The Cygarin Ratings has been updated for the Fiesta Bowl between Iowa State and the Oregon.
As has been the case season all season long, I continue to use a composite rating as a basis for the Cygarin Ratings, and then made adjustments to individual teams based on “fit” within the conference standings.
For the Fiesta bowl I have done that for both the Big-12 and the Pac-12 conferences. In both cases, my results are notably different than both polls and power ratings, though the Big-12 has evolved into being more in sync recently. That appears to be a direct result of more data points to evaluate.
The Oklahoma State and Texas bowl games suggest that the Big-12 was indeed a stronger conference than being credited for earlier in the season, probably mostly due to its mid-major non-conference losses as well as early losses by both Oklahoma and Texas that gave critics a false impression of their abilities.
The Pac-12 is woefully short on data points to evaluate, with some teams having only played four conference games. Four games in a twelve team conference make it difficult to “connect” teams to come up with a strength hierarchy within the conference. This is especially true when performances have been inconsistent, making it difficult to determine which performances are normal, and which are outliers.
The Cygarin Ratings has come up with rankings for the Pac-12 teams that is far different than most other ranking systems. What I show is that most teams in the Pac-12 appear to be incredibly average to slightly below average. Indeed, for many teams it is hard to discern any difference between them at all.
I have USC on top and standing alone. Still, the Trojans fail to crack the top-15, coming in 16th. Stanford is second, coming in at 26th, leaving USC as the only top-25 ranked team.
At the bottom end, Arizona comes in as the Pac-12’s “Kansas”, at 111th.
All others fall between 28th and 85th, ranging equal to about West Virginia to Texas Tech. I have no Pac-12 team except USC rated higher than the top five teams in the Big-12.
Interestingly, I have both Colorado and Oregon as way overrated, and sitting solidly in the middle of the pack of the Pac-12…. 5th and 7th respectively, though little difference between them. If Colorado’s game against Texas is any indication, my analysis of the Pac-12 is appearing fairly accurate.
OREGON
Analyzing Oregon is a statistical mess. For every solid positive data point, there is an equally solid bad data point. Even the two trend analyses oppose each other. This is a team that has trend results that look like a shotgun pattern.
To start with, Oregon owns the best performance of the season of any team in the conference by defeating top-16 ranked USC in the Pac-12 CCG. (To compare, Colorado had the second best with a road victory at Stanford, a 20th ranked equivalent.)
However, the Ducks also had the worst two combined losses in the conference, with losses at 85th California and at 82nd Oregon State. Indeed, Oregon and (winless) Arizona are the only two teams to have two loses in the bottom half of the conference strength of schedule.
Trends are also conflicting……
Over the regular season, Oregon was in death spiral, trending steeply down game-to-game from an early win over Stanford (at home, 30th ranked equivalent) in the first conference game of the season, falling consistently game-to-game until losing to both Cal and OSU in their last two regular season games.
Margin-wise this was a fall of nearly 35 points per game. However, Oregon followed those two lowly losses by beating USC in the CCG, a remarkable 30 points improvement over the Ducks’ previous two regular season games.
Conversely, the trend against the strength of schedule is the exact opposite……the better the opponent, the better Oregon played. At the bottom end is two embarrassing losses to Cal an Oregon State, and at the top are admirable wins over Stanford and USC. In the middle is a close, 3-point win over UCLA at home, about a 60th ranked equivalent, nailing their (Cygarin Rating) expected spread.
So, if you eliminate Oregon’s two worst losses, they are a pretty good team. But if you eliminate their two best wins….the Ducks are a gawd awful team without a single top-55 equivalent win.
Indeed, their USC game was the only game that the Oregan came within 14 points of their spread, on either side.
Therein lies the problem with having only five data points to work with, especially when only four of them came in the regular season. It makes it very hard to figure out which games are “normal, which games are “outliers”, and in the case of Oregon….is there even an “normal” to work with. Statistically speaking.…..Oregon has real no bell curve of performance, and instead it offers more of a random scatter diagram.
Forcing an “average”, I have Oregon ranked 41st in the country with a rating of 76, or about the same as Kansas State. But as noted, not only do they have two losses below their average and two wins above their average…they were 14 point or more deviations on both sides.
Indeed, the (very limited) data suggests that Oregon likely will not play “average”, and instead one could more likely expect a VERY poor performance…..or a VERY good performance.
THE GAME
Looking for an outcome game is more about probability that it is predicting an anticipated spread. That is, what is the probability that Oregon will play well enough above their rating to beat Iowa State. (Assuming that Iowa State plays relatively normal.)
After some final tweaking, I have Iowa State as an eleven point favorite (which means nothing, and that translates to Iowa State with a 77% chance of winning.
However, if you flatten the bell curve due to a high standard deviation of Oregon’s performance, that actually improves the Ducks’ chances of winning a bit, leaving Iowa State a 33% chance of winning. In other words, from a little better than 1 in 4 to right about 1 in 3.
My gut tells me that the key factor in this game is my belief Oregon is not a top ranked team.
Additionally, their only real test was against USC in the CCG. To compare, Iowa State played three teams better than USC, and two more teams that approach being equals. That’s a five to one experience factor in Iowa State’s favor, as well as rating credibility.
Futhermore, I put more weight into Oregon’s regular season downward trend with two bad losses at the end…..than I do their David vs. Goliath victory over USC.
Others who watched the game may be able to better explain that Duck victory, but if I understand right, three USC turnovers played a major factor.
Clearly, everyone has found that the best way to try to beat Iowa State is to force Purdy beat us with his arm. If you can do that, especially if you can hold a lead, you stand a good chance that Brock will force passes.
From my perspective, “IF” Oregon is rated as low as I have them, I believe that means that we hopefully be able to depend more on Breece, rather than having to put the game on Brock’s shoulders. Indeed, the conditions appear to be ripe for Breese to have a great day.
In summary….
Oregon has proven that they can beat pretty good teams. But, they have yet to prove they can beat a team as good as Iowa State. To be fair they haven’t yet had the chance.
Oregon’s wild range of performance means that the Ducks are unpredictable no matter who they face. Therefore, I will be thrilled Iowa State’s just winning their first appearance wearing big bowl pants.
But analysis-wise….and I do fear jinxing us…..Iowa State has a 50-50 shot at a double digit win margin, and a pretty good chance at a win similar to what Texas did to Colorado.
So….I’m liking the good guys in this one.
(In my opinion.)
Cygarin Ratings Linky......
As has been the case season all season long, I continue to use a composite rating as a basis for the Cygarin Ratings, and then made adjustments to individual teams based on “fit” within the conference standings.
For the Fiesta bowl I have done that for both the Big-12 and the Pac-12 conferences. In both cases, my results are notably different than both polls and power ratings, though the Big-12 has evolved into being more in sync recently. That appears to be a direct result of more data points to evaluate.
The Oklahoma State and Texas bowl games suggest that the Big-12 was indeed a stronger conference than being credited for earlier in the season, probably mostly due to its mid-major non-conference losses as well as early losses by both Oklahoma and Texas that gave critics a false impression of their abilities.
The Pac-12 is woefully short on data points to evaluate, with some teams having only played four conference games. Four games in a twelve team conference make it difficult to “connect” teams to come up with a strength hierarchy within the conference. This is especially true when performances have been inconsistent, making it difficult to determine which performances are normal, and which are outliers.
The Cygarin Ratings has come up with rankings for the Pac-12 teams that is far different than most other ranking systems. What I show is that most teams in the Pac-12 appear to be incredibly average to slightly below average. Indeed, for many teams it is hard to discern any difference between them at all.
I have USC on top and standing alone. Still, the Trojans fail to crack the top-15, coming in 16th. Stanford is second, coming in at 26th, leaving USC as the only top-25 ranked team.
At the bottom end, Arizona comes in as the Pac-12’s “Kansas”, at 111th.
All others fall between 28th and 85th, ranging equal to about West Virginia to Texas Tech. I have no Pac-12 team except USC rated higher than the top five teams in the Big-12.
Interestingly, I have both Colorado and Oregon as way overrated, and sitting solidly in the middle of the pack of the Pac-12…. 5th and 7th respectively, though little difference between them. If Colorado’s game against Texas is any indication, my analysis of the Pac-12 is appearing fairly accurate.
OREGON
Analyzing Oregon is a statistical mess. For every solid positive data point, there is an equally solid bad data point. Even the two trend analyses oppose each other. This is a team that has trend results that look like a shotgun pattern.
To start with, Oregon owns the best performance of the season of any team in the conference by defeating top-16 ranked USC in the Pac-12 CCG. (To compare, Colorado had the second best with a road victory at Stanford, a 20th ranked equivalent.)
However, the Ducks also had the worst two combined losses in the conference, with losses at 85th California and at 82nd Oregon State. Indeed, Oregon and (winless) Arizona are the only two teams to have two loses in the bottom half of the conference strength of schedule.
Trends are also conflicting……
Over the regular season, Oregon was in death spiral, trending steeply down game-to-game from an early win over Stanford (at home, 30th ranked equivalent) in the first conference game of the season, falling consistently game-to-game until losing to both Cal and OSU in their last two regular season games.
Margin-wise this was a fall of nearly 35 points per game. However, Oregon followed those two lowly losses by beating USC in the CCG, a remarkable 30 points improvement over the Ducks’ previous two regular season games.
Conversely, the trend against the strength of schedule is the exact opposite……the better the opponent, the better Oregon played. At the bottom end is two embarrassing losses to Cal an Oregon State, and at the top are admirable wins over Stanford and USC. In the middle is a close, 3-point win over UCLA at home, about a 60th ranked equivalent, nailing their (Cygarin Rating) expected spread.
So, if you eliminate Oregon’s two worst losses, they are a pretty good team. But if you eliminate their two best wins….the Ducks are a gawd awful team without a single top-55 equivalent win.
Indeed, their USC game was the only game that the Oregan came within 14 points of their spread, on either side.
Therein lies the problem with having only five data points to work with, especially when only four of them came in the regular season. It makes it very hard to figure out which games are “normal, which games are “outliers”, and in the case of Oregon….is there even an “normal” to work with. Statistically speaking.…..Oregon has real no bell curve of performance, and instead it offers more of a random scatter diagram.
Forcing an “average”, I have Oregon ranked 41st in the country with a rating of 76, or about the same as Kansas State. But as noted, not only do they have two losses below their average and two wins above their average…they were 14 point or more deviations on both sides.
Indeed, the (very limited) data suggests that Oregon likely will not play “average”, and instead one could more likely expect a VERY poor performance…..or a VERY good performance.
THE GAME
Looking for an outcome game is more about probability that it is predicting an anticipated spread. That is, what is the probability that Oregon will play well enough above their rating to beat Iowa State. (Assuming that Iowa State plays relatively normal.)
After some final tweaking, I have Iowa State as an eleven point favorite (which means nothing, and that translates to Iowa State with a 77% chance of winning.
However, if you flatten the bell curve due to a high standard deviation of Oregon’s performance, that actually improves the Ducks’ chances of winning a bit, leaving Iowa State a 33% chance of winning. In other words, from a little better than 1 in 4 to right about 1 in 3.
My gut tells me that the key factor in this game is my belief Oregon is not a top ranked team.
Additionally, their only real test was against USC in the CCG. To compare, Iowa State played three teams better than USC, and two more teams that approach being equals. That’s a five to one experience factor in Iowa State’s favor, as well as rating credibility.
Futhermore, I put more weight into Oregon’s regular season downward trend with two bad losses at the end…..than I do their David vs. Goliath victory over USC.
Others who watched the game may be able to better explain that Duck victory, but if I understand right, three USC turnovers played a major factor.
Clearly, everyone has found that the best way to try to beat Iowa State is to force Purdy beat us with his arm. If you can do that, especially if you can hold a lead, you stand a good chance that Brock will force passes.
From my perspective, “IF” Oregon is rated as low as I have them, I believe that means that we hopefully be able to depend more on Breece, rather than having to put the game on Brock’s shoulders. Indeed, the conditions appear to be ripe for Breese to have a great day.
In summary….
Oregon has proven that they can beat pretty good teams. But, they have yet to prove they can beat a team as good as Iowa State. To be fair they haven’t yet had the chance.
Oregon’s wild range of performance means that the Ducks are unpredictable no matter who they face. Therefore, I will be thrilled Iowa State’s just winning their first appearance wearing big bowl pants.
But analysis-wise….and I do fear jinxing us…..Iowa State has a 50-50 shot at a double digit win margin, and a pretty good chance at a win similar to what Texas did to Colorado.
So….I’m liking the good guys in this one.
(In my opinion.)
Cygarin Ratings Linky......
