Crawl Across the Ocean

Wednesday, January 05, 2005

Making it Look Easy

Congratulations to Canada's Junior Hockey team, whose overwhelming dominance of this year's tournament makes them clearly the best junior hockey team ever assembled.

As an aside (and to make a dull entry a bit more interesting), I find it odd how people have debates over which sports teams are the best ever, in which they compare teams from different eras. In every measurable field of athletic endeavour, the athletes of today are far better than those of the past.

Similarly today's athletes in team sports are bigger, stronger, faster, in better shape and better trained than those of the past. I for one, am very confident that the 2004 Edmonton Oilers could easily defeat the 1986 Oilers.

As a further aside, the argument that the NHL is suffering from dilution of talent is equally nonsense. Just the infusion of foreign talent alone would have been enough to compensate from going from 21 teams (as in the high scoring 80's) to the current 30. Add in the constant improvements in the overall talent base and it seems pretty clear that the league has a greater depth of talented players than ever before. In fact, it's more likely that scoring is down because there's too much talent (especially in net and on defense), rather than the other way around.

Labels: , , ,


  • Totally agree with you that the NHL's talent base has not been diluted over the past decade, despite expansion.

    However, I disagree that the 2004 Oilers are better than the 1986 Oilers. I'll try to do a little math to back that up and post again later.

    By Blogger dejour, at 2:25 PM  

  • OK - I've looked at the improvement in times for selected men's track and field sports in the Olympics between 1984 and 2004. What I've done is compare the 8th place finisher between the two decades (this is to avoid the distorting effect of an outlier in 1st)

    Long jump 1984 Kim Jong-IL 7.81m
    Long jump 2004 Bogdan Tarus 8.21m
    increase 5.1%

    High jump 1984 Cai Shu 2.27m
    High jump 2004 Svatoslav Ton 2.29m
    increase 0.9%

    Shot put 1984 Erik deBraun 19.65m
    Shot put 2004 Ralf Bartels 20.26m
    increase 3.1%

    Javelin 1984 Laslo Babits 80.68m
    Javelin 2004 Tero Pitkamaki 83.01m
    increase 2.9%

    Also of note Carl Lewis' 100m run in 1988 would have given him 5th in 2004.

    Steve Lewis' 400m run in 1988 would have given him gold in 2004.

    John Ngugi's 5000m run in 1988 would have given him gold in 2004.

    (Note, these sports were chosen before looking at the stats, so there is no intentional bias here.)

    So, if one were to assume that hockey players are 3% better now than they were in 1986, then perhaps two adjustments should be made to Gretzky's 215 point total in 1986. One to reflect the decrease in overall league scoring, plus an additional 3% deduction for tougher competition today.

    OK, so goal scoring per team was at 317 in 1986, but was 211 last year. Therefore multiply the 1986 player stats by a factor of .66. Change it to .64 to reflect the increases in quality seen in track and field.

    Now looking at the top 5 players from each team, the adjusted scoring stats (pro-rated to full seasons) are:

    Gretzky 138
    Coffey 89
    Kurri 86
    Anderson 73
    Messier 68

    R Smyth 59
    York 56
    Dvorak 53
    Nedved 47
    Horcoff 41

    Of course this analysis ignores defensive prowess, but in both cases the Oilers were just slightly better than the league average in defensive record.

    Anyways, my point is that I think the '86 Oilers were significantly better than the current version and the numbers support that view.

    By Blogger dejour, at 3:03 PM  

  • I think it was Ken Dryden that said that for pretty much everyone the NHL was never better than when they were ten years old. Therefore I expect some people our age to be offended by the Oilers comment Some people a decade older will likely chime in that the late 70’s Habs were the best ever conveniently ignoring that the rival WHA and lack of Europeans made the general level of talent the thinnest it has been in history.

    I do agree with the general argument that the calibre of player (different from the calibre of play!) is much higher today than it was just a few years ago, the three main reasons being the European influx, money, and culture. The last two are somewhat related in that as the money stakes have become higher players are doing more to stay in shape in the off-season and recuperation after games no longer consists of a few rounds with the guys. I’m not convinced of your specific case of Oilers vs. Oilers though. Most of the names on that ’86 team have played until just recently and a handful were key contributors to the Rangers’ Cup eight years later. I don’t think they could have been so effective for so long if the average player was evolving at the rate you suggest. Of course you could probably counterargue that the Oilers were the youngest ‘great team’ of all-time.

    I will say this though… if Ty Conklin could transport himself back to 1986 he could easily replace Grant Fuhr as the Oilers’ #1 and would be a first-ballot Hall of Famer today. The evolution of the goaltender in the past twenty years has been absolutely astounding. A .900 save percentage used to lead the league, now it won’t keep you in the league! You often hear about Orr and Gretzky changing how the game is played but I would argue that Roy and Hasek exceed them in that department. I wonder, when was the last time a goalie stood in the net with an *“In the Crease” pose?

    *painting by Ken Danby

    By Blogger Spearin, at 3:17 PM  

  • The current track results are closer to 1988 results than I thought. You have planted a small seed of doubt, and maybe now I would only say that the 2004 Oilers would beat the 1986 ones, but perhaps not with ease, rather with superior tactics, speed, checking and goaltending.

    One thing to consider: The 3% increase (roughly) in track results is the improvement on the base of a zero length jump. So if you imagine a 0% talent hockey player as being one who can't skate or even hold a stick (the equivalent to someone who can't jump even a single inch) then the 1986 hockey player would be 100% and the 2004 player would be 103%. This could be a very significant difference. If you put my hockey skills at 30%, then a decent beer league player could be 60% and a decent junior player would probably be upwards of 90%. The difference between 100 and 103% in hockey talent could be the difference between the best player in the NHL and the worst (or larger - or less, hard to say).

    Since you can't take the worst player in the NHL (with 0 points) as being the 0 against which you measure your percentage improvement, you can't just take 3% of the '86 Oilers points away and still have a valid comparison with the track results.

    That wasn't particularly articulate, but hopefully you get what I'm trying to say: A 3% improvement on a scale ranging from absolute worst to absolute best can look like a very big change in a league which only contains people from a very small range of the total distribution.

    If you measured the % increase in track results using the last place finisher in the finals of each event as '0' then you would have a more accurate number for adjusting points with.

    Also, I suspect the improvement in hockey players has been greater than that in track athletes over the last 20 years (especially in net as Spearin notes) but it's not really a fruitful line of debate, since there's not much to prove either way, it's just my opinion.

    By Blogger Declan, at 7:52 PM  

  • Two points to consider:

    1) The 1980s coincided with the peak physical years of the largest cohort of Canadians. Thinking back to demography class, I think the largest cohort of Canadians was born in 1962. Assuming peak hockey years are 25-30, that means that 1987-92 was a golden age.
    I seem to remember my year of birth, 1974, was a low water mark - meaning that 1999-2004 might be expected to provide a smaller level of talent.

    2) Maybe a better comparison would be to track how earlier generation athletes in track would do in later competitions. (eg. if the 2nd place finisher in track in 1984 would finish 5th in 2004, that might mean the 2nd place scorer in hockey would be 5th in 2004). I'll investigate this.

    By Blogger dejour, at 8:43 PM  

  • OK, the '86 Oilers had the #1,#3, and #4 scorers in the league.

    What I've done is look at a variety of track and field events and see where the #1,#3,and #4 finishers from 1988 would have been had they competed in 2004. (I decided against using 1984 as that was a boycotted Olympics.) If the results show that the average bronze medal from 1988 would be good enough for 7th in 2004, I would suggest that a similar movement could be expected amongst hockey players (the 3rd leading scorer in 1988 would translate to the #7 scorer today)

    First place

    100m 4.5 (meaning worse than 4th but better than 5th)
    400m 0.5 (meaning better than the 2004 gold medallist)
    5000m 0.5
    high jump 0.5
    long jump 0.5
    shot put 0.5
    javelin 3.5
    400m hurdles 0.5
    pole vault 2 (meaning tied for 2nd in 2004)
    discus 1.5

    All in all, 1st place in 1988 seems to translate to at worst 2nd place in 2004.

    Third place

    100m 6.5
    400m 1.5
    5000m 4.5
    high jump 1
    long jump 4.5
    shot put 0.5
    javelin 5.5
    400m hurdles 0.5
    pole vault 4
    discus 1.5

    3rd place in 1988 translates to roughly 3rd place in 2004.

    4th place

    100m 6.5
    400m 3.5
    5000m 4.5
    high jump 1
    long jump worse than 8th
    shot put 0.5
    javelin worse than 8th
    400m hurdles 1.5
    pole vault 7.5
    discus 3.5

    Incomplete data makes this difficult to read, but say that 4th in 1988 translates to 6th in 2004.

    I would suggest that the Oilers of '86 could expect to have three scorers amongst the top 6 in the league if they played in 2004. This would very likely translate to an elite team in this day and age.

    By Blogger dejour, at 10:05 PM  

  • I'll concede this much - if the rate of improvement of NHL players has been the same as the rate of improvement of track and field athletes (from mid 80's to current day) then the 1986 Oilers would beat the 2004 Oilers.

    It seems hard to imagine watching footage of those old games, but perhaps I am being fooled by some optical illusion which makes the players of the 80's look slow, small and defensively inept in comparison to today's players.

    By Blogger Declan, at 11:34 PM  

  • I'm a bit slow posting here, but the 86 Oilers vs the 04 edition you say... perhaps it would depend on whether Steve Smith figured out which net he was supposed to shoot on :)

    Interesting stat - last year Mark Messier, even as an old geezer, still managed to outscore Shawn Horcoff, 43 points to 40 (and he played a couple fewer games).

    If the 86 Oilers did play the 04 Oilers, what pads would the goalies wear? Would Conklin still be a first ballot hall of famer if he had to wear 80s size pads (Spearin)?

    For the record I guess, I think that if the two teams played a best of seven series, the 86 Oilers would win in six. I agree with Declan insofar as the average player today is superior to the average player 20 years ago, and an average team from last season probably would beat most of the teams from the 80s, but the 86 Oilers weren't an average team. Looking at how long players like Gretzky, Messier, Lemieux, Bourque and MacInnis contributed after their 80s heyday finished leads me to believe you're overrating the improvement of today's players. I think the 86 Oilers had enough elite level talent that they could beat what was a very mediocre 2004 Oilers team, regardless of how much stronger, faster and better conditioned their fourth liners were (sorry, no stats analysis).


    By Anonymous Anonymous, at 10:11 PM  

  • I think even with the little pads, Conklin would still be far superior to Fuhr. You guys have at least convinced me that it could be a pretty interesting series to watch. Perhaps someday we'll be able to realistically simulate it (probaly not, though).

    By Blogger Declan, at 10:17 PM  

Post a Comment

<< Home