I am consided pretty good at using stat's to analyze trends & develop models. (snip) I do stand behind the model that...
Who considers you pretty good? Can't be anyone here... you spew out copious amounts of weak crap to a bunch of guys that have WAAAAY more experience than you. i.e. You're not bringing anything new to the table and you're offending people by assuming they don't know what they are talking about and haven't tried the things you bring up. Anyway, stand by your hypothesis all you like, smokey, but here's some other stats for you to digest:
- 9218-4162 (0.689)
- Playoff Appearances
- Finals Appearances
I kinda have a little experience to back up my hypotheses... in fact, I've collected enough data over said amount of time to change those hypotheses to theories (and some of those theories to proven facts).
"I do most of my trend modeling from the differential (you vs opponents) stat's during the a league."
I find that you now are speaking about differentials as being what you pay attention to when you clearly were speaking in a completely different direction earlier (paraphrased: you can't average more than 35 assists per game so you shouldn't go higher than that. You can't get more than 55 rebounds per game, so you shouldn't worry about more). Nice change of story.
Oh... and one other thing:
"One thing I'm sure of is that stat's are a slippery road."
Hi. This is a simulation engine that uses stats (no apostrophe) to determine results.
I'll see your 30 years of science and raise you 5 years of (winning) experience in the subject at hand. I won't tell you how to make drugs since I clearly have no knowledge about the subject. You shouldn't tell people how to make teams for the same reason.
edit: A prog league as your example? Really?! And one from the 70s? Wow. I'm done wasting time responding to you.
3/27/2013 8:36 PM (edited)