Quick and dirty averages for 2023 season using QBs who qualify for CMP% on Profootball reference.

CMP% 65.4

YDS 228.3

TD/G 1.3

INT/G 0.76

Y/A 7.05

Rate: 89.3

Sack 2.47

Also feel like here I should mention his earliest game for the last data point in the original post dates back 10/13/2022 against Washington, which is only his 15th game, so there is about a 3 game overlap for the first and last data point. Part of the reason I decided to re-do this with the 8-game running average as well in my other comment

Decided to do an 8 game running average as well and plotted everything on its own graph. More noise, but with only 31 games I think it makes more sense

https://imgur.com/a/BNvMS5P

Pick up his 5th year option, it’s not that much. That gives you 2 more years to develop and evaluate and 2 years to take your time finding a coach/OC you trust and building the team to an acceptable level for a rookie QB.

IMO one of the biggest mistakes the Browns and Jets make is being in on the QB carousel every 1-3 years and investing too much draft capital there over the rest of the team.

It starts at the top.

GM has to be secure enough to take the 3-4 year approach. Coach has to be feel he’ll be fairly evaluated on his skills, not the W/L record with a bad team. Then you build the roster and bring in a QB when you have a team to support them.

That is laughable. Based on your exact description of their methodology there is only 1 reviewer and if they disagree it’s more of an arbitration between the original two scores than a review.

Peer reviewed journals publish many pages worth of data supporting their findings. Which is reviewed and critiqued by peers, often requiring additional experiments. Results are more often than not quantitative and if you approach a journal with qualitative data only it better be rock solid or the reviewers will have a field day.

PFF is purely qualitative data that is pseudo-quantified. Reporting their methods doesn’t make their methods good (this is true in peer reviewed journals as well).

Any time you take a qualitative source data and attempt to quantify it, the data should be questioned. Pseudo-quantification is good as a supporting measurement, but if it’s your primary measurement for your conclusion then you’re in trouble. Just own it as qualititative and you’ll get more respect.

How much cross reference does PFF do to make sure scores are consistent. I.e. do the same 3 people score all the OL in every game or is one guy assigned to each team, grading that team and the opposing team each week. Do they publish results of bias studies on their graders? Are some graders tougher on some teams than others? Or on some players than others (i.e. position, star status, race/ethnicity, etc)

The data is fine for fans who have no idea how to judge a player within a given season, but it is severely lacking in many aspects

Bears have most cap space in the league, they’d be happy to slap 2 franchise/transition tags on him (depending on who else needs a tag… Such as JJ), and then let him go after he reaches his age 30 season or at the first sign of injury trouble