Which teams won Free Agency?

I wanted to take a holistic look at free agency and compare how well the Washington Commanders filled holes and added talent before the draft. At the same time, it’s tough to evaluate our team without a point of comparison, so I’ll do the same for our NFC East rivals to see how the whole division has prepared itself before the draft. My basic methodology will be somewhat simple: I’ll chart the Pro Football Focus (PFF) grade for each projected starter at the number of positions that I think these teams want to fill on offense and defense. I’ll group these grades into quality tiers and look at the teams’ projected starting roster before and after FA.

Methodology

(Feel free to skip ahead to the charts if you’re impatient for results)

As always, there is some nuance in the details. I’m using my own judgement as to which positions to consider for the starting roster and some of you might disagree with some of my decisions (especially since I’m including 12 positions on offense and defense when a team can only field 11 at one time). For example, I think we should include 3 WR spots and 2 RB spots on these notional starting rosters because I think that’s the level of quality depth worth considering based on snap distributions in the NFL for those positions, but the results look a bit different if only 2 WR and 1 RB spots are considered. I’m also doing my best to project which players will be starters for each team based on depth charts from sites like Ourlads and ESPN as well as my own judgement, but you may think different players should be considered starters (if so, mention why in the comments below).

Another very important detail: in order to offset the effect of injuries and teams that had a down year, I’m actually using the best PFF grade of the last two years for each of these players, where possible.

A quick aside about PFF grades and why I use them. A detailed description of how PFF grades are calculated can be found here, but the gist is that they are determined by a group of analysts looking at every play involving a player, attempting to analyze that player’s responsibility and how well they executed, then assigning a score to the individual play. Those scores are then rolled up into an overall player grade for the year based on how high they scored on a per-play basis compared to other players and that final season grade is typically on a scale of 40 to 100. I considered other metrics like Approximate Value (AV) of Pro Football Reference, but decided that PFF grade (though flawed) is a better representation of the talent a player showed in a single season (though I think AV is a better measure of a player over their full career). I realize that PFF grade is not a perfect metric, but no metrics are perfect and I think the problems should get averaged out when considering the whole roster. If some players are over-valued, others will be under-valued, so the overall result when looking at entire rosters should be reasonable.

As shown in the legends below, I’m using tiers for PFF grades based on this official grouping with a slight modification to make the second tier (Pro Bowl/High Level) as wide as the others. Now onto the results!

Washington Commanders

The results for the Commanders are shown below (with new starters in bold).

The first thing that jumps out at me is the overall weakness of the roster before FA, with numerous “holes” (“replaceable” players denoted in red) and no elite (blue) players. This is improved quite a bit with FA signings. If we define “stars” as elite or high-level players, then the roster gains 3 stars in FA (Austin Ekeler, Frankie Luvu, and Bobby Wagner) and fills 4 holes (removing Saahdiq Charles, Logan Thomas, and David Mayo, moving Dyami Brown to a backup role). This is still a weak roster, with half of the starters graded as “backup” quality, but it has fewer holes and more star power than it had before FA.

A few other things stand out. This is a major roster overhaul, with 14 new projected starters out of 24 spots shown (by far the most in the division). Revamped are 3 out of 5 OL spots, 2 out of 3 LB spots, and 2 out of 3 CB spots. It strikes me that CB may not be as dire a need as I initially thought if Michael Davis and James Pierre can play as well as they did 2 years ago. Also, PFF seems to hate Jeremy Chinn, so let’s see if he’s as bad as his grades over the last 2 years indicate (remember, this is his highest grade in 2 years).

Philadelphia Eagles

The results for the Eagles are shown below (with new starters in bold).

Looking at this, my first thought is my goodness does Philadelphia have a talented roster! Before FA, almost half of their starters could be classified as stars (elite or high level). They lost net talent in FA, but still only have 5 starters rated as backup level or below (compared to over half of Washington’s roster). It’s worth mentioning that the Eagles defense is transitioning from a 4-3 defense to a 3-4, so there could be more growing pains in the defense than is indicated by this talent chart and some of their existing players may not be as good of a fit in their new roles. Still, this model would make me think that if they disappoint this season, it’s a result of coaching rather than talent.

Although the Eagles have a very talented roster, this model indicates they lost net talent in FA, both losing star power and gaining holes. Some of this was outside of FO control, like the retirement of Jason Kelce, but some of it was intentional. Signing Saquon Barkley was the most aggressive move the Eagles made in FA, but the PFF grades indicate it was largely a lateral move from D’Andre Swift. The Eagles did little to fill the hole opened up by Jason Kelce’s retirement and time will tell if that was a mistake. They also signed Devin White to lead the ILB group and he received some of the worst PFF grades on any of these rosters over the past 2 years. Overall, the Eagles still seem like a very talented roster, but with a few more holes and a bit less star power than last year heading into the draft.

Dallas Cowboys

The results for the Cowboys are shown below (with new starters in bold).

Note - DeMarvion Overshown does not have a PFF score due to not playing in the regular season. He was a 3rd round pick by the Cowboys in 2023 and tore his ACL in the preseason.

The first thing that strikes me about the Cowboys roster is how completely it embodies a “stars and scrubs” build. The Cowboys have more elite players and more replacement-level players projected to start than any team in the division, which is wild. The next thing that strikes me is how dependent they are on identifying talent in the draft. Literally every “star” Cowboys player (elite or high level) was taken by the Cowboys in the draft, as were the majority of their starter-quality players.

Overall, the Cowboys lost stars and gained roster holes in FA. This determination is a bit more cut-and-dry than others due to how inactive the Cowboys were in FA. Although they technically have 7 new starters after FA, the only projected starter who was a new FA signing this offseason is Eric Kendricks. The other new projected starters are players returning from last year who are being promoted.

It seems their stars and scrubs build is driven by success in the draft coupled with weakness in FA, since the draft is typically where good teams get talent and FA is typically where good teams fill holes. The Cowboys seem good at drafting talent, but not at filling holes in FA. The Cowboys had a notoriously bad 2023 draft class, so another poor draft class this year could really hurt them.

New York Giants

The results for the Giants are shown below (with new starters in bold).

The Giants are the opposite of the Eagles in that they have the weakest roster in the division (according to PFF grading) and any success they have on the field will likely be due to coaching rather than talent. After FA, the Giants only have 2 star players (Andrew Thomas and Dexter Lawrence), having lost Xavier McKinney to the Packers. At the same time, they have 5 major roster holes, almost as many as the division-leading Cowboys. They seem to have successfully built up the right side of their OL, but got weaker on the back end of their defensive coverage.

On the one hand, the Giants seem to have been active in FA, with 8 new starters and all but Gary Brightwell (rotational RB) a new addition to the team. However, the Giants expended a lot of energy for what on paper look like marginal gains. Brian Burns is a major improvement over Jihad Ward (who was terrible), but he still only grades as a pretty good player by PFF and the cost to acquire him was very high (a 2nd and 5th round pick in addition to extending him to a 5-year deal making him one of the highest-paid EDGE in the NFL). With 5 outstanding roster holes, the Giants will be another team heavily dependent on success in the draft in order to shore up weaknesses and improve the overall talent-level of the team.

It’s also worth mentioning the Giants will have a new defensive coordinator this season (Shane Bowen, former DC of the Titans), though it remains to be seen how that will impact their play.

Overall Roster Movement

The chart below summarizes the net changes in roster composition for each team in the division, side-by-side:

Simplifying it further as “stars added and holes filled”:

Amazingly, the Commanders were the only team in the division able to both fill overall roster holes (4) and increase their overall star talent (3) in FA! Part of this is probably because Washington was starting out with a very weak roster, which made it easier to improve and difficult to worsen. Likewise, it’s probably no coincidence that the most talented teams in the division (Eagles and Cowboys) had the most trouble improving in FA while managing their cap. Still, it speaks well to the evaluation and cap management process of the Commanders FO that they were able to make such marked improvement in a single period of FA. The Giants also started with a very poor roster, but their FO did less to improve it, netting out with a relatively neutral FA period (one less hole, but a little less star power). The Cowboys seem like the biggest losers in FA, tying with the Eagles for losing the most star power (net 2) while opening up the most roster holes in the division (4).

Post-FA Roster Standings

A detailed summary of the division’s roster status after FA is shown below:

The Eagles still look like the class of the division, with more talented players and fewer backup- or replacement-level projected starters than any other team. The Cowboys have a very high-ceiling, low-floor roster, with more elite players and more replacement-level projected starters than any other team in the division. It will be up to coaching to maximize those strengths while hiding those weaknesses. The Commanders are in many ways the opposite of the Cowboys, tied with the Giants for the fewest number of elite players (1), but with the fewest replacement-level players (1) as well. The Giants look like the weakest roster in the division, with the least number of talented players and almost as many replaceable players (5) as the Cowboys.

We can simplify this further by charting only the number of “stars” (elite and high-level players) and “holes” (replaceable players) on each roster:

With the fewest holes on their roster (1), the Commanders look well positioned to take a Best Player Available (BPA) approach to the draft. Adam Peters seems to have done his job well by filling holes in FA. Although the Eagles still net out to having the most stars (9) on their roster, the Commanders actually tie with the Cowboys for second most (7), though we know from the previous chart that the Cowboys stars are mostly in the elite tier. The Giants have the fewest stars (2) and second-most holes (5) in the division, so they will likely be dependent on coaching for a better record this year. It’s maybe worth noting that the Cowboys and Giants each have 3 picks in the top 100, making it difficult to fill all of their remaining needs in the draft. The Eagles also have 3 picks in the top 100, while the Commanders have 6 picks in the top 100.

What do you think about this analysis? Does anything stand out to you about these teams or how they went about free agency? The main consideration I didn’t address here is how teams managed their cap with their additions/subtractions, but I may do a follow-up article about that in the dead time after the draft.