, Dallas and San Jose all overtook Edmonton. That was it. Those were the changes from Games 29 to 48 back in the 1995 season. The teams who weren’t in playoff position at that point, averaging 65 points per 82 games through the first 28, averaged 67 through the remainder, a much less noticeable difference than in the longer season. Teams in playoff position stayed unchanged at 92 points per 82 both before in the first 28 and last 20.

Basically, I’d have to be convinced pretty hard at this point that anything after about mid-March will really matter. That’s about when the playoff picture should be easy enough to suss out, barring a team doing something other than what the 1995 Oilers did, which was go 5-14-1 over their last 20 games to miss the playoffs by a whoppingâ€¦ four points.