EAGAN — In 2021, Pro Football Focus released a data study that really made me see football differently than I ever had before.
They analyzed runs and passes that had perfect blocking (meaning nobody got a negative blocking grade) and plays that had at least one player who made a mistake on the play and received a negative grade.
The results were that successful runs dropped from 60.2% to just 25.7% of the time when at least one blocker produced a negative play.
The other crazy part of the study is that teams only perfectly blocked on 34.9% of their plays. That’s why it’s hard to dominate with the run game
On the other side of the coin, teams were able to perfectly block pass plays 68.9% of the time and the success rate only dropped by 10.5% when at least one blocker had a misstep.
Why is that?
Well, if the ball comes out quickly, then blockers are much less likely to have a negative grade because there isn’t time for them to be defeated and the success rate doesn’t drop that much because the quarterback controls how they react to a negative block, whether that’s in the form of pocket movement, scramble or getting rid of the ball before the rush arrives.
How does this all apply to the Minnesota Vikings?
On Wednesday, head coach Kevin O’Connell discussed the Vikings’ passing struggles and talked about how the team made “one-off” mistakes that added up on them. Here’s the entire comment:
“The little things that I could get into are what’s causing some massive chances not being had in the game that changed the whole outlook of how you look at the performance,” O’Connell said. “It doesn’t mean we’re not capable of it, but as I’ve told our guys, until we stop having those one-offs here and there by everybody, it’s not just Kyler, it’s everybody on the offense, we’ve got to continue to coach our way towards these guys having total clarity.”
If we go back to the study and the implication that there is more control by the quarterback or the scheme over negatively-graded individual efforts, you might be able to engineer a theory when combined with O’Connell’s takeaway that individuals were making little miscues. That theory would be: The more an offense can be designed for players to execute their individual assignments without errors, the more success it’s going to have.
If we take this a step further, we might even be able to suggest that the better the players understand their jobs and the most straight forward those jobs are, the higher success rate.
Watching back the Chicago Bears’ win over the Philadelphia Eagles brought this idea to the forefront as well.
When Case Keenum used play-action, he went 12-for-13 with 126 yards and had zero turnover-worthy plays. When he didn’t have play-action, he was 12-for-21 with 121 yards and two TWPs. He only threw twice over 20 yards in the air and went 13-for-19 between 0-10 yards.
Now the point isn’t to turn Kyler Murray into a QB who plays like 38-year-old Case Keenum. It’s that the throws were right there in front of him and he got rid of the ball faster than any QB in the NFL during Week 3.
Here’s a fun stat for you when it comes to degree of difficulty: Keenum and Kyler Murray had virtually identical Completion Percentage Over Expected metrics for last week.
You might expect that Murray’s number would have been atrocious but he was actually on the plus side (+1.0) and Keenum was +1.1.
How can that be? Based on the distance of the throws and locations of defenders, Murray’s throws were only expected to have a 51.9% completion percentage.
It’s just one game but we see this from O’Connell’s offense repeatedly. JJ McCarthy’s expected completion percentage was 8th lowest in 2025, Sam Darnold’s was 30th out of 43 QBs, Kirk Cousins was 17th in 2023 and and 27th in 2022.
In 2024, Murray’s expected completion percentage was 10th. In 2023 it was 2nd, in 2022 it was 4th, and in 2021 it was 12th. In only five games. Murray was No. 1.
What that tells us is simple: The passes that Murray was asked to make had a higher chance of being completed than the passes that any of the Vikings’ quarterbacks during KOC’s tenure.
In 2024, Murray was 12th best in Completion Percentage Over Expected, meaning he completed more than the expectation.
The argument that O’Connell might make is that the Cardinals were only 15th in total passing Expected Points Added in 2024 and the Vikings were 6th despite having a lower expected completion percentage.
But that required Darnold being 9th in Completion Percentage Over Expected and he was still only 17th in regular completion percentage.
Cousins’ outstanding 2023 start required him to be 4.6% above expectation.
Now this isn’t just a scheme and play calling stat. It’s a quarterback decision stat. But when the Vikings’ QBs consistently are nowhere near the top and quarterbacks like Murray previously, Patrick Mahomes under Andy Reid (top 10 every year since 2019), Jared Goff (6th in 2025) and Tua Tagovailoa have consistently high expected completion percentages, then you start to wonder how much the system is impacting the ability to simply keep the offense on track with completions.
There are other factors as well.
The Vikings are 7th in third downs with more than 7 yards to go since 2022. The combination between not being a high completion percentage offense and having a struggling run game play into that.
You’ll never guess where Murray ranks when passing on third down with more than 7 yards to go. Since 2022, he’s 35th in passer rating on third downs with more than 7 yards to go (min. 100 passes).
He’s 13th in terms of passer rating on third downs with 6 or fewer yards to go and that doesn’t factor for his 30 rushes for 207 yards on third-and-manageable or less either.
It’s pretty clear that until Jordan Mason comes back, it’s going to be a challenge to get Murray in those favorable situations simply by running the ball.
Aaron Jones is 9th lowest when it comes to Rush Yards Over Expected.
“We are seeing more eight-man fronts and a great example of it just throughout the game of some plays that we either made against it or could have,” O’Connell said. “We’ve got to continue to run the football. People are going to try to take it away, but it’s a part of how we want to play offense. And I think we’ll continue to kind of grow as we go, get Jordan Mason back at some point. Aaron’s done an unbelievable job, shoulder in that load.”
The passing game has to become the running game in some ways. It was far from that against the Bucs.
The plus side is that Murray can make throws if they are laid out for him (and he makes the right decisions). His On-Target Percentage since 2020 among QBs with at least 1,000 throws is 9th and his Success Rate is 12th. He is capable of racking up completions.
The Vikings need to get that going this week but it might start with O’Connell asking: Is the system/play-calling causing some of the errors? How can we increase the expected completion percentage of throws Murray’s being asked to make?
Or was it simply that Justin Jefferson went out?
I thought a lot of problems last week was stale play calling. It was not working. Outside runs (when the inside is owned by Vea) and quick passes to the edge... Let Kyler get a bit of rhythm. It seemed we ran the same losing plays over and over
A few things seem to be at play here so far:
- KOC's habit of mashing the same button on the controller over and over even when it's clearly not working. Last week felt a lot like the game at Detroit in '24—your QB is clearly off, but your game plan was to throw it all over the yard, so you keep hitting your head against the wall. "Hunting explosives" in an empty forest.
- Kyler not seeing the opportunities that were there. Andrew Krammer did a film review with several plays where guys are wide open or entering voids in coverage and Kyler doesn't see them, sees them but doesn't pull the trigger, or just misses the throw. Everybody's always like, "Where are the easy-button throws?" There were plenty of plays where he had open checkdowns or stick routes and he didn't get rid of the ball on those either. Wouldn't the expected completion percentage on those be pretty high?
- Maybe he's just not the right fit from a stature or play style standpoint. I'll give it a few more games on that one, but I was dubious that he was the guy to bring in since Coller first started bringing up his name. Kirk or Geno made more sense to me, and look way better so far this season. I hope I'm proven wrong, otherwise this is going to continue to be a tough watch.