kook_10 1,705 Report post Posted February 18, 2009 i stand corrected. i forgot how much ice time Marty actually got in his last season in Detroit. Share this post Link to post Share on other sites
eva unit zero 271 Report post Posted February 19, 2009 (edited) Lapointe's Red Wing linemates went with him as part of his signing that contract? Edited February 19, 2009 by eva unit zero Share this post Link to post Share on other sites
kook_10 1,705 Report post Posted February 19, 2009 no, but perhaps the word ideally in the question was confusing to some Share this post Link to post Share on other sites
Wombat 26 Report post Posted February 19, 2009 (edited) Of course it doesn't. Its not a perfect model. But this is based on statistics and hypotheticals, and the points rendered here. Hmm, lets see, that Hudler would produce less with more has been proved statistically null, and similarly in terms of his continual PPM drop-off. The fact that statistics suggest that Hudler WOULD produce more supports my argument - which is merely opinion as much as yours. But because the statistics support my opinion, and not yours, I'm not supposed to use them? The only reason you're deducing what I'm saying to irrational levels and presumptions is because they go against what you are saying. One of the first things you learn in a science or research methods class is that you work to prove a hypothesis wrong, not right. What I just did was use all the statistical evidence that we have of Hudler, and it did not prove my point wrong. Did it prove your point wrong either? No. But theres a reason why half of the popular research articles you read on the internet start with phrases such as "research shows..." or "The statistics suggest..." Going by all the evidence you half, you make solid theories based on them. Which exactly what I've done. This is just a sensitive subject for the people who want to assume that Hudler is destined to be an obsolete NHLer, yet have no statistical evidence to back it up, and must rely on biases to carry them. You need not give me a lesson in research methods or statistics... but I'm glad you came at it that way as it forced me to actually crunch the numbers. I took Hudler's game stats from ESPN.com and separated the data into 5 classes based on TOI. To try to get as many games as possible into each class, I set it up as: Under 12 Minutes - 11 games 12-13 minutes - 15 games 13-14 minutes - 9 games 14-15 minutes - 13 games Over 15 minutes - 10 games I feel it is better to use discrete classes as otherwise your statistics are skewed as you include data points in multiple break downs (ie: comparing playing over 15 minutes to playing over 16 minutes, in which all data above 16 minutes is included in both classes). The results: Games playing under 12 minutes - 10 points, average 10.76 TOI, 11.84 Minutes per point Games playing 12-13 minutes - 13 points, average 12.62 TOI, 14.56 Minutes per point Games playing 13-14 minutes - 3 points, average 13.58 TOI, 40.73 Minutes per point Games playing 14-15 minutes - 15 points, average 14.38 TOI, 12.46 Minutes per point Games playing over 15 minutes - 8 points, average 16.51 TOI, 20.64 Minutes per point While we are limited somewhat because of the sample size in each class, what I take from this data is that Hudler is most efficient either playing under 12 minutes a game, or between 14-15 minutes. He takes longer to score a point when he plays over 15 minutes per game. If you look at Hudler's last 22 games (since Jan 1st), he is averaging about 13.9 minutes per game, which is reasonably close to the 14-15 minute classification. In using per game statistics, his 14-15 minute class is where he has his 2nd highest goals per game (only .005 less than his highest class), 2nd highest assists per game (only .03 less than the highest), highest points per game (over a point per game), best +/- per game, fewest PIM per game, and highest SOG per game. Strangely enough, he also has his 2nd lowest PPG per game and 3rd PPA per game in the 14-15 minutes class, which I interpret as meaning that he is getting even more "bang for the buck" in even strength play than in any of the other classes. I think this is an important issue since we therefore cannot "blame" his production on higher PP production... he's just playing excellently in even strength situation in those games. No one is trying to say that if you give Hudler 2 or 3 more minutes a game that his season totals will be less. In a small increment like that, there most probably are not going to be negative outcomes. What we HAVE been trying to say is that Hudler is most efficient with his ice time, in terms of minutes per point scored, when he has less time than more. There is NOT a linear relationship between TOI and minutes per point; therefore, you cannot extrapolate that with added time played he will produce at the same rate. If you are willing to predict that he will score at the same rate playing 18, 19, or 20 minutes, you must also be willing to predict that he will score at the same rate playing 30 or even 60 minutes. That is your line of best fit and is your predictive model. Instead, what you must do is find his peak; in this case, he has two peaks (less than 12 minutes, 14-15 minutes) and his production falls at every other place. Summary: - I do not "hate" Hudler. I've loved his play this year and I am completely happy with his role - I trust that Babcock probably knows that Hudler produces more when his ice time is limited to fewer minutes than Dats, Z, Hossa, et al. - Hudler also suffers in the TOI department because he does not play in 4x4 or PK situations. The other top forwards do. - "The statistics suggest" that Hudler peaks at production around as either under 12 minutes or 14-15 minutes, and more TOI above 15 minutes leads him to have less efficient point production. (By the way, this is what I love about statistics... separate conclusions can be taken from the same data set. There isn't a right answer. It's the same way with message boards. It's a debate and it's fun to have. This has been a fun debate and I hope we can continue it. Have a great day everyone!) - edit: misstated his PPA stat as being 2nd worst when it's actually 3rd, or right in middle of his production of all the classes. Edited February 19, 2009 by Wombat Share this post Link to post Share on other sites
Datsyerberger 279 Report post Posted February 19, 2009 (edited) *snip* Good post, but I've already mostly given up here myself. Some people here just clearly do not understand the difference between total production and most efficient rate of production, and think the former is the end all be all. I'm going to go out on a limb here and guess the games in the 12-13 minute category are 'normal' games, and the ones where he's logging 14-15 are PP heavy scorefests. Hudler seems to thrive in those. Pure speculaton on my behalf, though. And by the way: Lies, damned lies, and... Edited February 19, 2009 by Datsyerberger Share this post Link to post Share on other sites
Wombat 26 Report post Posted February 19, 2009 Good post, but I've already mostly given up here myself. Some people here just clearly do not understand the difference between total production and most efficient rate of production, and think the former is the end all be all. I'm going to go out on a limb here and guess the games in the 12-13 minute category are 'normal' games, and the ones where he's logging 14-15 are PP heavy scorefests. Hudler seems to thrive in those. Pure speculaton on my behalf, though. Hudler doesn't have many of his points in 14-15 games on the PP, so I'm not sure that is how he's cashing in. Unless, the couple extra PP shifts rev him up for 5x5 play and then he scores. Let's keep up the fight for efficiency... we can't be the Bill James' of hockey, but maybe we can be one of his minions! Share this post Link to post Share on other sites
