The CardBoard
A question for the stat folks - Printable Version

+- The CardBoard (https://thecardboard.org/board)
+-- Forum: C-House! (https://thecardboard.org/board/forum-4.html)
+--- Forum: The CardBoard (https://thecardboard.org/board/forum-5.html)
+--- Thread: A question for the stat folks (/thread-6223.html)


A question for the stat folks - washingtonismoney - 12-01-2013

When's the last time a team beat five ranked teams at home in one season? Heck, when's the last time a team faced five ranked team at home in one season? We may have just had a home season for the ages...

Re: A question for the stat folks - Gauss - 12-02-2013

Sorry washingtonismoney, but this is a pet peeve of mine: we only beat 3 currently ranked teams at home (Washington and Notre  Dame are not ranked anymore).
At the beginning of the season some teams start seriously overranked, and others are underrated. I wish ESPN didn't confuse people with the old rankings on the schedule (which have their own use, like understanding the significance of the game at the time, of course).

That said, we really had a favorable schedule, with the most difficult teams at home.
That, plus the returning talent on the lines, were the main reasons why many of us thought that we had a (small) shot for an undefeated season. USC, and perhaps OSU, were the only road games that worried me, and they were not Oregon...
Oh well, let's see what we have in 2015.

Re: A question for the stat folks - Renaissance Man - 12-02-2013

(12-02-2013, 01:01 AM)Gauss link Wrote:Sorry washingtonismoney, but this is a pet peeve of mine: we only beat 3 currently ranked teams at home (Washington and Notre  Dame are not ranked anymore).
At the beginning of the season some teams start seriously overranked, and others are underrated. I wish ESPN didn't confuse people with the old rankings on the schedule (which have their own use, like understanding the significance of the game at the time, of course).

That said, we really had a favorable schedule, with the most difficult teams at home.
That, plus the returning talent on the lines, were the main reasons why many of us thought that we had a (small) shot for an undefeated season. University of Spoiled Children , and perhaps OSU, were the only road games that worried me, and they were not Oregon...

Of course, we do not play in the SEC, where a team dropping out of the top 25 simply means that two others will be moved in to take their place, so Stanford is very likely to beat teams that are not ranked later simply because they will have the loss to Stanford, along with a brutal conference slate, and will not have 4 home games vs. FCS opponents to pad their record.
Oh well, let's see what we have in 2015.

Re: A question for the stat folks - washingtonismoney - 12-02-2013

(12-02-2013, 01:01 AM)Gauss link Wrote:Sorry washingtonismoney, but this is a pet peeve of mine: we only beat 3 currently ranked teams at home (Washington and Notre  Dame are not ranked anymore).
At the beginning of the season some teams start seriously overranked, and others are underrated. I wish ESPN didn't confuse people with the old rankings on the schedule (which have their own use, like understanding the significance of the game at the time, of course).

No, I disagree. I don't think it's automatically appropriate to assume the later rankings are the better ones. Take Georgia. They started out the season ranked #5 and Clemson beat them. Obviously they're unranked now, but does that discredit Clemson's home win? I don't think so -- obviously Georgia suffered through one of the worst waves of injuries I've seen in a while in college football. We can't just casually assume that in every instance more data clarifies the situation.

At any rate, most computer rating systems agree we played one of the toughest schedules in the country -- top ten for sure -- so it's not as if a naive look at the schedule is totally unfounded, even if you subscribe to the belief that the later rankings are automatically a more accurate representation of a team's strength in other, past moments.

Re: A question for the stat folks - Gauss - 12-02-2013

(12-02-2013, 09:48 AM)washingtonismoney link Wrote:No, I disagree. I don't think it's automatically appropriate to assume the later rankings are the better ones. Take Georgia. They started out the season ranked #5 and Clemson beat them. Obviously they're unranked now, but does that discredit Clemson's home win? I don't think so -- obviously Georgia suffered through one of the worst waves of injuries I've seen in a while in college football. We can't just casually assume that in every instance more data clarifies the situation.

At any rate, most computer rating systems agree we played one of the toughest schedules in the country -- top ten for sure -- so it's not as if a naive look at the schedule is totally unfounded, even if you subscribe to the belief that the later rankings are automatically a more accurate representation of a team's strength in other, past moments.

Oh, I was by no means arguing that our schedule had not been tough, and even my point about Washington and Notre.Dame not being ranked I uttered half-heartedly: those teams are by most measures top-30 teams, so not far from being ranked at all.

There are indeed a few cases where a team's strength varies dramatically over the season, in which an accurate early ranking would be better. But, by and large, current rankings are more accurate than rankings that include less information.

In my book, Florida State cannot claim having beaten the #7 team in the nation (Miami), or Oklahoma having beaten the #10 (Texas Tech).

Re: A question for the stat folks - Oasis - 12-02-2013

What I find irksome is the SEC practice of scheduling teams like Furman (LSU), Georgia State, Chattanooga (Bama), Appalachian State (Georgia) and Idaho (Ole Miss).  The goal is just to avoid losses.  It doesn't always work (Georgia Southern beat Florida), but it's nothing but gaming the system.  And although I know the patsies don't have to agree to play SEC teams, I still feel sorry for the players who have to go in and usually (not always!) get massacred by a team they shouldn't be playing and wouldn't except for the money.  It's just bad.

Stanford's out-of-conference schedule is SJS (a traditional Bay Area rival and no patsy), Army (completely appropriate because like us they select football players seriously for academic ability) and the Hibernians (who would be shot up somewhere in the polls if they had beaten us).

Re: A question for the stat folks - Spiny_Norman - 12-02-2013

(12-02-2013, 10:52 AM)Oasis link Wrote:What I find irksome is the SEC practice of scheduling teams like Furman (LSU), Georgia State, Chattanooga (Bama), Appalachian State (Georgia) and Idaho (Ole Miss).  The goal is just to avoid losses.

The goal is also to play 4 out of conference home games or 3 at home plus one neutral site.  Most top level SEC teams play only 4 true road games each season, another major advantage for them.