Messing with the standings
Posted 2:01 PM ET | Comments 2
At differing points each year, I like to get out the spreadsheet and calculate how the standings might look if the NHL reweighted its points system. These adjustments must always be taken with a grain of salt -- changing the points system invariably would change team strategies, especially late in games -- but I do think they provide additional perspective that can help us evaluate and compare teams.
I made these calculations for the Western Conference only, but I can and will expand it to all teams at some point (I usually do this once by the all-star break, and once at the end of the year).
WIN-LOSS-TIE STANDINGS:
This is the system I prefer, since it is in practical terms the most comparable to the system used in playoff hockey.
To "calculate" a team's record under the W-L-T system, all regulation and overtime period wins are counted as "wins," all regulation and overtime period losses are counted as "losses," and all games that go to the shootout (either win or loss) are counted as "ties." Points percentage is the first tiebreaker. Wins is the second tiebreaker.
Team - Wins - Losses - Ties - Points
*1. Chicago, 10-7-4, 24
*2. San Jose, 10-6-2, 22
3. Detroit, 11-8-0, 22
4. St. Louis, 10-7-2, 22
*5. Dallas, 10-8-2, 22
6. Minnesota, 9-7-4, 22
7. Nashville, 10-8-1, 21
8. Phoenix, 9-7-3, 21
9. Vancouver, 10-9-1, 21
T10. Los Angeles, 9-8-3, 21
T10. Edmonton, 9-8-3, 21
12. Calgary, 8-11-0, 16
13. Colorado, 6-12-3, 15
14. Anaheim, 5-11-4, 14
15. Columbus, 5-13-2, 12
Chicago tops the leaderboard but is only fourth by points percentage, behind San Jose, Detroit and the surprising Blues.
The standings tighten considerably -- whereas the difference between first and elevnth under the current system is 6 points (3 games), the difference under W-L-T is only 3 points (1.5 games).
THREE-POINT "OLYMPIC-STYLE" STANDINGS:
This system, widely used by the IIHF in European leagues, awards 3 points for a win in regulation, 2 points for a win in overtime or the shootout, and 1 point for a loss in overtime or the shootout. I like this better than the current system, because all games are now equally weighted, but I like it less than the W-L-T system or the five-point system (which is coming up next).
Converting to the three-point system does not require that we pretend any games actually become "ties." We simply use the results already recorded and reweight to come up with the new figures.
First tiebreaker is percentage. Second tiebreaker is regulation wins. Third tiebreaker is OT/SO wins.
Team - RW - OT/SO W - OT/SO L - RL - PTS
*1. Chicago, 9-3-3-6, 36
*2. Minnesota, 8-4-3-5, 35
*3. San Jose, 9-3-1-5, 34
4. Detroit, 10-1-1-7, 33
5. Nashville, 9-1-4-5
6. Dallas, 9-3-0-8, 33
7. St. Louis, 9-1-2-7, 31
8. Phoenix, 8-2-3-6, 31
9. Edmonton, 9-1-2-8, 31
10. Los Angeles, 7-3-3-7, 30
11. Vancouver, 8-2-1-9, 29
12. Calgary, 8-0-1-10, 25
13. Colorado, 4-5-1-11, 23
14. Anaheim, 5-1-4-10, 21
15. Columbus, 4-1-2-13, 16
Some teams shift a bit in this system, too, though not as much as under the W-L-T system. Once again, Chicago leads in overall points but lags behind in points percentage, trailing San Jose, Minnesota, Detroit and St. Louis. Again, the standings tighten, with a two-game spread (six points) between first and tenth.
FIVE-POINT STANDINGS:
I have seen this system proposed by at least one hockey sabremetrician (I would credit him, but I can't remember where I saw it -- sorry!) as the fairest possible way of weighting the standings under the current rules. I tend to agree, although the system is far too complicated to ever be adopted.
Each game is worth five points. Regulation wins yield 5 points. Overtime period wins are worth 4. Shootout wins are worth 3. Shootout losses are worth 2. Overtime losses are worth 1. Regulation losses are worth zero.
Team - RW - OTW - SOW - SOL - OTL - RL - PTS
*1. Chicago, 9-1-2-2-1-6, 60
*2. Minnesota, 8-1-3-1-2-5, 57
*3. San Jose, 9-1-2-0-1-5, 56
4. Detroit, 10-1-0-0-1-7, 55
5. Dallas, 9-1-2-0-0-8, 55
6. Nashville, 9-1-0-1-3-5, 54
7. St. Louis, 9-1-0-2-0-7, 53
8. Phoenix, 8-1-1-2-1-6, 52
9. Edmonton, 9-0-1-2-0-8, 52
10. Los Angeles, 7-2-1-2-1-7, 51
11. Vancouver, 8-2-0-1-0-9, 50
12. Calgary, 8-0-0-0-1-10, 41
13. Colorado, 4-2-3-0-1-11, 38
14. Anaheim, 5-0-1-3-1-10, 35
15. Columbus, 4-1-0-2-0-13, 28
Once again, Chicago leads in points, though they are again bested by San Jose and Detroit in points percentage. The gap between first and eleventh is just two games (10 points).
RUDIMENTARY CONCLUSIONS:
-Generally speaking, teams that tend to win before extra time and have mediocre (or limited) shootout records improve their position using these adjustments. Detroit and especially St. Louis tend to look better under the adjusted standings than the "real" one.
-By contrast, teams that lose a lot in the 5-minute OT frame and/or rely on shootout prowess to accumulate wins tend to do poorly under these adjustments. Even though Chicago leads the conference in every adjustment, they also fall quite a bit back to the pack. Minnesota also loses some ground using these adjustments, especially in the W-L-T system. And Colorado is considered worse than Calgary in all three systems.
-The "extra point" is not causing standings parity. All three of these adjustments weight games equally, yet the standings actually get closer than they are in real life. Even if you wanted to make the simplest adjustment possible (count all losses, regardless of circumstances, as worth zero points), the standings would still be closer than they are today. The top 11 teams all have 10 to 12 all-purpose wins, a difference of two games (rather than three).
I made these calculations for the Western Conference only, but I can and will expand it to all teams at some point (I usually do this once by the all-star break, and once at the end of the year).
WIN-LOSS-TIE STANDINGS:
This is the system I prefer, since it is in practical terms the most comparable to the system used in playoff hockey.
To "calculate" a team's record under the W-L-T system, all regulation and overtime period wins are counted as "wins," all regulation and overtime period losses are counted as "losses," and all games that go to the shootout (either win or loss) are counted as "ties." Points percentage is the first tiebreaker. Wins is the second tiebreaker.
Team - Wins - Losses - Ties - Points
*1. Chicago, 10-7-4, 24
*2. San Jose, 10-6-2, 22
3. Detroit, 11-8-0, 22
4. St. Louis, 10-7-2, 22
*5. Dallas, 10-8-2, 22
6. Minnesota, 9-7-4, 22
7. Nashville, 10-8-1, 21
8. Phoenix, 9-7-3, 21
9. Vancouver, 10-9-1, 21
T10. Los Angeles, 9-8-3, 21
T10. Edmonton, 9-8-3, 21
12. Calgary, 8-11-0, 16
13. Colorado, 6-12-3, 15
14. Anaheim, 5-11-4, 14
15. Columbus, 5-13-2, 12
Chicago tops the leaderboard but is only fourth by points percentage, behind San Jose, Detroit and the surprising Blues.
The standings tighten considerably -- whereas the difference between first and elevnth under the current system is 6 points (3 games), the difference under W-L-T is only 3 points (1.5 games).
THREE-POINT "OLYMPIC-STYLE" STANDINGS:
This system, widely used by the IIHF in European leagues, awards 3 points for a win in regulation, 2 points for a win in overtime or the shootout, and 1 point for a loss in overtime or the shootout. I like this better than the current system, because all games are now equally weighted, but I like it less than the W-L-T system or the five-point system (which is coming up next).
Converting to the three-point system does not require that we pretend any games actually become "ties." We simply use the results already recorded and reweight to come up with the new figures.
First tiebreaker is percentage. Second tiebreaker is regulation wins. Third tiebreaker is OT/SO wins.
Team - RW - OT/SO W - OT/SO L - RL - PTS
*1. Chicago, 9-3-3-6, 36
*2. Minnesota, 8-4-3-5, 35
*3. San Jose, 9-3-1-5, 34
4. Detroit, 10-1-1-7, 33
5. Nashville, 9-1-4-5
6. Dallas, 9-3-0-8, 33
7. St. Louis, 9-1-2-7, 31
8. Phoenix, 8-2-3-6, 31
9. Edmonton, 9-1-2-8, 31
10. Los Angeles, 7-3-3-7, 30
11. Vancouver, 8-2-1-9, 29
12. Calgary, 8-0-1-10, 25
13. Colorado, 4-5-1-11, 23
14. Anaheim, 5-1-4-10, 21
15. Columbus, 4-1-2-13, 16
Some teams shift a bit in this system, too, though not as much as under the W-L-T system. Once again, Chicago leads in overall points but lags behind in points percentage, trailing San Jose, Minnesota, Detroit and St. Louis. Again, the standings tighten, with a two-game spread (six points) between first and tenth.
FIVE-POINT STANDINGS:
I have seen this system proposed by at least one hockey sabremetrician (I would credit him, but I can't remember where I saw it -- sorry!) as the fairest possible way of weighting the standings under the current rules. I tend to agree, although the system is far too complicated to ever be adopted.
Each game is worth five points. Regulation wins yield 5 points. Overtime period wins are worth 4. Shootout wins are worth 3. Shootout losses are worth 2. Overtime losses are worth 1. Regulation losses are worth zero.
Team - RW - OTW - SOW - SOL - OTL - RL - PTS
*1. Chicago, 9-1-2-2-1-6, 60
*2. Minnesota, 8-1-3-1-2-5, 57
*3. San Jose, 9-1-2-0-1-5, 56
4. Detroit, 10-1-0-0-1-7, 55
5. Dallas, 9-1-2-0-0-8, 55
6. Nashville, 9-1-0-1-3-5, 54
7. St. Louis, 9-1-0-2-0-7, 53
8. Phoenix, 8-1-1-2-1-6, 52
9. Edmonton, 9-0-1-2-0-8, 52
10. Los Angeles, 7-2-1-2-1-7, 51
11. Vancouver, 8-2-0-1-0-9, 50
12. Calgary, 8-0-0-0-1-10, 41
13. Colorado, 4-2-3-0-1-11, 38
14. Anaheim, 5-0-1-3-1-10, 35
15. Columbus, 4-1-0-2-0-13, 28
Once again, Chicago leads in points, though they are again bested by San Jose and Detroit in points percentage. The gap between first and eleventh is just two games (10 points).
RUDIMENTARY CONCLUSIONS:
-Generally speaking, teams that tend to win before extra time and have mediocre (or limited) shootout records improve their position using these adjustments. Detroit and especially St. Louis tend to look better under the adjusted standings than the "real" one.
-By contrast, teams that lose a lot in the 5-minute OT frame and/or rely on shootout prowess to accumulate wins tend to do poorly under these adjustments. Even though Chicago leads the conference in every adjustment, they also fall quite a bit back to the pack. Minnesota also loses some ground using these adjustments, especially in the W-L-T system. And Colorado is considered worse than Calgary in all three systems.
-The "extra point" is not causing standings parity. All three of these adjustments weight games equally, yet the standings actually get closer than they are in real life. Even if you wanted to make the simplest adjustment possible (count all losses, regardless of circumstances, as worth zero points), the standings would still be closer than they are today. The top 11 teams all have 10 to 12 all-purpose wins, a difference of two games (rather than three).
then whats the point of the shootout?
The point is that there wouldn't be any shootouts. When you adjust to W-L-T, any games that had/would have gone to the shootout are simply counted as if the ended as a draw, with the teams getting one point apiece.