Skip to main content
Menu
The Rams Love Trading Early Draft Picks. How Much Value Do They Find In The Later Rounds?

The Los Angeles Rams are built differently. In a league that tends toward conservatism and groupthink, Rams general manager Les Snead is nothing short of a wild-eyed contrarian. For years, the received wisdom in the NFL has been that building a successful team over the long term requires a star quarterback coupled with good, cheap, young players that provide surplus value over the period of their rookie deals. This structure allows a team to compete year in and year out, and it provides cap space to fill any gaps formed through injury or ineffectiveness by paying up for players at positions of need in free agency.

Or at least that’s how conventional thinking goes. Snead and the Rams have taken a different approach — one that’s fascinating in its divergence from NFL norms. Instead of coveting early round picks — particularly first-rounders, which have the best chance of providing a team with those crucial inexpensive young players — Snead has made a practice of trading them away. Whenever an opportunity arises to acquire a proven player on an expiring deal (often with a more expensive contract looming on the horizon), he’s had no issue with dealing first- and second-round picks.

Snead’s run of trading off premium picks began in 2016. After trading away the Rams’ 2016 and 2017 first-round picks1 to move up 14 spots to draft Jared Goff first overall in 2016, Snead has dealt a first-round pick in each year since

In 2018, Snead traded the Rams’  first-round pick and a sixth-rounder to New England for wide receiver Brandin Cooks and a fourth-rounder. In 2019, he traded out of the first round, dealing the 31st and 203rd overall picks to Atlanta for the 45th and 79th picks. In 2020, he dealt their 2020 and 2021 firsts2 to the Jacksonville Jaguars for cornerback Jalen Ramsey. And this past offseason, he moved the Rams’ first-round picks for the 2022 and 2023 seasons — along with a 2021 third-rounder and Goff — to the Detroit Lions for quarterback Matthew Stafford.

The Stafford trade was remarkable not merely for the sunk costs it represented — recall the haul that the Rams gave up, including two firsts, simply to draft Goff in 2016 — but also for the dead money the Rams had to eat to move him. $22.2 million of Goff’s salary remains on the books this season, a consequence of the $134 million extension, with a then-NFL record $110 million guaranteed, that Snead gave Goff in 2019.3 This wasn’t the first time Snead traded away an expensive player for a loss, either. After signing Cooks to an $80 million extension in 2018, last year the Rams traded Cooks and a fourth-round pick to Houston in return for a 2020 second-round pick — eating $21.8 million in dead money in the process.

From afar, the moves appear to reek of mismanagement. Paying premiums to take big swings on players and then paying again to move on from them doesn’t seem like a sustainable strategy. Things may not be so cut-and-dry, however. According to Jourdan Rodrigue, Rams beat writer at The Athletic, L.A. has a well-thought-out plan — and it’s more than just a “stars and scrubs” approach to team building.

“It’s about turning what is often perceived as an inefficient pick — or a total inefficiency in terms of third-, fourth-, fifth-, sixth-, seventh-round picks — [and] trying to turn them into efficiencies and advantages, and leverage the math,” Rodrigue said on a recent episode of The NFL Rhodes Show podcast with Lindsay Rhodes. “It’s how Sean McVay built his offense, it’s how they’re trying to reenergize this defense. … It’s how they operate in their personnel department in their front office — turning perceived disadvantages into mathematical advantages.”

NFL predictions illustration

Related: Our 2021 NFL Predictions Read more. »

Rodrigue did note that what the Rams look for in these late-round picks are not “complete players” but rather athletes who complement their stars — defensive lineman Aaron Donald, Ramsey and Stafford. Rodrigue said this approach, coupled with winning enough games each season to make the premium picks they trade away less valuable, is part of a recent philosophical change in L.A. The pivot represents a new, overarching strategy that may look inefficient from the outside but is internally consistent and is one the Rams believe can lead to lasting success.

What makes these claims about leveraging math and finding efficiency in the late rounds exciting is that they’re explicitly testable. If the Rams have found a way to find value where other teams are failing, that represents a massive edge over the league, and it’s one we should be able to identify in their draft results.

To ground our analysis in the context of what Rodrigue reports the front office is actually doing — finding complementary players — we developed a novel metric: Meaningful Snaps Over Expected (MSOE). A meaningful snap is one given to a player when a game is in doubt, and we reasoned that those snaps will best help us measure the contribution of complementary players across all positions on an NFL roster. We defined a meaningful snap as one given to a player when a team’s pre-snap win probability is above or equal to 10 percent and under or equal to 90 percent. 

To calculate MSOE, we took each player drafted in the fourth through seventh round4 in the NFL draft from 2007 through 2020, and built a model5 to predict expected meaningful snaps. Next we subtracted the expected snaps from a player’s observed meaningful snaps, and finally we totaled them by team. 

With this framework in place, we found that from 2017 (the first draft after the Rams began trading away their premium picks) through 2020,6 L.A. has been slightly above average in extracting meaningful snaps from its late-round picks.

L.A. is above average at finding late-round contributors

Teams ranked by Meaningful Snaps Over Expected (MSOE) from players selected in the fourth through seventh rounds of the NFL draft, 2017-20

team num. picks hit rate rank MSOE rank
Los Angeles Chargers 16 56% 1 2785 1
San Francisco 49ers 19 53 5 2007 2
Indianapolis Colts 23 39 21 1550 3
Washington Redskins 26 46 14 1423 4
Chicago Bears 16 50 6 1233 5
Buffalo Bills 17 47 10 1214 6
Dallas Cowboys 22 41 17 973 7
Seattle Seahawks 24 54 4 850 8
Atlanta Falcons 15 47 11 843 9
Detroit Lions 20 55 3 564 10
Philadelphia Eagles 18 56 2 542 11
Los Angeles Rams 24 33 27 449 12
Oakland Raiders 18 39 22 246 13
Jacksonville Jaguars 19 47 12 122 14
Miami Dolphins 18 50 7 101 15
Tampa Bay Buccaneers 13 31 30 57 16
Cleveland Browns 17 41 18 25 17
New York Giants 18 50 8 -31 18
Carolina Panthers 16 50 9 -61 19
New England Patriots 18 39 23 -161 20
Kansas City Chiefs 12 42 15 -162 21
New Orleans Saints 11 36 25 -306 22
Tennessee Titans 12 42 16 -355 23
Houston Texans 15 33 28 -706 24
Green Bay Packers 25 40 19 -775 25
Cincinnati Bengals 25 32 29 -794 26
Baltimore Ravens 20 35 26 -805 27
Arizona Cardinals 19 37 24 -1081 28
Minnesota Vikings 35 40 20 -1147 29
Denver Broncos 17 47 13 -1153 30
New York Jets 18 28 32 -1451 31
Pittsburgh Steelers 17 29 31 -2136 32

A meaningful snap is defined as a play for which the pre-snap win probability is above or equal to 10 percent and under or equal to 90 percent.

A hit is defined as a player who has amassed positive MSOE for their career, with MSOE calculated seasonally. Hit rate is a team’s number of hits divided by their total number of selections from rounds 4-7.

Source: ESPN Stats & Information Group

The Rams rank 12th in MSOE with 449 snaps over what we would expect from the 24 selections they made in rounds four through seven. Of those 24 picks, however, just eight have returned positive MSOE over the course of their careers, giving the Rams a hit rate of just 33 percent, good for 27th in the league.

Only a few late-round Rams picks have delivered

Players drafted in 2017-20 by the Los Angeles Rams in the fourth through seventh rounds, by Meaningful Snaps Over Expected (MSOE)

draft
player position year round Snaps Games MSOE
David Edwards OL 2019 5 1647 42 +805
Sebastian Joseph-Day DL 2018 6 980 39 +710
Samson Ebukam DL 2017 4 2068 73 +546
Jordan Fuller DB 2020 6 939 22 +456
Josh Reynolds WR 2017 4 1674 69 +340
Nick Scott DB 2019 7 809 41 +330
Brian Allen OL 2018 4 876 32 +42
Dakota Allen LB 2019 7 413 27 +9
Micah Kiser LB 2018 5 711 31 0
Sam Sloman K 2020 7 53 8 -4
Greg Gaines DL 2019 4 487 36 -30
John Franklin-Myers DL 2018 4 818 40 -73
Clay Johnston LB 2020 7 123 8 -109
Justin Lawler LB 2018 7 200 19 -116
Travin Howard LB 2018 7 377 24 -120
Tremayne Anchrum OL 2020 7 31 11 -121
Ejuan Price LB 2017 7 13 1 -123
Sam Rogers RB 2017 6 0 0 -149
Trevon Young LB 2018 6 5 2 -160
John Kelly RB 2018 6 87 11 -180
Tanzel Smart DL 2017 6 362 34 -307
Obo Okoronkwo LB 2018 5 412 27 -352
Brycen Hopkins TE 2020 4 38 5 -389
Jamil Demby OL 2018 6 119 6 -556

Source: ESPN Stats & Information Group

In general, MSOE tracks with Pro-Football-Reference.com’s Approximate Value (AV) metric for these Rams players, but MSOE also weights draft slot, position and early career playing time.7For instance, Jordan Fuller has a career AV of 5, and MSOE sees him as a very successful late-round pick based on his rookie season alone, when he was on the field for 939 meaningful snaps, over twice what we would expect. Meanwhile, linebacker Dakota Allen, a seventh-round pick with 1 career AV, is seen as a hit (albeit just a slight one) since late-round selections have a low probability of contributing meaningful snaps.

It’s worth noting that this analysis punishes teams with depth across their rosters, since having lots of players on the depth chart in front of a late-round pick makes it less likely they will see meaningful snaps. Early round picks are also given favor by coaches over later-round picks — sometimes even when the late-round players are performing at a higher level. In other words, this analysis was tailored to show the Rams’ stated strategy in the best light possible. They are top-heavy by design, and they have very few early round picks with whom their late-round selections must compete.

However, we fail to find convincing evidence of a late-round edge in the Rams’ recent draft classes. Snead has not been bad, but if you’re going to trade away your premium picks, you better have an ace up your sleeve — so long as building a long-term winner is the goal. The evidence suggests the Rams are merely mortgaging their future to attempt to win now, and that isn’t a particularly novel approach.

Since going all-in on the season by trading even more draft capital to acquire 32-year-old Von Miller from the Denver Broncos and signing Odell Beckham Jr. after he was released by the Cleveland Browns, the Rams have lost two straight, including a blowout against division rival San Francisco this past Monday. For the Rams, the pressure is mounting. This season has become one where anything less than a Super Bowl win will be viewed as a disappointment. And if their late-round draft returns are any indication, it’s an opportunity they simply can’t afford to squander.

Check out our latest NFL predictions.


Who’s the most chaotic fictional football coach? | FiveThirtyEight

Footnotes

  1. Along with two 2016 second-round picks, a 2016 third-round pick and a 2017 third-round pick.

  2. Along with a 2021 fourth-rounder.

  3. Patrick Mahomes now holds the record for largest guaranteed contract.

  4. We excluded third-round picks on the grounds that they are better characterized as midround, not late-round, selections. Moreover, third-rounders represent picks with substantial and well-recognized surplus value. This is a big reason teams like the New England Patriots — the archetype for conventional long-term NFL success — made a habit of collecting third-round compensatory picks. And ultimately, finding value in the third round does not fit the definition of turning an inefficiency into an efficiency, which is the crux of the Rams strategy.

  5. We used an ensemble regression model containing two sub-models: “ranger,” a regression forest model, and “glmnet,” a generalized linear model using penalized maximum likelihood. Hyperparameters were discovered via grid search. For glmnet, the predictors were 

    preprocessed by centering, scaling and applying principal component analysis. Data was split into 70/30 train and test sets, and 10-fold cross validation was used. Predictors included draft slot, draft round, combine testing data, position and season. Out-of-sample r-squared was 0.38.

  6. We removed 2021 picks from consideration since the current season isn’t over yet.

  7. According to Eric Eager at Pro Football Focus, wins above replacement is less flattering to the Rams’ strategy. Six players drafted in the fourth round or later since 2017 have a WAR over expected that is positive: safety Jordan Fuller, wide receiver Josh Reynolds, safety Nick Scott, guard David Edwards, edge rusher John Franklin-Myers and defensive lineman Sebastian Joseph-Day.

Josh Hermsmeyer is a football writer and analyst.

Comments