So, my beloved Kings, you're doing better than last year, but, ".500" is not the benchmark you seek.
It took 91 points to make the playoffs last year (Nashville). 91/82 is about 9 points from 8 games, or 4-3-1. Nashville had 42 wins in their 82 games, i.e., winning slightly more than half. A nominal .500 team is taking no more than 8 points from 8 games, and, as such, is never going to make the playoffs. This is an artifact of the absurdity of the current points system comprising both 2- and 3-point games, as a league of 30 teams in which 16 make the playoffs should mean that those last couple teams making it in are around .500, unless there is a significant outlier or two at the bottom driving the median points number up.
Simply put, .500 meaning "same number of W's and L's" is meaningless; if .500 meant "fraction of games won," it would be meaningful. Announcers across the league (and consequently, fans) are describing their teams as being some number of games above or below .500 as if .500 were a useful cutoff line corresponding to eventual playoff qualification.
If you need nine points from eight games, that's a little over 1.1 ppg. Taking one pity point for the OT/SO loss is leaving you behind the pace--not by much, but behind the pace. An 0-0-82 team would never make the playoffs. Nor would a team that's 41-41-0. In fact, none of the possible records involving a 1ppg average would ever qualify a team for the second season.
For .500 to mean something, it should be used properly: W/GP. Pity points will be amassed randomly (or consistently if you have Mathieu Garon), and will help...but not if you're not winning half your games in the first place.
Subscribe to:
Post Comments (Atom)
Posts
Since you asked me to, here was my response on LGK:
ReplyDeleteI suppose in your estimation a team that goes 20-0-62 for 102 points is a 20/82 or approximately .250 team? Those 102 points wouldn't get them in the playoffs?
I think the proper formula is Points/Expectation(Points Allotted Per Game Leaguewide) must be over .500 to make roughly the top "half" of the league as an "above average" record...(it seems our common assumption is that 16/30 make the playoffs, so a legitimate .500 record - one in which you acquire half the points "available" - should make it on average).
For instance, if one out of 5 games goes into OT, then expectation of points per game is .8*2+.2*3 = 2.2 points per game, to be truly ".500" you need 1.1 point per game average.
Excellent. As a followup, I checked the standings, and so far, 166/747 games have gone to OT, for a 22.2% rate. This projects to 273/1320 OT games for the whole season, or an average of 9.1 OTL per team. Since there's no reason to expect otherwise, and there's no such thing as a fractional OTL, the probability of finishing an 82-game season with a certain number of OTL should follow a Poisson distribution:
ReplyDelete# of OTL - probability
0 0.0%
1 0.1%
2 0.5%
3 1.4%
4 3.2%
5 5.8%
6 8.8%
7 11.4%
8 13.0%
9 13.2%
10 12.0%
11 9.9%
12 7.5%
13 5.3%
14 3.4%
15 2.1%
16 1.2%
17 0.6%
18 0.3%
19 0.2%
20 0.1%
21 0.0%
So it's most likely that a team would finish with 9 OTL, but every 30 years or so we'd expect to see a team with as few as 1 or as many as 20. Now, in an average season, we'd expect to see 9 as the mode OTL in the final regular season standings, but even that is only going to show up 4, maybe 5 times. Extending the above table to include the # of teams we'd expect with each # of likely OTL, as well as the number of wins each team would need to achieve 91 points...
OTL - Prob - # of teams - W required for 91pts
0 0.0% 0 46
1 0.1% 0 45
2 0.5% 0 45
3 1.4% 0 44
4 3.2% 1 44
5 5.8% 2 43
6 8.8% 3 43
7 11.4% 3 42
8 13.0% 4 42
9 13.2% 4 41
10 12.0% 4 41
11 9.9% 3 40
12 7.5% 2 40
13 5.3% 2 39
14 3.4% 1 39
15 2.1% 1 38
16 1.2% 0 38
17 0.6% 0 37
18 0.3% 0 37
19 0.2% 0 36
20 0.1% 0 36
21 0.0% 0 35
So your average 9 OTL team is still going to need 41 W to make the playoffs, or, 0.500 W/GP. Projecting this further to show these teams' records if they were to reach 91 (or 92) points:
[W - L - OT - Pts]
W L OTL PTS
46 36 0 92
45 36 1 91
45 35 2 92
44 35 3 91
44 34 4 92
43 34 5 91
43 33 6 92
42 33 7 91
42 32 8 92
41 32 9 91
41 31 10 92
40 31 11 91
40 30 12 92
39 30 13 91
39 29 14 92
38 29 15 91
38 28 16 92
37 28 17 91
37 27 18 92
36 27 19 91
36 26 20 92
35 26 21 91
So...it's pretty clear that no matter how many OTL you have, you're going to need to be at least +9 in the wins vs. losses department to make the second season. So, W=L is bad, no matter what happens...and teams with an above average number of OTL can still make it as long as they win 9 more than they lose in regulation.