WordsOfWisdom

Although the system is sunk, here were my revised divisions anyway: Revised division 1: Detroit, Toronto, Buffalo, Ottawa Revised division 2: Boston, Montreal, Rangers, Islanders Revised division 3: Pittsburgh, Philly, Capitals, Devils. Revised division 4: Lightning, Panthers, Hurricanes, Columbus That truly nails it I think. Yes I know about the Devils, but the New York area has too many teams. Someone gets left out no matter what. I think Rangers and Islanders fans would trade New Jersey for Boston and Montreal any day of the week. In baseball, the Yankees and Mets are in separate leagues. Actually, wait....... now that I think about it: I'd do this: Revised division 1: Detroit, Toronto, Buffalo, Ottawa Revised division 2: Boston, Montreal, Rangers, Devils Revised division 3: Pittsburgh, Philly, Capitals, Islanders. Revised division 4: Lightning, Panthers, Hurricanes, Columbus There. That covers all the rivalries.

Thanks. Now that I think about it, Boston is a long way from Toronto. How about this: Revised division 1: Detroit, Toronto, Buffalo, Ottawa Actually I just discovered a major flaw that sinks it: You can't have all four teams in the division making the playoffs. That's just not right, even with a strong division. It ruins division based rivalries if everyone gets in.

But what about my newly retooled eastern conference divisions? (shown below) Divisions of 4 seems small, but I don't see it as an issue. The 1st place team in each division would get home ice in the playoffs. Everyone else would fight over the remaining 4 wild card spots. (Assuming you want to keep 8 teams in the playoffs per conf.) Using the division format above, this would be the playoff picture today: Division 1: - Bruins - Leafs - Red Wings - Senators Division 2: - Devils - Rangers - Islanders - Canadiens Division 3: - Capitals - Penguins - Flyers - Sabres Division 4: - Lightning - Blue Jackets - Panthers - Hurricanes Team seeds would be: 1. Lightning 2. Bruins 3. Capitals 4. Devils 5. Leafs 6. Penguins 7. Blue Jackets 8. Flyers The Devils (being the product of an exceptionally weak division), would be seeded 4th. That would be the value of winning your division.

That's the main problem with having too many divisions. You get weak division leaders. However, given the parity in the NHL today, I don't think it really matters any more. With an 8-division format, you could have the division leaders get in automatically, and be seeded 1-4 (home ice), and then fill out the remaining spots with wild card teams.

The new eight division format would get the NHL primed and ready for expansion to 40 teams.

Thanks. I thought about the same thing. The difficulty is, you practically have to make a division out of the New York area teams because they have so many. Someone gets pushed out. If you put the Devils back in, then it's bye bye Capitals.

How about this: Division 1: Detroit, Toronto, Boston, Ottawa Division 2: Montreal, Rangers, Islanders, Capitals Division 3: Pittsburgh, Philly, Columbus, Buffalo. Division 4: Lightning, Panthers, Hurricanes, Devils

" Division #3: Boston Bruins, Buffalo Sabres, Ottawa Senators, and Montreal Canadiens. Division #4: Columbus Blue Jackets, Detroit Red Wings, Toronto Maple Leafs, and Pittsburgh Penguins. " That would suck so hard for Toronto. Would never fly. You're basically keeping the current division mostly intact but removing the Leafs from all of their current division rivals. Leafs fans would want to see Boston, Buffalo, Ottawa, and Montreal. Columbus would be of no interest to Leafs. Pittsburgh generates no intrinsic interest other than being a good team. Detroit is a classic rival of Toronto but they're rebuilding now and don't hold much appeal. Plus you could never split Pittsburgh and Philadelphia into separate divisions. Wouldn't fly. I think fewer divisions is the way to go. Four divisions. Shift someone out to make room for Seattle if needed but no other re-alignment please.

I went to the website but I didn't see any graphs or anything. I saw lots of tables with statistics but no charts or graphs. I do software development so it's something I might be able to help with. I would expect compensation though of course. (I only give my services away for free to family members lol.)

What do you need help with from a web design standpoint?

I don't know why but I hate the orange in the Ducks jerseys. It doesn't fit. They should just have black and gold. Orange and gold doesn't work, and I suspect those jerseys won't have a long future in the NHL. I do prefer the older logo though, although both are bad in my opinion. Can't take either seriously.

If I had to rank them, I'd say... 1. NYR 2. Islanders 3/4: Coyotes/Ducks (hate them both and hate both of those logos equally)

That would work if they made the arm strips orange OR one stripe orange and the other stripe blue. Like so:

Total Leafs ripoff, but I guess imitation is the greatest form of flattery.

What I find interesting is how teams essentially "swap" jerseys over time. Since there are only so many color combinations that work, and only so many colors that teams will use (pink, purple, and brown will never be used for example), it means that new designs become the old designs of other franchises. Example: Sabres -> Ducks Whalers -> Canucks Leafs -> Lightning Knights -> Canucks Flames -> Devils etc..

I was just about to say that they beat the Vancouver Canucks to their own jerseys! Vancouver was going to switch back to the 1990's style uniforms but now Vegas has the same colors. The transformation will be complete once the Knights switch the silver to black, and I think that change will happen sooner rather than later. Within five years they will look exactly like the Canucks from the mid 90's.

Something to ponder: People are always quick to point out Gretzky's advantage in points compared to Lemieux, and don't seem to care that Lemieux had injury problems and battled cancer. When it comes to Orr, people are willing to overlook the fact that Ray Bourque and Paul Coffey have double the points that Orr had. Orr manages to stay #1 on most lists despite his knee problems, but Lemieux always seems to be punished for his lymphoma.

" 15-save shootout should not be the same as 40-save shootout, although for any of the four stats listed above they create two identical entry." You really should adjust your terminology to avoid confusion. A "shootout" means something very specific in hockey, and what you're referring to is a 15 save or 40 save game. "Bonus = (Saves - 10) / 200 If there were less than fifteen saves in the shootout, the bonus is assigned the minimum value of 0.025. We consider adding this bonus necessary, because the opposing team is usually gives an extra effort to avoid being shut out even during the garbage time." Where and how are you coming up with these figures? Are you just assigning arbitrary bonuses?

One comment I would make is that new hockey stats still need to be "fan friendly". There is a lot of complex analysis work that can go into evaluating players, but it flies over the head of the average fan. Your average fan needs something simple and meaningful when it comes to stats. I do like reading your blog entries though.

I once constructed a system of allocating points to goaltenders called (not surprisingly) "goalie points" (or GPts for short). The idea being that they would accumulate as the season progressed just like points do for skaters. I scrapped it however because it didn't work well. Basically the way it worked was: 1 point for every period that the goalie plays 1 point for every 10 saves that the goalie makes -2 points for every goal allowed Thus, a goaltender that has a 30 save shutout performance would get 6 points for that game. (Basically the best performance a goaltender can do without stepping into record breaking 40, 50, 60 save performances and the like.)

I'd replace JVR on that list with Matthews... already. See this is what happens when you try to make a TOP 100 list while having only 50 suitable candidates.

I agree. The existing system encourages teams to get into OT. However, that can be easily solved in a few different ways: Back to 2-1-0 (wins, losses, and ties). No points just for reaching OT. My modern 2-1-0 + shootout idea (where the only way you can tie is after 3 shooters in a shootout)***. Introduce a 3-1-0 system and bring back ties. (Could remove the shootout for example.) Wins and losses only. (Teams play until someone wins. No more points system.) ***My 2-1-0 + shootout system was actually discussed over the radio one time (because I submitted it to the host as an idea). I can truly call it "my" idea without sounding arrogant (I hope) because I'm the only person on earth that has ever suggested it. It is truly unique. The notion that you could have a shootout and still have the game end in a tie is something that most people can't wrap their head around. It's a concept that blows people's minds. You could end the shootout after 3 shooters and call it a tie game. At that point, the fans would have seen 60 minutes + 5 minutes of OT + a shootout. They're not being shortchanged. If a team still can't win the game after all of those opportunities, they don't deserve 2 points in my book. And so it would be: the teams would split the 2 points, leaving with 1 each. Ties would be a rare occurrence, but they would happen from time to time. The NHL could stop with the gimmicks in overtime (4 on 4, 3 on 3, etc.) and just play a normal overtime session. Shootouts would be exciting because the stakes are raised. You could win (getting 2 points), tie (1 point), or lose (0 points). Ditto for overtime. There wouldn't be a safety net like there is now. It's what the NHL has been missing. The NHL had a problem of too many games ending in a tie back in 1997, 1998, whenever it was. I'm the only person that has ever created a points system that would have solved that problem 20 years ago without damaging the current points structure, without changing the value of wins and losses, without introducing loser points, and while allowing the NHL to either have or not have the shootout. (Okay I promise not to toot my own horn any more tonight.)

In leagues that have used a 3-1-0 point system (such as European soccer), a victory was worth 3 points, a tie was worth 1 point, and a loss was worth 0 points. Under such a system, wins and losses had a consistent value under all scenarios (regulation, OT, or whatever). This is a very different concept from what a 3-2-1-0 system offers. Although there may be 3 points available in each game, such a system creates an environment of "spoiled victories" and "good losses". You essentially create shades of grey. Some wins are good wins, others are bad wins, etc. Some losses are good losses to have, others become bad losses. We've already seen a bit of what this looks like with the NHL's current loser point system, and most fans dislike it. It's also fundamentally flawed in a purely mathematically way: 0-0-82 = 82 points > 40-42-0 = 80 points. A team with 0 wins (and 82 losses in overtime) can be positioned ahead of a team with 40 wins. It's the most outrageous math fail in NHL history that the league adopted such a system. A 3-2-1-0 system carries the same idea forward and applies it to both wins and losses, further magnifying the problem. At first glance it may seem like the answer, but I liken it to the scene in Indiana Jones where the Nazis open the ark. Good discussion though.

Although this topic is covered in other threads already, there is one fundamental problem with a 3-2-1-0 points system which is not solvable: If you install a 3-2-1-0 points system, then you admit that some victories are worth more than others and also that some losses are worse than others. To draw an analogy (and to shine a light on how big this flaw is), if the same system were applied to baseball it would mean that winning a game in the first 9 innings is worth 3 wins in the standings, a win in extra innings is only worth 2 wins, and if the game goes to a home run derby after 15 innings, the winner gets 1 win in the standings.

Oh yeah that's right. They have police sirens. Not as cool, but at least it's different.