[BostonCard]

Matt, thanks for your thorough analysis. One thing that I’d like your help interpreting is the idea that recruiting in the OL has gone down (I haven’t looked at the defensive line data because it is harder to review since there are position switches).  On the offensive line, there may be a slight trend to lower ratings, but it is hard to tell the difference due to the large variability in numbers and stars between years. If you look at a curve of either number of stars or 24/7 rating by year, it is pretty flat, with maybe a slight decrease at the end due to a below-average 2021 OL recruiting class.  A simple regression by year finds no statistically significant trend over time using either number of stars, 24/7 Rating, or normalized Rating (which I did to make the dynamic range of the ratings a little more noticeable, as they otherwise stretch between 0.85 and 0.99).

In order to try to get a little more power, I divided the twelve years into three period (2011-2014, 2015-2018, and 2019-2022). The average star rating does decrease a little bit (from 3.67 to 3.54 to 3.35) which supports your thesis. However, going from the first third of the Shaw years to the last third, a change of 0.32 stars basically corresponds to a little over one additional 4* player (instead of a 3* player) per year. However, the two 5* players we got in 2012 (Murphy and Peat) and again in 2017 (Little and Sarell), account for all the difference in the periods.  Moreover, we have recruited more players in the most recent 4 year span (17) than in either of the earlier spans (15 from 2011 to 2014 and 13 from 2015 to 2018), so I am not completely sure that the results don’t cancel out. For all the quality of the 2017 class, the headliners were often injured and in the end contributed less than their star ratings would have predicted, and the 2017 class was followed by the disastrous 2018 class, where we all but sat out OL recruiting. Had we had better numbers in 2018, even if they were three star players, it would have dragged down the average but may have averted bad OL play down the line.

Anyway, the TL; DR of this is that there has been a small but noticeable decline in OL stars and/rating, that I think may be made up for by greater OL numbers.

What do you think?

BC

---

One add, one way to look at the data to account for both quality and quantity is to look at the total number of stars.  From 2011 to 2014, we recruited 15 players with a total of 55 stars. From 2015 to 2018, we recruited 13 players with a total of 46 stars. From 2019 to 2022 we recruited 17 with a total of 57 stars.

BC

[BostonCard]

Matt, thanks for your thorough analysis. One thing that I’d like your help interpreting is the idea that recruiting in the OL has gone down (I haven’t looked at the defensive line data because it is harder to review since there are position switches).  On the offensive line, there may be a slight trend to lower ratings, but it is hard to tell the difference due to the large variability in numbers and stars between years. If you look at a curve of either number of stars or 24/7 rating by year, it is pretty flat, with maybe a slight decrease at the end due to a below-average 2021 OL recruiting class.  A simple regression by year finds no statistically significant trend over time using either number of stars, 24/7 Rating, or normalized Rating (which I did to make the dynamic range of the ratings a little more noticeable, as they otherwise stretch between 0.85 and 0.99).

In order to try to get a little more power, I divided the twelve years into three period (2011-2014, 2015-2018, and 2019-2022). The average star rating does decrease a little bit (from 3.67 to 3.54 to 3.35) which supports your thesis. However, going from the first third of the Shaw years to the last third, a change of 0.32 stars basically corresponds to a little over one additional 4* player (instead of a 3* player) per year. However, the two 5* players we got in 2012 (Murphy and Peat) and again in 2017 (Little and Sarell), account for all the difference in the periods.  Moreover, we have recruited more players in the most recent 4 year span (17) than in either of the earlier spans (15 from 2011 to 2014 and 13 from 2015 to 2018), so I am not completely sure that the results don’t cancel out. For all the quality of the 2017 class, the headliners were often injured and in the end contributed less than their star ratings would have predicted, and the 2017 class was followed by the disastrous 2018 class, where we all but sat out OL recruiting. Had we had better numbers in 2018, even if they were three star players, it would have dragged down the average but may have averted bad OL play down the line.

Anyway, the TL; DR of this is that there has been a small but noticeable decline in OL stars and/rating, that I think may be made up for by greater OL numbers.

What do you think?

BC

---

One add, one way to look at the data to account for both quality and quantity is to look at the total number of stars.  From 2011 to 2014, we recruited 15 players with a total of 55 stars. From 2015 to 2018, we recruited 13 players with a total of 46 stars. From 2019 to 2022 we recruited 17 with a total of 57 stars.

BC