The NFL Draft has come and gone, with 259 players – along with hundreds of others in free agency – earning initial spots in the NFL. With that, and the schedule release later this month, all the big moments in the NFL offseason are finished.
However, on Tuesday, there was a deadline that has increased in both interest and importance since the 2011 collective bargaining agreement. It was the deadline for teams to decide whether to exercise the fifth-year option on their 2020 first-round picks. These players are entering their fourth year in the NFL, and hence exercising the option ensures team control for two more years (and four more years if the team is willing to invoke the franchise tag twice on the player). For players taken from 2011 to 2017, the fifth-year option was worth the transition tag amount (the average of the top 10 salaries at the position) for the first 10 picks, and for the remaining picks in the first round the average of the third-to-the-25th-highest-paid players at the position was used, which is why you saw a kink in terms of rates at which the option was picked up right around pick 10 for the 2011 through 2017 drafts:
Note: We consider a player who signs a market-level extension prior to the option decision as players who have had their option picked up, and strange extension situations (like Jeff Okudah’s and Jordan Love’s) as players who have not had their options picked up. There are a few edge cases, but for the most part these are straightforward to sort out.
Only 12 players had their 5th-year options picked up:
Joe Burrow: $29.5M
Andrew Thomas: $14.1
Tua Tagovailoa: $23.1M
Justin Herbert: $29.5M
Derrick Brown: $11.6M
Jedrick Wills: $14.1M
Tristan Wirfs: $18.2M
Jerry Jeudy: $12.9M
AJ Terrell: $12.3M
CeeDee Lamb: $17.9M
Justin…— Ari Meirov (@MySportsUpdate) May 2, 2023
The option was also not guaranteed until the start of the next league year, except for injury (e.g. Ryan Shazier) which of course made it more attractive for the team to pick up.
When the 2020 CBA was agreed upon, the fifth-year option became much more player friendly. For players who didn’t meet playtime requirements, it became the average of the third-to-the-25th-highest-paid players at the position. This playtime requirement is 75% or more of the team’s snaps in two of his first three years, an average of 75% percent over all three seasons, or 50% or greater over all three seasons. For some players – for example situational pass rushers or two-down running backs, or oft-injured players taken in the top 10 – this is a bargain relative to previous fifth-year options for the team.
Players who meet that playtime threshold earned at least the average between the third-and-20th-highest-paid-players at the position.
14 players officially had their 5th-year options declined:
No. 2: Chase Young
No. 8: Isaiah Simmons
No. 9: C.J. Henderson
No. 11: Mekhi Becton
No. 14: Javon Kinlaw
No. 18: Austin Jackson
No. 20: K'Lavon Chaisson
No. 21: Jalen Reagor
No. 23: Kenneth Murray
No. 24: Cesar Ruiz
No.…— Ari Meirov (@MySportsUpdate) May 2, 2023
For players that made one Pro Bowl (on the original ballot), the fifth-year option is equal to at least the transition tag, which was the previous option number for top-10 picks.
For players that made multiple Pro Bowls (on the original ballot), the fifth-year option is equal to the franchise tag number (the average of the top-five earners at the position). This is a big mark up over the previous fifth-year option.
Furthermore, the fifth-year option is now fully guaranteed, as is any money in the player’s fourth year that wasn’t already so.
Hence, what we’ve seen is an increase in success rates in the top part of the draft, and a decrease in the bottom half of the draft, and no kink in the middle of the graph:
In this article I want to look at some position- and team-level trends in the success rate of the fifth-year option data.
Examining positions
One of the first things to look at when it comes to the fifth-year option is how well different positions have done in terms of success rate. That’s not necessarily as easy as it sounds, though, as some positions are selected earlier than others (for example, quarterback), and the price of the option is different depending on the position. Hence, I built a LOESS curve to model the success rate of the pick, using pick position and CBA (2011 or 2020) as features. We then constructed a “option exercised over expected” by subtracting the rate at which the option was selected minus the expectation of the pick. Here’s how the positions have done:
| Position | N | Average Pick | Option Exercised | Option Exercised OE |
|---|---|---|---|---|
|
TE |
8 |
19.4 |
87.5% |
28.1% |
|
DI |
38 |
17.1 |
71.1% |
8.9% |
|
C |
6 |
20.5 |
66.7% |
8.5% |
|
S |
17 |
19.8 |
64.7% |
7.5% |
|
CB |
42 |
18.1 |
64.3% |
5.9% |
|
T |
39 |
15.5 |
64.1% |
2.9% |
|
WR |
38 |
17.4 |
57.9% |
0.1% |
|
QB |
32 |
9.6 |
59.4% |
-7.3% |
|
LB |
23 |
18.5 |
47.8% |
-10.1% |
|
G |
16 |
18.4 |
50.0% |
-10.4% |
|
ED |
46 |
14.9 |
52.2% |
-11.0% |
|
RB |
14 |
17.9 |
35.7% |
-13.3% |
This somewhat confirms our priors – that running backs are the worst first-round investment, even after adjusting for pick position. Guard and linebacker round out three of the worst four investments in the first round of the draft and are largely considered non-premium positions. It’s also instructive to see that running back, guard and linebacker have some of the lowest sample sizes of picks taken.
However, at the top of the list are three positions in four that we would not consider premium positions: tight end, center, and safety. One might be inclined to say that in these cases teams have more success picking these spots high because they are skillfully picking their spots. This is a dubious assumption considering both the small sample size for each of these positions and the existence of the bottom of the list, which is mostly non-premium positions.
It’s likely more of a signal that four of the top six premium positions in football – quarterback, tackle, wide receiver, and cornerback – comprise the middle of the list. They are picked roughly the most often and are entering the league through a reasonably-efficient market in terms of evaluation. The fact that non-premium positions comprise the top and the bottom of the list might signal more inefficient markets at those positions.
Examining Teams
When looking at success rates for positions, sample size is the overarching caveat, especially for non-premium positions where relatively few players are taken. For teams, where in some cases there are just half-a-dozen picks to choose from, we press further into that caveat. Be that as it may, here’s the above data for teams:
| Team | N | Average Pick | Option Exercised | Option Exercised OE |
|---|---|---|---|---|
|
HOU |
8 |
17.1 |
100.0% |
38.3% |
|
CAR |
10 |
16.2 |
90.0% |
29.7% |
|
ATL |
10 |
17.2 |
80.0% |
23.7% |
|
STL/LA |
8 |
11.5 |
87.5% |
21.6% |
|
BAL |
9 |
22.9 |
66.7% |
20.0% |
|
SD/LAC |
12 |
16.6 |
75.0% |
16.4% |
|
KC |
7 |
17.3 |
71.4% |
15.3% |
|
NO |
11 |
20.4 |
72.7% |
14.9% |
|
NE |
8 |
26.3 |
50.0% |
9.6% |
|
GB |
10 |
24.1 |
60.0% |
9.2% |
|
CIN |
11 |
16.4 |
72.7% |
8.1% |
|
DAL |
9 |
17.4 |
66.7% |
7.9% |
|
IND |
7 |
16.4 |
71.4% |
6.8% |
|
MIA |
12 |
14.3 |
66.7% |
2.0% |
|
PIT |
9 |
22.4 |
55.6% |
0.8% |
|
WAS |
9 |
13.1 |
66.7% |
0.7% |
|
MIN |
14 |
21.3 |
50.0% |
-0.2% |
|
BUF |
9 |
12.3 |
66.7% |
-1.0% |
|
TB |
11 |
12.1 |
63.6% |
-3.2% |
|
DEN |
9 |
18.9 |
55.6% |
-4.0% |
|
NYG |
12 |
15.3 |
58.3% |
-4.5% |
|
NYJ |
11 |
12.3 |
63.6% |
-4.9% |
|
SF |
13 |
17.0 |
53.8% |
-5.4% |
|
DET |
10 |
14.7 |
60.0% |
-6.1% |
|
PHI |
9 |
16.0 |
55.6% |
-6.5% |
|
TEN |
11 |
13.8 |
45.5% |
-15.6% |
|
ARI |
10 |
13.7 |
50.0% |
-15.8% |
|
CHI |
7 |
14.4 |
42.9% |
-23.1% |
|
CLE |
15 |
13.2 |
40.0% |
-27.2% |
|
OAK/LV |
11 |
14.5 |
27.3% |
-34.6% |
|
JAX |
11 |
8.8 |
27.3% |
-36.0% |
|
SEA |
6 |
25.7 |
0.0% |
-39.5% |
A lot of these results check – the Raiders have famously struggled with their first-round capital over the course of the years, as have the Jaguars. Seattle not exercising one (1) fifth-year option this entire time – while making the playoffs in all but three seasons since the 2011 CBA – was a surprise. While Houston has gone through some rough stretches of late, their high-end draft capital has panned out well with J.J. Watt, DeAndre Hopkins, Jadeveon Clowney, and Deshaun Watson. We did count Whitney Mercilus as an exercise of the option since he signed a deal right around the option date in 2016.
It’s fun to see the CEO of SumerSports, Thomas Dimitroff, the former Falcons GM, ring in with one of the best marks in this metric, having hit on all but two of his first rounders.
Conclusion
Fifth-year option data is a fun look into the temporal success rate of teams in what is the most important source for high-end, premium-position talent for NFL rosters. The change in the NFL CBA in 2020 has created some heterogeneity in expectation, which we had to account for in our simple model for the success rate of a pick.
Success at the position level was clamped near expectation for most premium positions, suggesting that theirs is an efficient market, while non-premium positions comprised both the top and the bottom of the list, with tight end surprisingly taking the top spot with a small sample size. Running back, to no surprise, was the least-successful position in the first round, even after adjusting for the value of the pick.
The teams largely shook out the way we expected, with teams like the Rams, Ravens, Chiefs, Saints, Patriots, Packers, Bengals, and Cowboys experiencing both team and first-round success, while teams near the bottom of the list at times struggling on the field as well. There were some surprises, like Houston (good) and Seattle (bad), which is the nature of the small-sample beast.
