The League of Extraordinary Statisticians (LOES) is a weekly forum bringing together the top analytical minds in the hockey world to answer a variety of questions that straddle the line between stats analysis and something you might hear floating around section 304. They have agreed to answer these questions in a few paragraphs or less, and with minimal formulae. Because this is a forum, we'd encourage you to use the comments section to answer the questions yourselves, or to discuss or debate the answers given.
The LOES is not meant to represent the entire of the hockey stats community. There are a number of people that either were too busy or too difficult to contact for the purposes of the forum.
While it is fun to explore what a Howie Morenz or Joe Malone would do in the 1980s NHL, or Wayne Gretzky in the 1920s, most statisticians would agree that it is difficult to compare performances from either of those periods because the NHL has changed significantly from 1920 to 1989. Within that wealth of changes are stretches of major change that either ushered in or ushered out new approaches to the game and/or different statistical performances. This is probably the quickest definition we could have of "eras" in the NHL, but where the debate lies is when these shifts occurred. There is a tendency to divide time up in decades, with each ten-year span having distinct characteristics, but would the NHL's historical eras break down in this way?
Fair warning, this one involves some reading; but it's important reading, so I'll just let 'er rip:
If you had to organize NHL history into "eras", what eras would you identify? What would be distinct about them?
You try to split based on the change in talent distribution or style of play. And you try to have at least ten years in there (more for logistic purposes than anything). So, the expansion from 6 to 12 teams is an easy one. The merger with the WHA is another. In between, you can even argue for multiple eras, but then you are left with a four or six year era. The next big one is early 90s when the scoring dropped substantially. The lockout is another one.
- Tom Tango, tangotiger.net and author of The Book: Playing the Percentages in Baseball
Wow, that could be a whole article in and of itself. Taking in expansion, major rule changes, changes in scoring levels, and other major events (ie: arrival of Soviet players).
In very broad strokes I would think pre-Original 6 (1917), Original 6 (1942), Expansion (1967), and Modern (1992).
- Rob Vollman, Hockey Prospectus
I would try to define the NHL's eras based on the game on the ice, not on external factors. People often refer to the "Original Six Era", for instance, from 1942-43 to 1966-67. Well, the only change from 1941-42 to 1942-43 was the disappearance of the Brooklyn Americans; nothing changed on the ice. Even the great expansion in 1967-68 didn't really change anything on the ice; the game was the same, but it was being played by more people in more places. But to me, it's not who's in the league, it's about the game.
Significant rule changes can obviously affect the game on the ice. The introduction of modern forward passing rules in 1929-30 obviously had a large effect on how the game was played. Changes in strategy can also change the game; going from a starter/sub approach to rotating forward lines, for example. So one way you might organize the NHL's eras might be:
1917-18 to 1928-29 (the Early Era): This era features increasing use of substitutes, and eventually most teams having two forward lines. Forward passing was still not allowed in the offensive zone, which makes goal-scoring tactics much different from the modern game. This era existed before 1917-18, but I'm using the arbitrary dividing line of the beginning of the NHL. It would certainly cover the NHA years as well (1909-10 to 1916-17).
1929-30 to 1967-68 (the Middle Era): The rules are now basically in their modern form, and the early parts of this era saw the development of three forward lines being used in rotation. There is a blip from 1942-43 to 1944-45, where the quality of play dropped significantly due to players going off to war.
1968-69 to present (the Modern Era): Bobby Orr revolutionizes the role of the defenceman, and the game becomes more free-wheeling and less locked into up-and-down-the-wings hockey, if you know what I mean. There are blips in the early 1980s, which has been called the "Age of Air Hockey" due to its offensive extremes, as well as in the late 1990s to early 2000s, which I would call the "Age of Clutch and Grab", which depressed scoring. Whether or not those should be categorized as seperate eras is up for debate.
- Iain Fyffe, Hockey Prospectus
The game before 1950 is barely recognizable.
1950-1967: This is the modernization of the NHL. The ice was painted white, teams started playing 70-game schedules. There's considerable change in this era (slap shots, goalie masks, some helmets), but nothing compared to what was to come.
1967-1979: Pro hockey was like the Wild West. Canada almost lost to the Russians in 1972 and clearly wasn't by far the best hockey country in the world. Expansion went crazy - in 1976, there were 32 pro teams in North America, even though almost every player was from Canada. Thirty years later, we have two fewer teams and a much larger Canada produces only half of the players in the NHL. There was massive talent dilution, particularly in the goaltending department. We don't have full stats from this era, but there were some seriously bad professional hockey players.
1979-1991: This is an era marked by high offense, superstars, and dynasties built from the ground up in a matter of years. The disparity in competition was way larger than it is today.
1991-2004: Parity and low-offense. I have to admit that like many people, I didn't pay much attention to the game in this era.
2005-2010: The new game. Teams have cost constraints and we have a resurgence in offense.
- Gabriel Desjardins, behindthenet.ca and, of course, Behind the Net blog
Whenever you organize history into eras, you have to define some boundaries which appear sharp, even though they are boundaries; they are easy to be seen as something far more significant than they actually are. For example, it is clear that in 1967, things changed significantly with expansion, things didn't change too significantly in 1967 in many ways. For example, Stan Mikita won the Hart Trophy in the year before and after the expansion and that clearly is not a significant change. With that disclaimer, I would set up eras roughly as follows:
1917-1926 - The early years. There was a small league with a maximum of seven teams and a minimum of three. The NHL was not the only league running. The PCHL and WHL also existed in the west and competed with the NHL for the Stanley Cup. The league was high scoring and did not have all of the best talent in the world as some played out west and in the US leagues.
1926-27 - This is a transition season. With the death of the leagues out west, the NHL began expanding in earnest to the US. Western players were sold to US owners and the NHL ballooned to ten teams. The best players in the world all started to show up in the NHL since they lost the leagues in the west and the American players who hesitated to leave the US now how NHL teams to play with. The next season had a significant rule change with the introduction of forward passing in the offensive zone, so this an outlier year. It doesn't fit in the previous era, but it is clearly different from the next era.
1927-1938 - The first NHL expansion. The NHL was recruiting the best talent in the world. For the most part the league was stable, although the Great Depression slowly dropped us from ten teams to eight. The league had some of its first big name hockey stars who were known throughout the hockey playing regions of the NHL in Eddie Shore, Howie Morenz and to a lesser degree others. The NHL was an established league.
1938-1945 - The war years. World War II broke out in 1939. That was when Canada entered the war and Canada was the country that produced almost all NHL players. This was a period of instability financially and lack of talent, due to players overseas at war. Many players who had been mid-level players became stars in the depleted league and were never stars again once the war was over. I start this era a year before Canada entered the war because the NHL fell to seven teams in 1938 with the folding of the Montreal Maroons. The NHL lost some talent due to the folding of franchises. It wasn't clear that the all the best players were in the league, some legitimate talent left the league to early retirements when their teams folded. Maurice Richard's famous 50-goals-in-50-games season occurred in the final year of this era and this result should be considered as suspect since he played against lesser talent.
1945-1955 - The establishment of the six-team league. Toronto and Detroit dominated these years. Gordie Howe and Maurice Richard dominated on the ice. It became clear that there was little parity across the league as certain well run teams, did well every year and Boston, Chicago and the NY Rangers became bottomfeeders.
1955-1967 - The heart of the six-team era. Montreal replaced Detroit as a top team. Toronto remained at or near the top. Chicago was good enough to make a run as well. The NHL had its best era in terms of overall talent levels in this period. The league was a legitimate big league on par with baseball or football in its American markets. There is a clear increase in overall talent in the league from the earlier decade. The exact year to make this divide is hard to pinpoint, I am using the start of the Montreal five-cups-in-five-years run and the decline of Detroit as an era cutoff, but it could easily be moved a year or two in either direction.
1967-1972 - Expansion. The league grew from six to fourteen teams in this era with expansions in 1967 and 1970. There was little parity. The established teams won and the new teams lost. The early Stanley Cup format of that time with the best Original Six team against the best expansion team made for a final that was not as exciting as the Original Six years.
1972-1979 - Over-expansion. The NHL and the WHA existed and we often had about 30 professional hockey teams in North America. There wasn't enough talent to go around and some teams were very bad (Washington, Kansas City/Colorado as examples). The best teams remained quite talented (Montreal, Philadelphia and at the end of the era the New York Islanders). Some of the expansion teams had become solid teams.
1979-1991 - The 21 team era. This was a high scoring era when the NHL significantly improved in talent due to the loss of the WHA. This was a Wayne Gretzky-dominated period looked back upon as one of the most stable periods of high-level hockey.
1991-1994 - Transition from 21 teams. The NHL began expanding, adding 5 new teams in this era. Europeans began coming to the NHL in record numbers, with the fall of the Berlin Wall opening up Eastern Europe and stronger scouting in Scandinavia. The NHL remained high-scoring.
1994-2004 - Dead Puck Era. I hate this name. Scoring dropped significantly with the successes of the trap, which was first seen in New Jersey. It was largely high-quality hockey, except in recent expansion cities, but it was far more defensive than the league had seen in years.
2005-present - Post-lockout years. The lockout lost us a season and when we got back, we had rules put in place to prevent the formation of the elite teams we had seen in years past with the salary cap and increased free agency. The formation of the KHL began to draw talent away from the NHL. Alex Ovechkin and Sidney Crosby both made their debuts and dominated the next several years. No team could remain atop the league for a long period of time and we saw a series of one-and-out Stanley Cup champions. Exactly where we are going and when it will end is unclear at this point.
Now as a counter-example to all this work, Cy Denneny was in the league from 1917-1929 and would have argued there were there were no major changes in his career. Dit Clapper was in the league at that time and stayed around until 1947 and would have argued there were no major changes in his career. Gordie Howe was in the league at that time until his first retirement in 1971 and he would argue for little to no change in his career. Bobby Clarke was in the league at that point, and lasted until 1984 and would have argued the league changed little in his career. Chris Chelios was an NHLer at that point and finally retired this summer and he might argue that hockey didn't change much in his career. Hence, hockey is the same as it ever was from day one until today. Except clearly it wasn't. Exactly where we draw the boundaries is the question and this has been my best attempt at defining some. They are relatively logical points, but some could be moved by a year or two.
- Greg Ballentine, The Puck Stops Here at Kukla's Korner
I don't like the use of eras because if you have to separate time frames into eras that means there is a start point and an end point. I see a league's development as a gradual process, with multiple variables, where changes happen over an extended period of time without any hard form of "This is Era X from Point A to B and this is Era Y from Point C to D."
The only times I would draw exception to that is when league structure, rule alterations or any form of significant immediate change(s) impacts the game. Things like the Original Six expanding to twelve teams or the '05 Lockout and the resulting changes in the league rules and CBA would be examples of that.
- Corey Pronman, Hockey Prospectus
I'm not qualified to speak of the NHL before the 1940s, and it was a vastly different game from today's anyhow. In the post-1950 period, I see 4 eras: the "Original Six" era, from 1950 to 1967. This was
characterized by low scoring, a high level of play and stagnant rosters. The heroes of this era have become legends: Howe, Beliveau, Hull, Plante.
The second era is the "Expansion" era, from 1967 to 1979. Between rampant expansion and the arrival of the WHA, the quality of NHL play dropped dramatically, with a massisve gulf between the best teams
and the worst, and higher scoring to go along with it.
The period from 1979 to 1994 should be known as the "Gretzky" era, the highest-scoring era in NHL history. With the WHA consolidation and the arrival of Finnish and Swedish (and later Czech and Russian) players, caliber of play crept back to a high level, but two teams, the Islanders and Oilers, still dominated for a decade.
The last era is the "Bettman" era, starting in 1994 after the last lockout and continuing today. Like the Original Six era, this is characterized by low scoring and a high level of play, but extreme roster
fluidity, as the bottom halves of NHL rosters churn constantly. While I'd prefer scoring levels half-a-goal higher, either by shrinking goaltending equipment or increasing the size of the net (it's not normal that goaltenders can stop more than 92% of shots at even-strength), there's little doubt we are witnessing some of the best hockey the NHL has ever put on ice.
- Tom Awad, Hockey Prospectus
A lot of information to digest, I know, but it is interesting to see some of the responses focus on eras as larger categories and some involving smaller periods in time. Universally, the notion of breaking down the NHL into eras identified by the decades (1920s, 1930s, and so on) is not suggested, and judging by the analysis presented here it makes sense to not do that. Do you think that NHL history is better expressed in larger identified eras or smaller ones? It's also interesting to think about what influences these shifts; in many cases, it is expansion that shakes up talent distribution. Does Bobby Orr's dominance get noticed in a tougher league, and therefore influence what we think a defenseman should do? Or the high-flying Oilers approach, offensive-based and built on outscoring (but not outshooting) their opponents? Rule changes might also affect things, as the lockout marked a shift yet did not involve expansion. There's also the question of European players: how have they impacted the game? In many ways, as Ballentine's reponse proposes, this is a continuous discussion, one that we haven't quite figured out yet.