Another NFL draft has come and gone, but this year’s draft was anything but ordinary. In an age of unparalleled data access, information about the 2021 draft class was almost as limited as pre-internet days. Thanks to the wide-reaching effects of the COVID-19 pandemic, player background info and tape was much harder to come by, with many players over a year removed from their last game snaps taken. For NFL teams, the lack of a Combine this year meant less person-to-person interaction with players, and perhaps more importantly, less off-the-clock conversations between scouts and front offices to come to a league-wide consensus on player evaluation. The media couldn’t ply team sources with top shelf liquor at any of Indianapolis’s many bars in exchange for draft information. Fans couldn’t watch players run 40-yd dashes and overreact to that one athlete who seemingly breaks the Combine every year. Consequently, draft analysts, reporters, and fans were flying blind entering the draft, relying mostly on tape and the little bits of (often false) information the media could glean from teams. What a year to try to grade a team’s draft performance…
Nevertheless, the world keeps turning and so does the 3GML league. Recently, league GM TH posted a thorough review of the 3GML draft using his always outstanding draft board and team fits to gauge GM performance. GM TH concluded that this was one of the stronger drafts of the 3GML league, paying high praise to the effort of the league members. While I echo his sentiments strongly, I had to have my say in the matter too. Just like last year, I’m grading the 3GML draft using the The Athletic’s Consensus Big Board. But unlike last year, there is a twist: player-position weighted grades.
The Consensus Big Board
Last year I introduced the Consensus Big Board as a tool to establish draft prospects’ Estimated Draft Pick (EDP), and in combination with Jimmy Johnson draft trade chart, draft value. For more on those methods, I point you to the April 30, 2020 blog post. As a refresher, the Consensus Big Board (CBB) is an aggregation of numerous big boards posted by NFL draft analysts and media. The board is divided into the 3 versions: Forecaster Board, Evaluator Board, and Consensus Board. As I wrote last year:
“I use those terms specifically, as they state, ‘one type of analyst, the “forecasters,” tend to have much more access to NFL and college personnel, which gives them information about injury concerns, character, off-field issues and behind-the-scenes information that could change our understanding of a player one way or the other. The other group, the “evaluators,” rely on public data — primarily college game film and advanced statistics.’ The Athletic creates a board for each. They suggest the Forecaster Board better reflects the actual draft than the Evaluator Board, which should be no surprise given the information they are gathering. However, the Evaluator Board, they claim, is better at projecting NFL success. The Consensus Big Board is a combination of both the Forecaster and Evaluator Boards – ideally the best of both worlds.”
And last year, the CBB was the best of both worlds, barely nudging out the Forecaster Board in terms of predictive ability. This year, however, would not be the case. Once again I plotted the 3 boards against actual NFL draft picks (figure below). The Forecaster board ran away with the show this time, posting a R2 = 0.67, which topped not only the other 2 boards this year, but all of the boards last year. To be honest, I was quite surprised. Given the lack of info, I expected teams would draft closer to the Evaluator board, which relies mostly on tape. Despite all of the smoke this year, it appears that the media largely understood where NFL teams valued prospects.
The thing is, an analysis of the past CBB has shown that the Evaluator board is a better predictor of NFL success, so placing too much weight on the Forecaster board is perhaps not the best way to grade draft performance. So, as with last year, I chose to stick with the Consensus Board - the best of both worlds (I hope).
Weighting the Grades
Like last year, I used the CBB and weighted each pick to its respective trade value based on the Jimmy Johnson (JJ) trade chart. While this provided an objective grade, it failed to account for the tendencies of teams to “reach” for positions the NFL values above others, particularly QB. This resulted in poor grades for QB-needy teams taking QB prospects early in the draft – picks which no analyst would otherwise grade as a poor move. To temper this effect, I wanted to introduce positional value into the CBB grading method.
My first thought was to use the Franchise Tag value as quantitative metric to weigh the CBB grades. Given that the Franchise Tag is derived the average of the top 5 salaries at each position normalized to the Salary Cap (check out OTC for more), I felt this would be a good indicator for how the NFL valued positions. In testing Franchise Tag as a weighting method, I found that it is a strong indicator, but it suffers from one fatal flaw we see discussed every year: it lumps edge rusher with linebackers and tackles with interior offensive line. The result artificially inflates the value of the off-ball linebackers and interior offensive line, while bringing down the value of stand-up edge rushers and tackles.
I decided instead to simply use the top 5 salaries at each position prior to the draft, but 1) separate tackles from interior line, 2) separate stand-up rushers from off-ball linebackers and 3) combine stand-up rushers with 4-3 defensive ends as “Edge” rushers. To determine a weight, I first took the average of the Top 5 salaries across all positions, then divided each positional average by this value. The idea is that weights greater than 1 are positons that the NFL places more value on, while weights less than 1 are positions that are less valued. However, the results using the overall top 5 average were too top heavy thanks to the enormous value of QB contracts and as such only a few positions fell below a weight of 1. So instead, I repeated the process using the median overall value, which better reflected how I believe the NFL values positions. These are shown in the table below.
As we would expect, QB is valued far above every other position. The next closest is Edge rusher, which again is no surprise. Then WR, OT and perhaps most surprisingly, DT all fall closely together. Even more surprising, CB was only slightly above 1 despite the need for shutdown corners in a pass happy NFL. This is perhaps proof that the CB market has not yet caught up with other positions on defense. On the flip side, LB, IOL, S, RB, and TE all were valued less than 1, with TE surprisingly below RB. A quick check of the projected 2022 Franchise Tag indicates that this trend is real with the 2022 TE franchise tag falling below the RB tag despite being higher than RB in 2021. Special teams (K, P, LS) live in the basement here, valued at only a fraction of the next lowest position (TE).
With these weights in hand, I re-ran the CBB grades, multiplying each pick value by the corresponding positional weight. The results were promising, but I quickly realized this method overvalued positions in later rounds of the draft, when positional value should be less of a factor. So instead, I decreased the value of the positional weights by round. To do this, I subtracted the nominal “position-less” weight of each pick (i.e. “1”) from the positional weights to establish the surplus or deficit value at each position, then reduced this value by a fraction of the round. For example, the weight for picking a QB in the 1st round was 2.293 (2.293 – 1 = 1.293*7/7 = 1.293 surplus over nominal pick value), but by round 7 it was only 0.185 (2.293 – 1 = 1.293*1/7 = 0.185 surplus over nominal pick value). Thus positions become less weighted each round. This method was applied to every draft pick by every team, including the 3GML members. Like last year, I also graded the 3GML UDFA’s, but UDFA’s were ascribed no positional weighted value and did not factor into overall grades.
Overall Team Performance
As a baseline for comparison to last year, I first ran the CBB grades without positional weights. Like last year, I compared the player selected at each team’s pick (with corresponding JJ value) against the players’ respective EDP (and corresponding JJ value) on the 3 Big Boards to establish the value of each pick. I totaled this up for each team, both in total points and percentage of original pick value. To establish a grade, I repeated this, but then normalized the value difference between original pick value and player EDP value by the number of picks in the draft (259). Since these numbers are not easily interpretable, I as established a GPA grade based on the Consensus Grade, and curved all GPA to the top grade. The results are shown below.
The big winner here is GM TH, who not only topped the 3GML league members, but the entire NFL, grading out with a 4.0 (A+). In comparison, last year the highest 3GML member was GM CD, who ranked 4th in the NFL at 3.6 (A-). GM CD did not repeat this performance this year, falling all the way down to 22 with a 2.5 (B-). Both GM’s SD and EP had strong showings using the unweighted CBB, finishing 7 with a 2.9 (B) and 12th with a 2.8 (B) respectively.
With this baseline in hand, I next calculated the grades using the positional weights from the previous section. The results are shown in the table below.
Right away we see differences. GM TH was supplanted as the top performer by the Chicago Bears, who got an added boost thanks to picking a young QB, Justin Fields, at 11 (for the sake of comparison, NFL team grades are based on real draft picks, not what they picked in the 3GML world). GMs SD, EP, and CD all got a Consensus Grade boost using this method, but EP’s curved GPA (2.7), grade (B-), and position (15) were all lower than the unweighted grades. GM SD’s GPA improved to a 3.1 (B+) good for 5th overall. GM CD’s grade was the most improved, now a 2.8 GPA (B) putting him at 11th overall.
The overall change between unweighted and weighted Consensus Grades are shown in the figure below. Most teams benefited from the weighted method, indicating that teams were by large using premium picks on positions of higher value, as we would expect. In the next section, I will dive into the picks made by the 3GML team members and discuss these in light of the weighted CBB grades.
3GML Breakout
As with last year, I looked at each 3GML team pick and UDFA’s (all UDFA were ascribed pick 260 to keep it even between teams). Each team is discussed below in alphabetical order, with best value, worst value, and overall grade discussed. As with the curved grades in the previous section, I did not factor in value for trades, assuming that the move up or down accounted for the value of the new pick. Therefore, when I say “original pick value,” I mean the value of the final picks for each team.
Green Bay Packers
Best Value: WR Terrace Marshall Jr.
WR is considered a plus value position, and Terrace Marshall at 46 was extra value for GM SD. Marshall was the 5th ranked WR on the CBB and many thought he could hear his name called at the end of Day 1. However, questions about his snap to snap effort and injury history pushed him down the board a little bit. If he reaches his projections, GM SD will get good WR value for the 2nd draft in a row.
Worst Value: OT Liam Eichenberg.
Despite this being a premium position, Eichenberg at 32 is considered a bit of a reach on the CBB. In real life, he heard his name called at pick 42, closer to where he was ranked on all 3 boards. Eichenberg is a high floor technician who will likely be in the league 10+ years, but may have to transition to the interior due to measurables. Regardless, GM SD would be wise to let him fail outside first before sliding him into the interior.
Overall Grade: B+
A solid draft for GM SD, who snagged a 3.1 GPA good for a B+. Despite having the lowest draft capital of the 3GML members, GM SD’s picks were worth 126% of the original draft available to him. Aside from the Eichenberg pick, every other pick made by GM SD was graded positively using the position-weighted CBB. With a new QB at the helm for GB, GM SD hopes this draft will help keep his team a contender for years to come.
Houston Texans
Best Pick: Edge Azeez Ojulari
Just about everyone felt Azeez Ojulari wouldn’t escape the 1st round, but the NFL clearly had larger concerns about his knee that was previously reported. Don’t mention that to Ojulari though, who is already using his tumble down the boards as motivation for this season. GM TH chose to end his tumble at pick 38, netting him a +1.6 Consensus Grade. This just edged out (pun intended) the grade he received for OT Christian Darrisaw, which was another excellent value pick. But the edge rusher wins the day here and GM TH hopes this won’t be the only time Ojulari beats a tackle.
Worst Value: S JaCoby Stevens
If your worst value pick is your last and it barely even grades negatively (-0.0009 Consensus Grade), it is safe to say you nailed the draft. Ask GM TH and he will identify this as his least favorite pick, but I don’t think he’s losing sleep over it.
Overall Grade: A
What a draft. GM TH stole the show with a 3.9 GPA good for an A. He grossed 157.8% of his original pick value by being patient and knowing when to strike for a falling player. The bar has been set high for the rest of the league in 3GML 2022 Draft.
Miami Dolphins
Best Value: OC Creed Humphrey
Arguably the top center in the draft, Creed Humphrey brings intensity and leadership GM EP’s O-line. Projected to be an early-mid 2nd round pick, Humphrey fell further than expected to the surprise of many. GM EP happily traded up to end Humphrey’s tumble and in the process secured his best value pick. Look for Humphrey to be a starting IOL for the Miami Dolphins for years to come.
Worst Value: LB Jeremiah Owusu-Koramoah
Whether you consider JOK a LB or S, both positions carry less weight to NFL teams. So even though GM EP selected JOK only 2 picks before his CBB rank, his positional value drove this grade down. JOK ended up falling out of the 1st and to the waiting arms of the Cleveland Browns in real life, which looks to be a good place for his development. Despite this grade, make no mistake – JOK is a baller and GM EP is pleased to have him on his squad.
Overall Grade: B-
GM EP fell victim to the position weighting in the CBB. Despite the negative grade on his first pick, GM EP still finished above average with a 2.7 GPA good for a B- and 15th overall (thanks in part to his next plus value pick of Edge Kwity Paye at 18). GM EP still got plus value out of his picks with 110.1% of his original pick value. With 6 of his 10 picks going towards plus-valued positions on defense, this class might be the future the Miami Dolphins defense.
San Francisco 49ers
Best Value: QB Justin Fields
This is perhaps the first pick to benefit from positional weighting. Drafting a young QB at pick 3 despite a lower consensus rank was recipe for a negative grade last year (see GM SD’s pick of QB Tua Tagovailoa). Thanks to huge value placed on QB’s by the NFL, this pick ends up being the best value pick by GM CD. It is safe to say that his pick signals the 49ers are heading in a new direction and hope that they will get even more value from Justin Fields in years to come.
Worst Value: TE Tommy Tremble
While only selected 1 pick before his CBB rank, TE is not a high-value position to the NFL and this grade suffered for it. Tremble is an excellent scheme fit for GM CD’s team and he identified Tremble early on as one of “his guys.” With great blocking ability and a high floor, Tremble may prove this grade wrong in years to come.
Overall: B
GM CD was the biggest beneficiary of the position-weighted CBB grades, vaulting from 22 to 11 with a 2.8 GPA and solid B grade. This draft was all about scheme and resetting the roster for the future of the 49ers, and scheme picks often don’t always equal draft value. In this case, GM CD scraped by with 100.1% of original draft pick value.
Conclusion
In 2020, the CBB grading suffered from a lack of positional value. After refining the methodology, the positional weighting added a new dimension to the CBB grades in 2021. The new methodology better reflected the significance of spending premium picks on positions more highly valued by the NFL. Last year, I ended this blog post by pointing out the next step to improve this grading. I will do the same this year. I hope to continue to develop the CBB grading by incorporating historical draft trends and success rate in the 2022 version of this blog post. Until then, I commend the 3GML GM’s for another excellent showing in the NFL draft!
