Some performances are downright offensive. At times, last season was offensive in all the wrong ways. The season culminated in the dually most and least offensive moment of the season: a 7-0 humbling to Union at home in the playoffs. The scoring production hit rock bottom. The usually rare disillusionment of the Lynah Faithful soared for the first time in a decade.
This is not to say that last season was without its memorable moments and high notes (think second win in the Frozen Apple, historic defeat of Denver, and two emotional Harvard games). None can erase what lies ahead of this coming season. The Cornell hockey program will be on a path of proving next season. All the questions with few exceptions encircle addressing the Red's offensive woes throughout the entirety of last season. In a multipart series, this contributor of Where Angels Fears to Tread will address the ways in which the coming team will redress this fever-inducing problem.
This is not to say that last season was without its memorable moments and high notes (think second win in the Frozen Apple, historic defeat of Denver, and two emotional Harvard games). None can erase what lies ahead of this coming season. The Cornell hockey program will be on a path of proving next season. All the questions with few exceptions encircle addressing the Red's offensive woes throughout the entirety of last season. In a multipart series, this contributor of Where Angels Fears to Tread will address the ways in which the coming team will redress this fever-inducing problem.
The Team
Readers, here it is, the last piece dedicated exclusively to predicting through statistical prognostication from where the offense for the Cornell hockey team will come and at which level it can be expected to produce during the 2015-16 season.
The first entrant in this series takes aim at the newcomers. That piece predicts to what degree freshmen will erase offensive losses to graduation and ease the scoring woes of the Big-Red program. The second entry takes a different approach. It chooses to highlight the integral role that offensive-minded defensemen play in all of Cornell hockey's championship teams. In this way, it is both carrot in promising possible immortality and stick in placing pressure on the carnelian and white's talented blue liners. The penultimate installment addresses the role of the sophomore and senior classes. The development of those classes from their previous season historically and statistically provides the greatest reservoir for internal scoring growth. This writer concludes that if those classes improve by average increments, Cornell's offense will remain deficient even compared to that of last season.
These chapters of this statistical saga are incomplete. Cornell will not play games as mere classes. The freshmen, sophomores, and seniors will not lace up against an opponent's entire roster without the aid of other classes. The blue liners will not stare down and rally answers alone to a challenger's salvos. And, anyways, what of the junior class? The Red will play as a unit. How will this Cornell aggregation enter every contest from Niagara until what will become its final grapple? What essential unit will determine the success or failure of this season?
Bo Schembechler knew the answer: "The Team, The Team, The Team, The Team."
Now, I do not recommend stealing much from Michigan (much less anything from Princeton), but the sentiment rings true across the spans of geography, universities, and sports. So, reader, in this final chapter, let's put the data together and see what we can expect this team for Cornell hockey to play like. The team is our team after all.
Offense is the big concern of everyone. This is justifiable. The biggest concern to outsiders of East Hill is how to close the gap to the offensive contributions that left Central New York to graduation. However, classes are not replaced, players are. Relying on this truism, this writer reduced the statistics of each player from last season's senior class to their goals and points per game totals. When aggregated to determine the contributions of the class as a whole, measures of player-goals and player-points per game emerged.
The reliance on player-goals and player-points per game allows a better reduction of the value that a player conferred while playing than resorting strictly to absolute goals and points. Joel Lowry scored only the fourth-greatest number of goals for the team last season. Injury shortened the star forward's senior campaign. Lowry's goals and points per game led the team last season. His lead over his nearest competitor, Christian Hilbrich, enjoyed a buffer of 12.5% of Hilbrich's output.
One reasonably could value Cornell's offensive losses from Joel Lowry's graduation to only four goals. This contributor believes that choice devalues Lowry's contributions last season. For example, half of the goals that Lowry scored decided games.
Using Lowry as an example indicates how relying on player-goals and player-points per game better approximates the actual contributions lost to graduation. Furthermore, the methodology dually appreciates the offensive losses of the team. Players like Joel Lowry and Joakim Ryan who lost significant percentages of their senior years have their contributions weighted equally with those of their classmates who enjoyed more game action. Specifically, the player-goals per game approach places the Big Red's offensive deficit 41.4% larger than does an approach relying on absolute goals per game lost.
In other words, use of player-goals per game estimates what Cornell's offense ideally would have been last season but for injuries to key personnel. This means that any expected contributions that erase the deficiency already correlate to an anticipated improvement in the Red's offensive prowess relative to its offense of last season because of the 41.4% overestimate. Calculating player-goals and player-points per game metrics for this season's freshman, sophomore, and senior classes is the basis of two entries in this series that approximate what percentage of the contributions of last season's senior class should be erased with average performances.
None of those classes eliminates alone the 0.82 player-goals per game of last season's senior class.
Do not get too worried. Using a model that bases expected freshman-season performances on correlating junior league-specific data of every freshman under Schafer to that player's freshman offensive output, a projection for each of this season's freshman is made. Remember, this is an average based upon several assumptions that depreciate the expected contributions of players like Chad Otterman and Beau Starrett because of prep hockey's limited data availability. The model for freshman scoring predicts that the newcomers's performances will supplant 64.6% of the player-goals per game that Cornell lost.
Another method determines the incremental improvements that upperclassmen should manifest in their playing based upon the data from over 100 players under Coach Schafer. This model, much like the overestimate in offensive debt, is painstakingly conservative to provide readers and this writer with a worst-case scenario. For example, players who experience infinite relative growth between any two consecutive seasons (i.e. going from zero goals scored to a non-zero goals scored season) have their data removed for that increment. This preserves the conservatism of the analysis.
The model divined that the sophomore and senior classes on the 2015-16 team will account respectively for 47.6% and 96.3% of the 0.82 player-goals per game lost. So, among three classes, no class is expected with historically average contribution to account for much less than half of the Red's lost offense. Two classes are expected to make up large majorities of the losses.
Remember the words of Bo. Schafer is not trying to replace a class as commentators, critics, and fans will have you believe. The coach is building a team. So, reader, compare likes to likes with last season's team to what is expected of this coming season's team. Multiple sets of data and their associated expectations lie before us. Let's put 'em together and see what we get. What exactly compromises this team's bibbidi-bobbidi-boo?
Wait. There are four classes on a team. What about those juniors? If one applies the same approach for determining upperclassman improvement to the data from last season's sophomores to project their contributions as this season's juniors based upon average relative scoring changes between sophomore and junior years, the role of the junior class becomes clear. The juniors alone are expected to erase the deficit of player-goals per game of last season's seniors.
The efforts of Holden Anderson, Matt Buckles, Eric Freschi, Jeff Kubiak, Patrick McCarron, Gavin Stoick, and Jake Weidner should deliver Cornell with 0.85 player-goals per game. So, there it is, the Rosetta Stone that can translate Cornell's incumbent talent into a way of eclipsing the scoring lost to graduation for which all media profess pursuit.
The model across all three related pieces projects the team, this Cornell hockey team for the 2015-16 season, to produce 2.56 player-goals per game. The team's immediate predecessor totaled only 2.28 player-goals per game. A model ripe with conservative estimates to provide worst- and average-case scenarios predicts that this season's team will be 12.3% more lethal at scoring than was the team that represented Cornell last season.
An improvement of that scale to Cornell's offense last season would have given the Big Red seven more goals. Eleven games over the course of last season saw the carnelian and white fail to win by one goal or less. Seven additional goals might have decided six tied contests in the Big Red's favor which would have given Cornell a win total of 17, identical to the win total of the 2013-14 season.
Player-goals per game and team-goals per game do not exactly correlate. Some intellectual massaging relates the two. Assuming that catastrophic injuries do not plague a team, player-goals per game and team-goals per game will approach one another. Additionally, the more productive players whose production in a player-goals per game model is weighted equally with the production of less productive players contribute disproportionately over a season which forces team-goals per game to drift. In reality, more productive players will see more game time which allows their contributions to push upward on the team-goals per game without throttling while less ice time will mitigate the lower production rate of other players. These factors reflect why player-goals per game for a team can serve as a crude surrogate for estimating team-goals per game.
If Cornell is able to parlay a conservative estimate of latent talent of 2.56 player-goals per game into a 2.56 team-goals per game for the season, the outlook of the season may be very different than naysayers predict. Cornell would hit its highest offensive output in three seasons, surpass the production of the Red's last run to the ECAC Hockey championship game, and tie for third-best in the last ten seasons. The Big Red's offense would remain depressed more than 10% relative to when Cornell defeated Michigan in the 2012 NCAA tournament. A deficit of nearly 25% of offensive potency would separate this expected scoring rate from that of the last team that brought Cornell a postseason crown in 2010.
A conservative model predicts that Cornell's offense should be improved by a not insignificant margin this season. Freshmen, sophomores, and seniors should become role players. Statistically, it is the juniors who should be the tip of the Red's spear. Another edged weapon is present in these data; a double-edged sword. The Red's offense should approach the scoring touch of the 2010-11 team that carried Cornell to its last Eastern title-game appearance but fall short of possessing the weapons of the 2010 and 2012 postseason that won a title and defeated Michigan, respectively.
This is what one should be able to expect. The task falls to the team to rise above the base, average projections of this contributor's model if its members want to grasp greatness. Statistics indicate that it should be a good year. It is up to the players and coaches to mature this team's latent talent to give the Lynah Faithful a great year.
The good news? All this prognostication becomes moot the moment the puck strikes the ice at Dwyer Arena. The bad news? Mootness erases the relevance of all the hope that these projections may imbue.
The first entrant in this series takes aim at the newcomers. That piece predicts to what degree freshmen will erase offensive losses to graduation and ease the scoring woes of the Big-Red program. The second entry takes a different approach. It chooses to highlight the integral role that offensive-minded defensemen play in all of Cornell hockey's championship teams. In this way, it is both carrot in promising possible immortality and stick in placing pressure on the carnelian and white's talented blue liners. The penultimate installment addresses the role of the sophomore and senior classes. The development of those classes from their previous season historically and statistically provides the greatest reservoir for internal scoring growth. This writer concludes that if those classes improve by average increments, Cornell's offense will remain deficient even compared to that of last season.
These chapters of this statistical saga are incomplete. Cornell will not play games as mere classes. The freshmen, sophomores, and seniors will not lace up against an opponent's entire roster without the aid of other classes. The blue liners will not stare down and rally answers alone to a challenger's salvos. And, anyways, what of the junior class? The Red will play as a unit. How will this Cornell aggregation enter every contest from Niagara until what will become its final grapple? What essential unit will determine the success or failure of this season?
Bo Schembechler knew the answer: "The Team, The Team, The Team, The Team."
Now, I do not recommend stealing much from Michigan (much less anything from Princeton), but the sentiment rings true across the spans of geography, universities, and sports. So, reader, in this final chapter, let's put the data together and see what we can expect this team for Cornell hockey to play like. The team is our team after all.
Offense is the big concern of everyone. This is justifiable. The biggest concern to outsiders of East Hill is how to close the gap to the offensive contributions that left Central New York to graduation. However, classes are not replaced, players are. Relying on this truism, this writer reduced the statistics of each player from last season's senior class to their goals and points per game totals. When aggregated to determine the contributions of the class as a whole, measures of player-goals and player-points per game emerged.
The reliance on player-goals and player-points per game allows a better reduction of the value that a player conferred while playing than resorting strictly to absolute goals and points. Joel Lowry scored only the fourth-greatest number of goals for the team last season. Injury shortened the star forward's senior campaign. Lowry's goals and points per game led the team last season. His lead over his nearest competitor, Christian Hilbrich, enjoyed a buffer of 12.5% of Hilbrich's output.
One reasonably could value Cornell's offensive losses from Joel Lowry's graduation to only four goals. This contributor believes that choice devalues Lowry's contributions last season. For example, half of the goals that Lowry scored decided games.
Using Lowry as an example indicates how relying on player-goals and player-points per game better approximates the actual contributions lost to graduation. Furthermore, the methodology dually appreciates the offensive losses of the team. Players like Joel Lowry and Joakim Ryan who lost significant percentages of their senior years have their contributions weighted equally with those of their classmates who enjoyed more game action. Specifically, the player-goals per game approach places the Big Red's offensive deficit 41.4% larger than does an approach relying on absolute goals per game lost.
In other words, use of player-goals per game estimates what Cornell's offense ideally would have been last season but for injuries to key personnel. This means that any expected contributions that erase the deficiency already correlate to an anticipated improvement in the Red's offensive prowess relative to its offense of last season because of the 41.4% overestimate. Calculating player-goals and player-points per game metrics for this season's freshman, sophomore, and senior classes is the basis of two entries in this series that approximate what percentage of the contributions of last season's senior class should be erased with average performances.
None of those classes eliminates alone the 0.82 player-goals per game of last season's senior class.
Do not get too worried. Using a model that bases expected freshman-season performances on correlating junior league-specific data of every freshman under Schafer to that player's freshman offensive output, a projection for each of this season's freshman is made. Remember, this is an average based upon several assumptions that depreciate the expected contributions of players like Chad Otterman and Beau Starrett because of prep hockey's limited data availability. The model for freshman scoring predicts that the newcomers's performances will supplant 64.6% of the player-goals per game that Cornell lost.
Another method determines the incremental improvements that upperclassmen should manifest in their playing based upon the data from over 100 players under Coach Schafer. This model, much like the overestimate in offensive debt, is painstakingly conservative to provide readers and this writer with a worst-case scenario. For example, players who experience infinite relative growth between any two consecutive seasons (i.e. going from zero goals scored to a non-zero goals scored season) have their data removed for that increment. This preserves the conservatism of the analysis.
The model divined that the sophomore and senior classes on the 2015-16 team will account respectively for 47.6% and 96.3% of the 0.82 player-goals per game lost. So, among three classes, no class is expected with historically average contribution to account for much less than half of the Red's lost offense. Two classes are expected to make up large majorities of the losses.
Remember the words of Bo. Schafer is not trying to replace a class as commentators, critics, and fans will have you believe. The coach is building a team. So, reader, compare likes to likes with last season's team to what is expected of this coming season's team. Multiple sets of data and their associated expectations lie before us. Let's put 'em together and see what we get. What exactly compromises this team's bibbidi-bobbidi-boo?
Wait. There are four classes on a team. What about those juniors? If one applies the same approach for determining upperclassman improvement to the data from last season's sophomores to project their contributions as this season's juniors based upon average relative scoring changes between sophomore and junior years, the role of the junior class becomes clear. The juniors alone are expected to erase the deficit of player-goals per game of last season's seniors.
The efforts of Holden Anderson, Matt Buckles, Eric Freschi, Jeff Kubiak, Patrick McCarron, Gavin Stoick, and Jake Weidner should deliver Cornell with 0.85 player-goals per game. So, there it is, the Rosetta Stone that can translate Cornell's incumbent talent into a way of eclipsing the scoring lost to graduation for which all media profess pursuit.
The model across all three related pieces projects the team, this Cornell hockey team for the 2015-16 season, to produce 2.56 player-goals per game. The team's immediate predecessor totaled only 2.28 player-goals per game. A model ripe with conservative estimates to provide worst- and average-case scenarios predicts that this season's team will be 12.3% more lethal at scoring than was the team that represented Cornell last season.
An improvement of that scale to Cornell's offense last season would have given the Big Red seven more goals. Eleven games over the course of last season saw the carnelian and white fail to win by one goal or less. Seven additional goals might have decided six tied contests in the Big Red's favor which would have given Cornell a win total of 17, identical to the win total of the 2013-14 season.
Player-goals per game and team-goals per game do not exactly correlate. Some intellectual massaging relates the two. Assuming that catastrophic injuries do not plague a team, player-goals per game and team-goals per game will approach one another. Additionally, the more productive players whose production in a player-goals per game model is weighted equally with the production of less productive players contribute disproportionately over a season which forces team-goals per game to drift. In reality, more productive players will see more game time which allows their contributions to push upward on the team-goals per game without throttling while less ice time will mitigate the lower production rate of other players. These factors reflect why player-goals per game for a team can serve as a crude surrogate for estimating team-goals per game.
If Cornell is able to parlay a conservative estimate of latent talent of 2.56 player-goals per game into a 2.56 team-goals per game for the season, the outlook of the season may be very different than naysayers predict. Cornell would hit its highest offensive output in three seasons, surpass the production of the Red's last run to the ECAC Hockey championship game, and tie for third-best in the last ten seasons. The Big Red's offense would remain depressed more than 10% relative to when Cornell defeated Michigan in the 2012 NCAA tournament. A deficit of nearly 25% of offensive potency would separate this expected scoring rate from that of the last team that brought Cornell a postseason crown in 2010.
A conservative model predicts that Cornell's offense should be improved by a not insignificant margin this season. Freshmen, sophomores, and seniors should become role players. Statistically, it is the juniors who should be the tip of the Red's spear. Another edged weapon is present in these data; a double-edged sword. The Red's offense should approach the scoring touch of the 2010-11 team that carried Cornell to its last Eastern title-game appearance but fall short of possessing the weapons of the 2010 and 2012 postseason that won a title and defeated Michigan, respectively.
This is what one should be able to expect. The task falls to the team to rise above the base, average projections of this contributor's model if its members want to grasp greatness. Statistics indicate that it should be a good year. It is up to the players and coaches to mature this team's latent talent to give the Lynah Faithful a great year.
The good news? All this prognostication becomes moot the moment the puck strikes the ice at Dwyer Arena. The bad news? Mootness erases the relevance of all the hope that these projections may imbue.
RSS Feed
