An Interesting Theory

[Scout]

with Gregg Drinnan

Thursday, November 13, 2008

An interesting theory on the circus. . .

The hockey game is in overtime. A defenceman has the puck in the left corner of his zone. In trying to clear the zone, he throws the puck into the middle of the ice. An opposing forward corrals it, skates in and scores the winning goal.
In the dressing room after the game, the head coach of the losing team berates the defenceman for what was a critical error.
“What’s the big deal, coach?” responds the player. “We still got a point.”
Yes, indeed, the team lost the game in overtime but still was rewarded with a point.
And therein lies the rub – of the ‘major’ sports, hockey is the only one that rewards failure.
--------
Obviously, this has bastardized hockey statistics to the point of ridiculousness. For example, a player scores the game-winning goal in a shootout but statistics don’t credit him with a goal. A goaltender stops 10 shooters in the shootout but his save percentage doesn’t reflect it.
The standings show one team with a 6-9-0-5 record. But hasn’t that team really won six games and lost 14?
Never mind that the shootout takes what coaches preach is a “team game” and turns it into an individual sport.
And we won’t even get into coaching records. It used to be that a head coach’s record was measured in wins, losses and ties. These days, if a coach has been around long enough, you need five categories – wins, losses, ties, overtime losses and shootout losses – to chart his career. And how dumb is that? Go ahead, you try and figure out Don Hay’s actual winning percentage as a WHL head coach.
Geez, remember when a loss was a loss was a loss?
---
Bill Motiuk is a hockey fan. And like a lot of us – hello there, Dan Russell – he despises the present system of awarding two points for a regulation-time victory and three points – two to the winner, one to the loser – of anything that goes beyond that. Loser points really are the scourge of hockey, aren't they?
So Motiuk put on his thinking cap and came up with a system that he says is “based on the business principle that the longer it takes you to get the job done, the more it will cost you.”
When I first looked at his idea, I thought, ‘Ah, what’s the big deal?’ But the more I thought about it and more I re-read it and the more I absorbed it, the more it grew on me. And now I’m sold.
But in order to be sold on this, you have to go along with this premise: Some games are worth three points, some are worth two and some are worth one. At first, I wasn’t enthused about that part, but the more I thought about it, well . . .
To begin with, this system puts a three-point value on a regulation-time victory. The winner gets three points; the loser doesn’t get a thing. “To get the three points,” Motiuk says, “you must win in regulation time.”
OK, but what if the game goes into overtime?
An overtime victory is worth two points. Again, the loser doesn’t get anything. “Being tied at the end of regulation deserves nothing,” he explains, “because it wasn’t a tie game and overtime only signified that the end of the first stage did not produce a winner. The winner in overtime only gets two points (as opposed to three) because he didn’t get the job done in regulation time.”
OK, but what if the game goes to a shootout?
The winner of the shootout would receive one point. The loser, again, wouldn’t get a thing.
“The eventual winner ends up forfeiting two points he could have obtained, again for not getting the job done in regulation time,” Motiuk reasons. “At the end of the night, the loser gets nothing because we should not be rewarding losers and there are no longer any tie games under the current system.”
The CFL has long gotten raked over the coals in some corners for awarding a single point on some missed field-goal attempts. But the NHL, the WHL and many other leagues also reward failure by giving points to teams that lose in overtime or a shootout.
“We only have tie segments,” Motiuk points out. “We do not give points for ties at the end of the first and second periods. So why give a point for a tie at the end of the third period? Why reward a team with a point when they lose in overtime? Ditto in the shootout.”
And that, folks, is my favourite part of all this and the thing that really validates it. If you are going to play overtime and if you are going to have shootouts, why give anything to teams for being tied at the end of the third period? As Motiuk points out, teams don't get anything for being tied at the end of the first and second periods.
The way he has it figured, this “would change the whole dynamic of the game.”
“The intensity of the game would be predicated by the three points available before the end of the third period,” he says.
Under this system, should teams be tied near the end of the third period they would be pulling out all stops in an effort to score. After all, score and win and you get three points; go to overtime and you will get two, one or zero points.
Hey, from where I sit, it makes sense. And it certainly is better than the system now in place. Which is why it doesn’t stand a snowball’s chance.


Scout

[Bocephus]

I like it!

[Scout]

Alan Caldwell

Thursday, November 13, 2008

Running with Gregg's idea

....or more accurately, Bill Motiuk's idea.

What I've done is re-worked the current WHL standings to use Motiuk's proposed system, just to see how things would be different. First, the current standings the way the WHL does it now:

(I've ranked the teams by win% in order to eliminate variance in games played)



GP W L OTL SOL PTS Win %
1 Vancouver Giants 19 15 1 0 3 33 0.868
2 Calgary Hitmen 24 19 4 1 0 39 0.812
3 Spokane Chiefs 20 13 4 0 3 29 0.725
4 Tri-City Americans 21 14 5 0 2 30 0.714
5 Saskatoon Blades 22 15 6 1 0 31 0.705
6 Swift Current Broncos 22 14 8 0 0 28 0.636
7
Brandon Wheat Kings 20 12 7 1 0 25 0.625
7 Regina Pats 24 13 7 1 3 30 0.625
9 Medicine Hat Tigers 22 11 7 2 2 26 0.591
10 Kelowna Rockets 21 12 9 0 0 24 0.571
11 Everett Silvertips 21 10 8 3 0 23 0.548
12 Kootenay Ice 24 10 9 2 3 25 0.521
13 Prince Albert Raiders 25 13 12 0 0 26 0.520
14 Kamloops Blazers 23 10 10 0 3 23 0.500
15 Lethbridge Hurricanes 20 9 10 0 1 19 0.475
16 Edmonton Oil Kings 24 9 12 1 2 21 0.438
17 Prince George Cougars 22 9 12 0 1 19 0.432
18 Chilliwack Bruins 21 7 11 1 2 17 0.405
18 Seattle Thunderbirds 21 7 11 1 2 17 0.405
20 Red Deer Rebels 25 7 13 0 5 19 0.380
21 Moose Jaw Warriors 20 6 13 0 1 13 0.325
22 Portland Winter Hawks 19 5 14 0 0 10 0.263

And now, the way the standings would look using Motiuk's idea:



GP W OTW SOW L PTS Win % Shift











1 Vancouver Giants 19 15 0 0 4 45 0.789 0
2 Calgary Hitmen 24 17 0 2 5 53 0.736 0
3 Saskatoon Blades 22 12 2 1 7 41 0.621 2
4 Spokane Chiefs 20 12 0 1 7 37 0.617 -1
5 Swift Current Broncos 22 12 1 1 8 39 0.591 1
6 Kelowna Rockets 21 12 0 0 9 36 0.571 4
7 Tri-City Americans 21 9 2 3 7 34 0.540 -3
8 Brandon Wheat Kings 20 10 0 2 8 32 0.533 -1

9 Regina Pats 24 11 0 2 11 35 0.486 -2
10 Medicine Hat Tigers 22 10 0 1 11 31 0.470 -1
11 Everett Silvertips 21 8 0 2 11 26 0.413 0
12 Prince Albert Raiders 25 7 3 3 12 30 0.400 1
12 Lethbridge Hurricanes 20 7 1 1 11 24 0.400 2
14 Kamloops Blazers 23 8 1 1 13 27 0.391 0
15 Prince George Cougars 22 7 1 1 13 24 0.364 2
16 Edmonton Oil Kings 24 7 1 1 15 24 0.333 0
16 Kootenay Ice 24 7 0 3 14 24 0.333 -4
18 Seattle Thunderbirds 21 5 0 2 14 17 0.270 0

19 Moose Jaw Warriors 20 5 0 1 14 16 0.267 2
20 Portland Winter Hawks 19 4 1 0 14 14 0.246 2
21 Chilliwack Bruins 21 4 0 3 14 15 0.238 -3
22 Red Deer Rebels 25 4 1 2 18 16 0.213 -2

The last column, "Shift", is the number of places up or down in the overall standings that the team would move from their current position if Motiuk's system was used. So the biggest beneficiary of this system would be Kelowna, who are 4 positions better using this method, while the biggest drop is Kootenay, who drop four places.

A quick explanation about the winning % calculation too since you can see the winning percentage is different in the second method. I have calculated winning % in Motiuk's method as a "points percentage", the same as it is in the current standings. It's not a reflection on the number of games a team actually wins, but rather the percentage of possible points they actually get. Each game is worth three points maximum so a team that has played 20 games could have a maximum of 60 points. Thus, they would need 30 points to be a .500 team and so on.

I'm off to recalculate last year's final standings now using this method. I'll post those results later this afternoon when I'm done.

UPDATE: I've redone last year's final standings using Motiuk's idea...

Last year's final standings the way the WHL counts it up:



GP W L OTL SOL PTS Win %
1 Tri-City Americans 72 52 16 2 2 108 0.750
2 Spokane Chiefs 72 50 15 1 6 107 0.743
3 Vancouver Giants 72 49 15 2 6 106 0.736
4 Calgary Hitmen 72 47 20 1 4 99 0.688
5 Lethbridge Hurricanes 72 45 21 2 4 96 0.667
6 Regina Pats 72 44 22 4 2 94 0.653
7 Medicine Hat Tigers 72 43 22 5 2 93 0.646
8 Kootenay Ice 72 42 22 5 3 92 0.639
9 Seattle Thunderbirds 72 42 23 5 2 91 0.632
10 Brandon Wheat Kings 72 42 24 3 3 90 0.625
11 Swift Current Broncos 72 41 24 1 6 89 0.618
12 Moose Jaw Warriors 72 37 21 6 8 88 0.611
13 Kelowna Rockets 72 38 26 2 6 84 0.583
14 Everett Silvertips 72 39 30 0 3 81 0.562
15 Saskatoon Blades 72 29 34 3 6 67 0.465
16 Chilliwack Bruins 72 28 35 4 5 65 0.451
17 Kamloops Blazers 72 27 41 2 2 58 0.403
18 Prince Albert Raiders 72 26 41 3 2 57 0.396
19 Edmonton Oil Kings 72 22 39 4 7 55 0.382
20 Prince George Cougars 72 20 48 1 3 44 0.306
21 Red Deer Rebels 72 18 47 4 3 43 0.299
22 Portland Winter Hawks 72 11 58 2 1 25 0.174

And the final standings they way they would be using Motiuk's system:



GP W OTW SOW L PTS Win % Shift
1 Vancouver Giants 72 44 3 2 23 140 0.648 2
2 Spokane Chiefs 72 44 1 5 22 139 0.644 0
3
Tri-City Americans 72 41 5 6 20 139 0.644 -2
4 Calgary Hitmen 72 41 3 3 25 132 0.611 0
5 Lethbridge Hurricanes 72 38 2 5 27 123 0.569 0
6 Seattle Thunderbirds 72 39 1 2 30 121 0.560 3
7 Medicine Hat Tigers 72 35 5 3 29 118 0.546 0
8 Brandon Wheat Kings 72 37 1 4 30 117 0.542 2
9 Regina Pats 72 32 5 7 28 113 0.523 -3
10 Kootenay Ice 72 31 6 5 30 110 0.509 -2
11 Swift Current Broncos 72 30 7 4 31 108 0.500 0
12 Everett Silvertips 72 29 6 4 33 103 0.477 2
13 Kelowna Rockets 72 29 4 5 34 100 0.463 0
14 Moose Jaw Warriors 72 28 4 5 35 97 0.449 -2
15 Saskatoon Blades 72 23 3 3 43 78 0.361 0
16 Chilliwack Bruins 72 23 1 4 44 75 0.347 0
17 Kamloops Blazers 72 22 1 4 45 72 0.333 0
17 Prince Albert Raiders 72 22 2 2 46 72 0.333 1
19 Edmonton Oil Kings 72 18 0 4 50 58 0.269 0
20 Prince George Cougars 72 17 1 2 52 55 0.255 0
21 Red Deer Rebels 72 12 1 5 54 43 0.199 0
22 Portland Winter Hawks 72 9 0 2 61 29 0.134 0

Seattle would gain the most, being three positions higher in the standings. Regina would lose the most, being three positions lower. Tri-City would not have been first overall, they would have been third and Vancouver would have been #1.

But would any of the playoff matchups last year have been affected? As it was we had:

Calgary vs. Moose Jaw
Regina vs. Swift Current
Lethbridge vs. Brandon
Medicine Hat vs. Kootenay
Tri-City vs. Kamloops
Vancouver vs. Chilliwack
Spokane vs. Everett
Seattle vs. Kelowna

With the alternate system, we would have seen:

Calgary vs. Moose Jaw
Brandon vs. Swift Current
Lethbridge vs. Kootenay
Medicine Hat vs. Regina
Vancouver vs. Kamloops
Spokane vs. Chilliwack
Tri-City vs. Kelowna
Seattle vs. Everett

So the matchups would have been completely different. Only #1 vs. #8 in the Eastern conference, Calgary vs Moose Jaw, would have remained the same. Most notable is that not only would Regina not have been the #2 seed, they wouldn't even have had home ice advantage in the first round.

An interesting proposal, for sure. Thoughts?


Scout

[HURRICANE'S ROCK]

I think you simplifiy it. 3 points total per every game. Win in regulation = 3pts for winner and zero pts for the loser. Win in OT or Shootout and winner gets 2, loser gets 1. This way you at least get rewarded for winning in regulation. Motiuk's way is goofy in that a "winner" can get as little as one point. At least my way a team that wins in regulation can go on a good point run. Win 5 or so games and you get 15 points and can move up on those teams that win in OT or SO.
The way it is now is plain wrong. Lets' say you are chasing two teams who are a few points ahead of you. They keep winning or losing in extra time, you win in regulation and they keep pulling ahead.They get rewarded with 3 total points between them and you only get 2. There has to be a reward for winning in regulation time.

[rinkrat]

How about two points for a win, no ties, play ot until there is a winner.
Simple, why complicate things? Oh and you can put the shoot out in the all star game (if we had one).

[Scout]

with Gregg Drinnan

Friday, November 14, 2008

The theory put forth by Bill Motiuk and expounded upon in this space on Thursday has prompted some response.
For starters, Dan Russell, who is the host of SportsTalk, the Vancouver open mouth show (it is heard nightly on CKNW in Vancouver and numerous stations throughout B.C.) that is the longest-running show of its kind in Canada, bandied it about a bit Thursday night. Like me, he wasn’t enthused about it upon first reading. But the more he thought about it, the more he liked it. And by the end he was positively enamoured of it.
Now here’s a note from Garth MacBeth, who supplies this blog with the MacBeth Report on a regular basis:
“I liked your bit on points - I've always contended that a loss is a loss is a loss. Period. Does nobody remember when the WHL (and the WHA) played 10 minutes of 5-on-5 OT?
“At least this is somewhat addressed in Europe (and in IIHF tournaments). Three points for a win in regulation, two points for a win in OT or shootout, one point for a loss in OT or shootout, no points for a regulation loss. Everyone knows that there are three points awarded in every game.
“In the WHL and NHL, some games have two points awarded (all regulation games) while all OT/shootout games have three points awarded. As a result, mathematically, .500 is different for every team. A team that wins in regulation gets 100 per cent of the points awarded for the game, but if they win in OT/shootout, they get only 67 per cent of the points available.
“However, there is no way a team could get 100 per cent of the points in an OT/shootout.
“I think that it should be two points for a win and no points for a loss, no matter when the game finally ends.
“(Going into this weekend), Kamloops has 23 points in 23 games, so the WHL says that they have a .500 winning percentage. Their record is 10-10-0-3 with one OT win and a 1-3 record in the shootout. So they have played in five games where three points were awarded and 18 games where two points were awarded. Thus, there were 51 points awarded in their games so far. So really, they are playing at .451 (23/51). Silly . . .
“Another way to look at the absurdity of it all: you could lose every game in OT/shootout and still be at .500. Think about that. You could lose every game in OT/shootout and still have 72 points at the end of the season. Seventy-two (72) points in the Western Conference would have put you in seventh place last season.
“Of course, you still could miss the playoffs because you would give an extra point to each opponent but it could be possible to make the playoffs with no wins.”
MacBeth, with some time on his hands, decided to take this a little bit further.
“I took a look at what would have happened last season if a team went 0-0-72-0 or some combination of 72 OT and shootout losses in the WHL.
“I picked the Edmonton Oil Kings because they had the worst OT/shootout record in the league: 0-4 in OT, 4-7 in the shootout. Edmonton finished 22-39-4-7 for 55 points and 11th place in the 12-team Eastern Conference. Had it lost every game in OT/shootout, it would have finished with 72 points. If you add the extra points to the records of the teams the Oil Kings beat last season, they would have finished in ninth place, one point ahead of the Saskatoon Blades.
Original standings:
1 – Calgary, 99
2 – Regina, 94
3 – Lethbridge, 96
4 – Medicine Hat, 93
5 – Kootenay, 92
6 – Brandon, 90
7 – Swift Current, 89
8 – Moose Jaw, 88
9 – Saskatoon, 67
10 – Prince Albert, 57
11 – Edmonton, 55
12 – Red Deer, 43
---
If Edmonton had gone 0-0-72-0:
1 – Calgary, 103 (no change)
2 – Regina, 94 (no change)
3 – Lethbridge 99 (no change)
4 – Kootenay, 96 (+1)
5 – Medicine Hat, 93 (-1)
6 – Moose Jaw, 92 (+2)
7 – Swift Current, 91 (no change)
8 – Brandon, 91 (-2)
9 – Edmonton, 72 (+2)
10 – Saskatoon, 71 (-1)
11 – Prince Albert, 61 (-1)
12 – Red Deer, 47 (no change)
---
"There would have been no change in position of the Western Conference teams, although Kamloops, Portland, and Prince George would have picked up two points each and Everett one.
---
"So . . . while it didn’t affect the Oil Kings’ playoff run, they finished two places higher without ever winning a game. Granted the chance of this happening is pretty slim but . . .
"Overall," MacBeth concludes, "it's probably a moot point but the possibility is there.”
And the fact the possibility is there proves the absurdity of it all and should be enough to force a change.
---
For more on all of this, zip on over to Alan Caldwell’s blog (Small Thoughts at Large) and check out the breakdown he did Thursday.


Scout

[HURRICANE'S ROCK]

with Gregg Drinnan

Friday, November 14, 2008

“At least this is somewhat addressed in Europe (and in IIHF tournaments). Three points for a win in regulation, two points for a win in OT or shootout, one point for a loss in OT or shootout, no points for a regulation loss.

Scout

Like I said earlier, this makes the most sense. 3 points for every game. Win in regulation and you get all 3 points. If you rework the standings from last year this way, I bet Edmonton does not move up in the standings because the teams they passed would have got 3 points, not 2 for a win.

[sbtatter]

We should have a change, to this system or 3 points for a regular time win, 2 points for OT/SO winner and 1 point for OT/SO losser

[Allin44]

But the point of this wasnt to improve statistics it was to increase the chances would take in OT. If you introduce no points for an OT loss it just causes teams to become more conservative knowing its 2 pts or 0

[TwoBits]

Why must everything be so complicated? I'd like to know what was wrong with the original Win 2points Tie split 1 -1 point Lose 0points?

[scrunt]

Why must everything be so complicated? I'd like to know what was wrong with the original Win 2points Tie split 1 -1 point Lose 0points?

Hear hear! Nothing wrong with a hard fought tie, both teams split the points on the line.

I think everyone has gotten used to overtime, so that could be kept, but keep it 5 on 5 and give 2 points for a win, 0 for a loss. (while 4 on 4 increases the chance someone scores, I don't like the idea that a team that earns a tie in regulation could lose that point under a different format).

If the Bettman Plan - no ties! - cannot be given the death it deserves, the next best solution is for three points to be awarded in every game, 3-0 regulation, 2-1 OT or S/O. There is NO WAY the number of points awarded for a game should be contingent upon the specific outcome of that game. Otherwise, why not open it wide open and award the winner 2 points plus 0.1 point per goal scored, or some other crazy means of affecting the standings.

The 3-2-1 system of declining points for a win the longer it takes to decide the game is, IMO, a better idea than the current system, but inferior to the old school system or the 3 points per game system. I would hate to see a game where one team only one point to eliminate its opposition from the playoffs - just reaching the shootout would be as good as a win, so think how dull OT would be...

Oh, for the good old days...vci34

[Vik]

But the point of this wasnt to improve statistics it was to increase the chances would take in OT. If you introduce no points for an OT loss it just causes teams to become more conservative knowing its 2 pts or 0
But you'd go for the win in OT because otherwise it's 1 or none. Personally, I'm all for the 3 points awarded in every game system.