There’s been a proliferation of takes on the Sam Bradford trade, where the Vikings sent their first-round pick to the Philadelphia Eagles for Bradford, hitting sports news, generally precipitated by a poor performance against his former team on November 23rd.
It’s probably telling that most of the articles you can find questioning the Bradford trade come from his worst performance of the year, with his lowest passer rating (71.6) and adjusted net yards per attempt (3.49).
If one judged Russell Wilson’s season—where he ranks 16th in adjusted net yards per attempt—by his worst game, you’d think he was worse than Blaine Gabbert. After all, a 17/33, 151-yard performance with no touchdowns and two interceptions looks abysmal. The same is true of Jameis Winston (27/51, 243 yards, one touchdown and four interceptions), Cam Newton (21/35, 262 yards and three interceptions) as well as MVP candidates Derek Carr (17/41, 117 yards with no interceptions or touchdowns) and Matt Stafford (23/36, 213 yards with no touchdowns and two interceptions).
All of those quarterbacks, as well as Alex Smith, Philip Rivers, Andy Dalton, Ben Roethlisberger, Joe Flacco, Carson Wentz, Ryan Fitzpatrick, Tyrod Taylor, Andrew Luck and Eli Manning have had worse “worst games” than Bradford.
Despite that, a short period after that game was when Bill Barnwell published his piece criticizing the trade and Steven Ruiz at For the Win (of USA Today) published his take around that time as well. Both of them laid out their reasoning with specificity, though Chris Chase at Fox Sports made the same argument without as much thoroughness.
Not everyone is pessimistic. My colleague at 1500ESPN, Matthew Coller, responded to Barnwell’s piece with the opposite argument. Most recently, Cian Fahey (who now also writes for 1500ESPN) published an article on Football Outsiders that essentially makes the same argument, in impressive detail.
Both Barnwell and Ruiz acknowledge that Bradford is playing well; Barnwell starts off his argument that he thinks “it’s fair to say Bradford has played the best football of his career,” while Ruiz says “Bradford, himself, is not to blame for the Vikings’ failures. He’s actually played well considering what he has around him.”
The issue for both of them is that the Vikings were a lost cause as soon as Bridgewater went down. Barnwell argues that general manager Rick Spielman should have been able to anticipate all the injuries that hurt the offense, and the offensive line in particular. Ruiz argues that the team was not particularly set up for success in the first place.
I’m not sure either of those arguments make sense—the Vikings have had eleven different offensive linemen play in games, and two more who many anticipated would be available but couldn’t make it to the regular season (Mike Harris and Phil Loadholt). A third (John Sullivan) was cut, perhaps on request. Between Matt Kalil, Alex Boone, Joe Berger, Mike Harris and Phil Loadholt, the Vikings are left with one of their expected starters for the year, and even he has missed snaps.
With Berger still on the concussion protocol, the Vikings are left with Nick Easton to play in his third game (his second start) directing the movements of the backup to the backup left tackle (T.J. Clemmings replaced Jake Long, who replaced Matt Kalil) and the backup to the backup’s backup at right tackle (Andre Smith replaced Phil Loadholt, and himself was replaced by Clemmings before Jeremiah Sirles took the job).
All of that ignores the fact that Zac Kerin and Willie Beavers have had to take significant snaps as well, and they were not good snaps for Minnesota.
Injury prediction websites like injurypredictor.com label the most injury-prone players as having a 55 percent risk of being injured throughout the season, with players who sustain additional injuries early on rising as high as 85 or 90 percent.
If we make the riskiest assumptions—that Kalil was at a 55 percent risk of getting injured and that Loadholt and Harris were at 90 percent, there’s a 44.6 percent chance the Vikings would be saddled with their backups: Long, Smith and Fusco.
If one additionally applies an extreme risk probability to Long and a high-risk probability to Smith, one ends up with a 22 percent chance of ending up with a third-string left tackle and a third-string right tackle with no injuries along the interior. An additional high-injury risk probability for an interior player drops it to 12 percent.
These are all injury risks for any injury that forces missed time, not season-ending injuries to players.
Even in the most extreme scenarios, which are not the actual risks, the Vikings are left with a rosy interpretation of their situation only 12,1 percent of the time. Consider the fact that Sam Bradford is listed as a 41 percent injury risk, when intuitively, you’d think he’d hit the high marks. Meanwhile Robert Griffin III was at 27 percent, and one comes to realize that most injury risks are far lower than people think.
If one drops those probabilities to something much more reasonable—especially considering that while Kalil has been struggling with injuries throughout his career, he hadn’t actually missed a game before this season—and those scenarios drop from 44.6 percent to 28.8 for the first three injuries, 22 to 10.4 for the second two, and 12.1 to 4.6 percent for the final injury.
Should Spielman have anticipated a 4-10 percent risk of the current offensive line situation going in? Probably not. This doesn’t include injuries to Adrian Peterson and occasional absences from players like Stefon Diggs. The Vikings’ injury situation overall is closer to something like one percent, not 50.
Barnwell even acknowledges that he overstates the injury risk when calculating the trade: “Research I’ve done suggests that injuries are mostly random, especially for younger players.”
The rest of his argument relies on the conceit that Shaun Hill would generate as many wins in a similar situation. This is his argument from September, when the Vikings made the trade and it matches my own from a few months ago when I criticized the move.
Cian Fahey at Football Outsiders would not be impressed with any arguments that equate Bradford with Hill. He persuasively makes the case that Bradford is playing at an extremely high level given his surroundings and likely would have been in the MVP race had it not been for the cascade of problems on the offensive line.
It’s easy to jump on him now because the Vikings’ season has fallen apart, but Bradford hasn’t been the reason why. When the offensive line was simply problematic and not disastrous before the bye, Bradford was a legitimate MVP candidate. His ability to negate pressure and throw receivers open so consistently was spectacular. His play as an individual has suffered the more of a beating he has taken as the season has gone on, but he is still elevating his teammates and making the situation better than it realistically should be. You could see that against the Cowboys.
Those aren’t exactly the same argument; Bradford may be performing significantly better than Hill would have, but the number of wins (for whatever reason) could be pretty close. From Barnwell:
It’s possible that Shaun Hill might have thrown a game away with some sort of awful three-interception performance, but given how little the Vikings are asking Bradford to do, it’s also difficult to imagine Hill would been in a position to make those sorts of mistakes. Bradford has played well in games where the Vikings were blowing out the competition, which isn’t a knock on him as a player. It’s not Bradford’s fault his defense blew that lead the first time around against Detroit. Given the outcomes of this specific season, though, you have to really squint to make a case for Bradford pulling out any games the Vikings wouldn’t have otherwise won with their defense.
Still, Barnwell and Ruiz rely on the argument that fundamentally, the difference between Bradford and Hill is small. So, if Cian is right that the difference is in fact very large, Ruiz and Barnwell are probably also wrong that the win total would look similar.
It makes some sense; Bradford this year is averaging 6.8 yards per attempt—the same as Shaun Hill’s career number. But Bradford has demonstrated a better touchdown and interception rate, which is one reason why his adjusted net yards per attempt and passer rating are better.
That also doesn’t give us many clues as to how Hill would perform under the pressure that Bradford has been under; Shaun Hill built the bulk of his career passes while under an abnormally low amount of pressure. It helps that they have him as one of the most accurate quarterbacks under pressure in the NFL, but they noted much of that had to do with his mobility—something he clearly doesn’t have any more.
His accuracy (per ProFootballFocus) dropped from sixth-best in the league in 2010 to 16th-best in 2014 with the Rams, and a big reason is because he was worse under pressure and under pressure more often. In his 41 attempts with the Vikings, his completion rate (51.2 percent) and accuracy rate (64.1 percent) have been abysmal and would rank dead last among qualifying quarterbacks in 2015.
His true ability is likely somewhere in between that 2014 performance and the 2015-2016 numbers he put up, but it’s clear that the difference between Bradford and Hill may have one point been small but is no longer. Not to mention, Hill looked awful in training camp (though improved into the preseason).
Even if one only uses data from when he was doing his best, his passer rating under pressure was only 58.7, with an unpressured passer rating of 93.5. Compare that to Bradford this year, where he has thrown for a passer rating for 85.4 with pressure and 104.5 without pressure for the games that Pro Football Focus has made data available for.
It’s easy seeing the Vikings losing the two close games they’ve won this year and that enough might warrant the cost of the trade; the Vikings are currently on the knife’s edge in the playoff race and two losses would make this a lost season.
Even some of the bigger wins would be suspect under Hill; Bradford had above-90 passer rating games against all five of his wins—Carolina, Arizona, Green Bay, New York and Houston.
It seems unlikely that the Vikings start off the 31-13 win against the Houston Texans with an 84-yard touchdown drive capped by a 36-yard pass to Adam Thielen with Shaun Hill throwing the ball; especially with the 2.7 yards per carry they managed in the game.
Even if you grant Hill the easy win against the Texans, it’s hard to imagine he’d pull off the Carolina win even with the gift Marcus Sherels provided with the punt return touchdown. The running backs averaged 2.7 yards per carry once more and did not pull ahead until the second half, on the back of a 79-yard touchdown drive.
Regardless, some close wins would have become close losses and some big wins would have become close games without Bradford avoiding turnovers (his 0.8 percent interception rate is one of the best in the NFL; better than Hill’s 2.5 career rate)—with leads, the Vikings needed to generate first downs and take care of the ball.
Not only that, Hill has turned from a player who prevents pressures from turning into sacks to one who invites sacks through pressure.
When comparing Bradford’s first down rate to all quarterbacks since 2010, he ends up in the 48th percentile, while Hill is at the 40th percentile. Those are close, until accounting for interceptions, where Bradford shoots up to the 66th percentile and Hill drops to the 34th percentile. And using Bradford’s numbers this year, he ends up in the 85th percentile.
A “take care of the ball” offense really cannot be sustained with Hill, who causes punts and interceptions at a far greater rate than Bradford. Increasing the number of short-field position drives by opposing offenses would have put much more stress on the defense; that’s one reason why offensive passing output correlates better with total point differential than total points scored.
The expected win rate for Bradford’s current passing output is 0.55, or essentially a 9-7 record. Hill’s career numbers show something close—a 7-9 record. But his most recent numbers, after adjusting for pressure, are much closer to 6-9-1.
So, even if one knows about all the injuries coming in beforehand, there’s a reasonable argument to make that the Vikings made the right move—two or three wins is certainly the difference between the playoffs and sitting at home.
Without foreknowledge of the Vikings’ unique injury situation, the difference becomes even clearer; with normal pressure rates, Bradford’s performance projects to a 10-6 record while Shaun Hill’s declining performance projects to a 7-9 record.
None of this takes into account defensive or special teams performance, which would likely push both win totals up by 1.5 or even two wins (which is to say a functioning offensive line may have resulted in a 12-4 record).
There are essentially three questions being asked that have been conflated: 1) Has Bradford’s performance suggested a first-round pick caliber quarterback? 2) Has that difference resulted in a meaningful impact in playoff or Super Bowl odds? 3) Has that shift in odds been impacted by foreseeable situational circumstances?
Barnwell and Ruiz largely answered the second two questions while Fahey answered the first one. I think all of this information combined suggests that the Vikings likely made the right move.
