2015-16

[Tom Lento]

(3) If you're not BC or Minnesota, you might as well build out from a stingy defense.

I'm not so sure that's still the case. Even if it is, I'd rather lose at Lynah by a score of 4-3 than 1-0. I think the "it's all your fault" chant is the most intimidating chant we have and I'd rather hear it as much as possible. Of course, I'm an offense and speed freak in any sport.

Looking at the last 10 national champs (of which there have been 8 different schools) offense appears to be carrying the game. 6 of the last 10 champs have had an offense which ranked higher nationally than its defense, which doesn't seem to be that big a deal. However, only once in the past 7 years had a champion had a higher defensive rank than its offense. The average goals scored per game for the last 10 champs is 3.60 with only one team averaging less than 3 goals per game (Yale, 2.89). The average goals against for those champs is 2.30 with only one team holding its opponents to less than 2 per game (Wisconsin, 1.84).

Maybe it's just cyclical, but it seems like the rules will continue to favor offensive play going forward.

You can't get from that bag of meaningless numbers to an assertion that rules will continue to favor offensive play going forward. You can't even get from there to an assertion that the rules have been favoring offensive play over the last 10 years. 6 of the last 10 is roughly half, so that tells you nothing regardless of the direction. 1 of the last 7 sounds impressive in favor of offense, except it means nothing if those teams all had top 10 offenses and top 15 defenses. 3.60 GFA certainly sounds high in the context of current season scoring for Cornell, but in 2003 the Big Red averaged 3.69 goals per game.

I don't feel like looking this up, but maybe someone out there will do this for me:

Plot mean, median, top quartile, and top 10% (i.e., p90) goal scoring for NCAA D-1 teams during the Schafer era. X axis year, Y axis GFA for each data series. Include Cornell, just for shits and giggles, or perhaps to bemoan the fact that 2003 was really the only consistently effective team offense in the Schafer era (disclaimer - I don't know if that's actually true, but I suspect it was - 2005-06 and 2009-10 both had sub-3 GFAs despite some good scorers at the top of the lineup).

Do the same for goals allowed.

Then maybe we can talk about cyclical scoring trends and rule changes and whatnot. My hypothesis: all of the "pro-offense" rule changes people keep going on about and the "fundamental change in the way the game is played" that has "passed Schafer by" has had basically zero impact on actual scoring.

These charts won't test that hypothesis directly, but they'll at least be a more reasonable indicator than the cherry-picked numbers or blind assertions that go along with all of these arguments about larger trends in the way the game is played.

[BearLover]

bemoan the fact that 2003 was really the only consistently effective team offense in the Schafer era (disclaimer - I don't know if that's actually true, but I suspect it was - 2005-06 and 2009-10 both had sub-3 GFAs despite some good scorers at the top of the lineup).

2005-06 yes (2.83), but 2009-10 was 3.15. 2004-2005 was 3.20.  Cornell is currently scoring way too few goals per game to even come close to competing at a nationally competitive level.

[Jim Hyla]

bemoan the fact that 2003 was really the only consistently effective team offense in the Schafer era (disclaimer - I don't know if that's actually true, but I suspect it was - 2005-06 and 2009-10 both had sub-3 GFAs despite some good scorers at the top of the lineup).

2005-06 yes (2.83), but 2009-10 was 3.15. 2004-2005 was 3.20.  Cornell is currently scoring way too few goals per game to even come close to competing at a nationally competitive level.

Wow, is that really a surprise to anyone? Even Coach Schafer said he tried changing the checking to generate more offense, it just didn't work.

[RichH]

Cornell is currently scoring way too few goals per game to even come close to competing at a nationally competitive level.

Absurd hyperbole. Cornell was one tie away last year of getting an at-large bid, and the number of times we're in the discussion for at-large bids is a majority of the years. Parity of the national landscape also serves to sink this statement.

[BearLover]

Cornell is currently scoring way too few goals per game to even come close to competing at a nationally competitive level.

Absurd hyperbole. Cornell was one tie away last year of getting an at-large bid, and the number of times we're in the discussion for at-large bids is a majority of the years. Parity of the national landscape also serves to sink this statement.

Cornell also usually scores considerably more goals each year.  Even last year they scored over 1/2 more goals per game.  

Here is the trend, based on data as far back as I can find:

2015: 1.84
2014: 2.41
2013: 2.44
2012: 2.86
2011: 2.53
2010: 3.15
2009: 2.56
2008: 2.83
2007: 2.90
2006: 2.83
2005: 3.20
2004: 2.66
2003: 3.69
2002: 3.37
2001: 2.21
2000: 3.25
1999: 3.10

Notably, there is a general downward trend.  Also notably, 2015 is stunningly low--by far the worst Cornell has ever been at scoring under Schafer.  So I don't retract my statement that we are scoring at way too low a rate to be close to competitive on a national level.  This year may have been aberrational--but last year was equally aberrational in that we almost made the tournament despite being +3.  2009 was the only year we made the tournament without scoring at least one more GPG.

[RichH]

Notably, there is a general downward trend.  Also notably, 2015 is stunningly low--by far the worst Cornell has ever been at scoring under Schafer.  So I don't retract my statement that we are scoring at way too low a rate to be close to competitive on a national level.  This year may have been aberrational--but last year was equally aberrational in that we almost made the tournament despite being +3.  2009 was the only year we made the tournament without scoring at least one more GPG.

Well, geez. Here's the thing...I can never tell about which timeframe you're making each of your arguments. Sometimes it's over a 4-year span. Sometimes 10. Sometimes its the selected range where Yale was kicking ass. Sometimes, like this instance, it's one season.

We got shut out 7 times. I don't think it's any big revelation to anybody why this season sucked. We lost scoring, we needed to replace that lost scoring, and it didn't happen.

[KeithK]

There is a downward trend, but it's not that strong.  The slope is -0.053 goals per year with a R2 of 0.3351.  If I start the data set in 2008 (where visually I might see a little change in the trend) the slope goes to -0.11 with a R2 of 0.5024.  (Excel "statistics".)  This year strongly affects these results, doubling the manitude of the slope in both cases. We need more data to see whether this year was an outlier or not.

Of course, what really matters is goals relative to the rest of the NCAA.  But you didn't give me that data to play with.

[BearLover]

Notably, there is a general downward trend.  Also notably, 2015 is stunningly low--by far the worst Cornell has ever been at scoring under Schafer.  So I don't retract my statement that we are scoring at way too low a rate to be close to competitive on a national level.  This year may have been aberrational--but last year was equally aberrational in that we almost made the tournament despite being +3.  2009 was the only year we made the tournament without scoring at least one more GPG.

Well, geez. Here's the thing...I can never tell about which timeframe you're making each of your arguments. Sometimes it's over a 4-year span. Sometimes 10. Sometimes its the selected range where Yale was kicking ass. Sometimes, like this instance, it's one season.

We got shut out 7 times. I don't think it's any big revelation to anybody why this season sucked. We lost scoring, we needed to replace that lost scoring, and it didn't happen.

Well, yeah, I guess I can be vague, but you can generally assume that when I talk about lengths of time, I am referring to all recent years, with each year closer to the present weighted more heavily than the last.  So here I was referring to all recent years, but with the most emphasis on 2015, and next most on 2014, etc.  

There is a downward trend, but it's not that strong. The slope is -0.053 goals per year with a R2 of 0.3351. If I start the data set in 2008 (where visually I might see a little change in the trend) the slope goes to -0.11 with a R2 of 0.5024. (Excel "statistics".)

You mean goals per game per year, right?  Assuming the rest of the NCAA is constant, even the first number is a fairly strong trend, no?

[KeithK]

There is a downward trend, but it's not that strong. The slope is -0.053 goals per year with a R2 of 0.3351. If I start the data set in 2008 (where visually I might see a little change in the trend) the slope goes to -0.11 with a R2 of 0.5024. (Excel "statistics".)

You mean goals per game per year, right?  Assuming the rest of the NCAA is constant, even the first number is a fairly strong trend, no?

Yes, goals per game per year ( i just fit the numbers you posted).

I'm not sure one twentieth of a goal is that significant given the noise. The 2008-15 number is much more so.  Of course, we don't know (unil someone bothers to look) whether the rest of the league is static.

[pfibiger]

There is a downward trend, but it's not that strong. The slope is -0.053 goals per year with a R2 of 0.3351. If I start the data set in 2008 (where visually I might see a little change in the trend) the slope goes to -0.11 with a R2 of 0.5024. (Excel "statistics".)

You mean goals per game per year, right?  Assuming the rest of the NCAA is constant, even the first number is a fairly strong trend, no?

Yes, goals per game per year ( i just fit the numbers you posted).

I'm not sure one twentieth of a goal is that significant given the noise. The 2008-15 number is much more so.  Of course, we don't know (unil someone bothers to look) whether the rest of the league is static.

collegehockeystats.net only has all of D1 starting in '02-'03 ('02 is missing hockey east, and before that i'd have to pull each conference individually, which i'm not gonna do. so feel free to excluse '02 if you want.)

NCAA Average Scoring Offense (gpg):

2002: 3.06
2003: 3.12
2004: 2.90
2005: 2.87
2006: 2.91
2007: 2.88
2008: 2.74
2009: 2.79
2010: 2.92
2011: 2.91
2012: 2.79
2013: 2.72
2014: 2.80
2015: 2.69

[Trotsky]

Notably, there is a general downward trend.  

Full history.

IMHO, this is a far more important stat.

[pfibiger]

2002-2015 scoring offense, cornell vs. d-1 average:

[ugarte]

From what I saw, Cornell's problem this year wasn't "scoring" exactly, it was an overall inability to use sticks. The passing was inaccurate, so it is no surprise that they also couldn't hit the net. The end boards at Lynah took more abuse this year than they probably have in years.

[RichH]

2002-2015 scoring offense, cornell vs. d-1 average

Thanks very much for doing this, pfibiger. That's the exact data I've craved this week, but didn't have the time/patience to find myself.

[pfibiger]

2002-2015 scoring offense, cornell vs. d-1 average

Thanks very much for doing this, pfibiger. That's the exact data I've craved this week, but didn't have the time/patience to find myself.

Be interesting to look at

1) scoring differential vs. just scoring and
2) redoing the graph above with only tourney teams included in the average set.

If anyone wants the set of excel sheets to monkey around w/ this, let me know