poorgriz said:
PlayerRep said:
poorgriz said:
Still waiting for a response to my question earlier, quoted in this post for convenience. I'm assuming you are punting.
"Really? Are you still trying to argue this or are you guys just playing lawyer and picking a side to argue just for fun... when you really don't seriously believe in what you're arguing? You have thrown out the blanket statement that six data points are always better than two. Do you really believe that, regardless of what those data points are? Are you saying that if two FCS teams have played six games each, one team played two FCS schools and four NAIA schools, and the other FCS team played two FCS schools and four FBS schools... that those stats would hold any relevance whatsoever?"
Again, I am saying that stats for the first 6 games of the Big Sky conference teams are more valid than conference only stats for 2 games?
Let me ask you a question. Are you saying that stats for 2 conference games are more valid than stats for the first 6 games of the season? That is the heart of the point that was being discussed in this thread.
No... you started by saying that, but then that led to your blanket statement saying that 6 data points are always more valid than 2. That's simply not true in all cases, per my comment above. Therefore, it all depends on the quality of the opponents within your sample. Lastly, my answer would be YES, in this particular situation it would be more accurate to look at only two BSC conference games than it would 6 games, which includes a mix match of NAIA, D2, FCS, and FBS schools. If all games were FCS I'd be with you. Having said all that... NEITHER option is an accurate reflection of where teams are at yet. :lol:
Nice to see someone gets it, even if they are a Cat fan. PR doesn't understand the concept of true statistics, and the sampling variable. He's to hung up on 6 games or 2 games. It has nothing to do with the number of games, if you are doing a true comparison. It has to do with the characteristics of the sample you are using.
Thus if I am doing a study on hamburgers at drive through restaurants, I would not include Chic-Fil-A simply because they are a drive through business. Relating back to football, because the sample size for non-conference games varies so widely (teams playing down vs teams playing up, some 2 FBS games, some none), then in its truest sense only conference statistics make sense. I could careless if its even 1 game, and by the end of the season the best overall statistics to use will still be conference stats. This is sampling in its pure form.
But I understand, he's a lawyer that likes to pretend he stayed at every Holiday Inn every night of the year, and there is nothing he doesnt know or can't do. But we all know what he is like, be we Griz, Cat, Eagle or any fan that frequents this board. He's overthinking the process....and missing the most basic concept.