[d4cohn]

Here is how the teams on the schedule rank based on Sagarin ratings heading into the weekend's action:

7) Notre Dame
20) Clemson
21) Louisville
29) TCU
38) Virginia Tech
40) SMU
41) Cal
59) Syracuse
64) NC State
84) Wake Forest
97) San Jose State
209) Cal Poly

Using the home field advantage adjustment, Notre Dame, Clemson, Louisville, and Cal are harder games than Syracuse while the others are easier.

I definitely think we should strive for a bowl, but analytically I am not predicting it until we get at least one more surprise win. Lots of good in the game that I highlighted in my post, but fundamentally we still played like a bad team that will be expected to lose more than win. If we played like we did against Syracuse we will still lose most of our games, unless we can get a pick six in all of them.

---

This may be an ironic argument to make, given professionally I analyze data for a living, but just looking at these Sagarin ratings, I think they are too biased toward prior year performance (either explicity or implicitly; I haven't seen a description of the formula yet). I can tell that because, based solely on data from performances this year, no rating system would have Notre Dame ranked #7 after losing to Northern Illinois at home etc. On the flip side, I think Sagarin (massively) undersells Syracuse because of how bad they had been under their previous head coach. Particularly in this NIL era with so much roster turnover, even if Sagarin doesn't explicitly factor in prior seasons' performance, a team with a lot of transfers (like Syracuse) may be better later in the season because of the time it takes for a lot of the new players to learn a new system and implement it in game action. 

If Sagarin doesn't take into account prior years performance, then it is a problem of very small sample sizes.

So either way, I see these Sagarin ratings similar to how I would see polling data: snapshots and sketches in time, but with (many) inherent limitations and biases.

On the yards per play stats, I understand a pick-six is a rare occurrence, but I also think defensive busts like the one we saw in the first half by the defense are also reasonably rare too (particularly given Syracuse had so little time left in the half). I also think on offense, our yards per game stats reflect QB limitations more than anything (Daniels completed three passes for positive yards in the final 7-8 minutes of the first half). I think Brown, when he is healthy, would give this team a (much) better chance to win (if TT is willing to make a change). 

Stanford dominated both lines of scrimmage in a way that the final score, and even the yards per play stats, belie.

[d4cohn]

Here is how the teams on the schedule rank based on Sagarin ratings heading into the weekend's action:

7) Notre Dame
20) Clemson
21) Louisville
29) TCU
38) Virginia Tech
40) SMU
41) Cal
59) Syracuse
64) NC State
84) Wake Forest
97) San Jose State
209) Cal Poly

Using the home field advantage adjustment, Notre Dame, Clemson, Louisville, and Cal are harder games than Syracuse while the others are easier.

I definitely think we should strive for a bowl, but analytically I am not predicting it until we get at least one more surprise win. Lots of good in the game that I highlighted in my post, but fundamentally we still played like a bad team that will be expected to lose more than win. If we played like we did against Syracuse we will still lose most of our games, unless we can get a pick six in all of them.

---

This may be an ironic argument to make, given professionally I analyze data for a living, but just looking at these Sagarin ratings, I think they are too biased toward prior year performance (either explicity or implicitly; I haven't seen a description of the formula yet). I can tell that because, based solely on data from performances this year, no rating system would have Notre Dame ranked #7 after losing to Northern Illinois at home etc. On the flip side, I think Sagarin (massively) undersells Syracuse because of how bad they had been under their previous head coach. Particularly in this NIL era with so much roster turnover, even if Sagarin doesn't explicitly factor in prior seasons' performance, a team with a lot of transfers (like Syracuse) may be better later in the season because of the time it takes for a lot of the new players to learn a new system and implement it in game action. 

If Sagarin doesn't take into account prior years performance, then it is a problem of very small sample sizes.

So either way, I see these Sagarin ratings similar to how I would see polling data: snapshots and sketches in time, but with (many) inherent limitations and biases.

On the yards per play stats, I understand a pick-six is a rare occurrence, but I also think defensive busts like the one we saw in the first half by the defense are also reasonably rare too (particularly given Syracuse had so little time left in the half). I also think on offense, our yards per game stats reflect QB limitations more than anything (Daniels completed three passes for positive yards in the final 7-8 minutes of the first half). I think Brown, when he is healthy, would give this team a (much) better chance to win (if TT is willing to make a change). 

Stanford dominated both lines of scrimmage in a way that the final score, and even the yards per play stats, belie.