OCinBuffalo Posted August 30, 2006 Share Posted August 30, 2006 Now this post is a clear case of congruence between a man's words and his avatar! 755934[/snapback] Nah more like I NEEDED to take a piss right then. Link to comment Share on other sites More sharing options...
Orton's Arm Posted August 30, 2006 Author Share Posted August 30, 2006 Anyway, your analogy is oversimplified in the sense that it does not acount for "meaningful" participation of the ninth graders, or the seventh graders for that matter. What if, like the NFL, some of those ninth grader get hurt and can't play? How about, like the NFL, some of those ninth graders do well the first year but get detention the next and can't play? What if a team with 4 ninth graders ends up benching them all and still wins the championship? Did they win because of ninth graders? The original question posed is not about general relationships but DEFINITIVE evidence that high o line drafting =(this means is equal to not might or maybe) wins/playoffs. There's a technical term for the factors you've described in this paragraph: noise. There's also a signal: the fact that the average ninth grader is better than the average seventh grader. Given a large enough sample, the signal can usually penetrate the noise; even without any attempts at noise correction. In fact, misguided attempts to correct for noise can actually interfere with proper random sampling technique. I'm not saying that's an issue with the study you've suggested. But other studies obtained flawed results by trying to obtain balanced samples with fixed percentages of specific demographic groups, instead of just taking a nice, random sample like they should have been doing. Link to comment Share on other sites More sharing options...
Orton's Arm Posted August 30, 2006 Author Share Posted August 30, 2006 That's right(& supports my point).Before him, the team was (on average) in need of someone at that OL position. Not only is that position filled successfully in your scenarios but filled successfully with 1st day talent so (on average) one can expect a higher upside to the player. Now we have a (on average?) close to pro bowl player where previously there was a (on average) player you really wanted to replace. All you are proving is the equivalent of 'guys like lots of sex'. You argue your case eloquently, and that bit about the quarterbacks was especially convincing. Let me mull this over . . . Link to comment Share on other sites More sharing options...
OCinBuffalo Posted August 30, 2006 Share Posted August 30, 2006 You argue your case eloquently, and that bit about the quarterbacks was especially convincing. Let me mull this over . . . 755940[/snapback] Why is it than when Dibs says it you say "mull it over" and when I say it you say noise/signal(btw I am a business intelligence expert - so I know all the technical terms)? DIBS AND I ARE SAYING THE SAME THING - just differently. The data has already been sampled - we are using the O lineman in the NFL!?!?! You cannot escape my conclusion by running to your stats/marketing/psychology text book - nice try though . In this case, Meaningful performance must be evaluated equally for each lineman. Those whose data does not meet the criteria are IRRELEVANT and should be rejected as an OUTLIER(see I can use fancy stats terms too ). Example: If a team drafts an O lineman 1st round and that guy: get's hurt/get's in trouble/sucks, doesn't play, but his team wins the Super Bowl - you are saying it doesn't matter since there is still a corellation(you contrived) between high draft picks and wins. This guy should still count(and convienently continue to support your corellation) because his team won the super bowl without him? I would think that it would matter very much to that team's fans since they would be wondering why the GM didn't do his homework. It would also be exceedingly meaningful to the guy that replaced your 1st round draft pick come contract time Moreover, you would have that same GM draft O lineman high again next year because of your "corellation". This is silly. Link to comment Share on other sites More sharing options...
Orton's Arm Posted August 31, 2006 Author Share Posted August 31, 2006 Why is it than when Dibs says it you say "mull it over" and when I say it you say noise/signal(btw I am a business intelligence expert - so I know all the technical terms)? DIBS AND I ARE SAYING THE SAME THING - just differently. The data has already been sampled - we are using the O lineman in the NFL!?!?! You cannot escape my conclusion by running to your stats/marketing/psychology text book - nice try though . In this case, Meaningful performance must be evaluated equally for each lineman. The point Dibs made is that in general, you'd expect early round success stories to correlate with winning games. The fact I've shown that early round offensive line success stories correlate with winning games may therefore be irrelevant. In other words, Dibs is saying I've answered the wrong question. This may or may not be true, and I'll probably have to do another regression on, for example, linebackers to see if Dibs' objections are valid. If the same correlations I observed for first day starting linemen also exist for first day starting linebackers, Dibs is correct in saying that I've discovered men like to have sex. If Dibs focused on the big picture of whether I've asked the right question, you have focused a little more on which sampling techniques I should be using, whether I should be throwing out outliers, and other details about the methodology. But before you can even begin discussing methodology, you have to figure out exactly what question it is you want answered, and why. My methodology was correct if the question you're interested in is, "Do teams which fill their offensive line starting slots via the first day of the draft tend to win more games?" You suggest meaningful performance evaluations be given to each lineman. What question are you hoping to answer by doing so? The success rates of the various rounds are already known. In regression part 2, I showed that having first day picks as starting OL correlates strongly with wins. In part 1, I sort of showed that merely picking offensive linemen early does not correlate with wins. The next question worth asking is whether using the first day of the draft to fill a non-OL position (such as linebacker) correlates with winning. Link to comment Share on other sites More sharing options...
Dibs Posted August 31, 2006 Share Posted August 31, 2006 Another factor which is very hard to quantify but I feel is very relevent is.... I've noticed that often, when a team picks a successful 1st day pick, they can also 'get on a roll' & have success year after year with said picks. If there is some merit to this, this would skew the results even further due to the compound nature involved. Examples: Ravens 96-03 Ogden, R.Lewis, Boulware, Sharper, McAlister, J. Lewis, Heap, Reed, Suggs Colts 96-03 Harrison, Tarik Glenn, Manning, E. James, Peterson, M.Washington, R. Wayne, Freeney, Tripplett, Doss Some of the players might get injured or move on but you get the idea. Link to comment Share on other sites More sharing options...
Chilly Posted August 31, 2006 Share Posted August 31, 2006 There's a technical term for the factors you've described in this paragraph: noise. There's also a signal: the fact that the average ninth grader is better than the average seventh grader. Given a large enough sample, the signal can usually penetrate the noise; even without any attempts at noise correction. In fact, misguided attempts to correct for noise can actually interfere with proper random sampling technique. I'm not saying that's an issue with the study you've suggested. But other studies obtained flawed results by trying to obtain balanced samples with fixed percentages of specific demographic groups, instead of just taking a nice, random sample like they should have been doing. 755939[/snapback] Heh, thats one of the things I thought was bad about your analysis. Its not big enough to take all of the other factors into account eh? Link to comment Share on other sites More sharing options...
Recommended Posts