We are sports data scientists with a deeper, faster, more accurate analysis of the game’s most critical decisions. Our EdjFootball model assesses a team’s Game-Winning Chance (GWC) or win probability at any moment of the game. Each week we analyze the NFL games and bring you the most interesting scenarios, including play execution and coaching decisions and how they affect a team’s GWC. Here are our most notable plays for NFL Week 6.
Green Bay vs. San Francisco
49ers lose nearly 5% GWC (Game-Winning Chance) on 4th and 1 punt
The situation: SF leads GB 30-23, 5:45 remaining in the 4th quarter, 4th and 1 at SF 13 yard line – SF punts
In this situation the 49ers missed an excellent opportunity to increase their GWC (Game-Winning Chance). It may seem crazy to most to consider anything but a punt, however attempting a first down with a rush or pass play increases the 49ers GWC by almost 5%. An average NFL offense would be expected to convert a fourth and one approximately 70% of the time. This figure is well established from more than a decade of play-by-play data and confirmed by an EdjSports simulation. While it is certainly costly to fail the conversion, it is far from catastrophic. Green Bay would still need to score a touchdown and the 49ers would have at least one more possession in regulation. Alternatively, a punt from the end zone typically allows the Packers to begin their drive from near midfield. For further confirmation, we adjusted the parameters of the simulation to reflect a very weak NFL offense against a top NFL defense and reassessed the merits of a first down attempt vs a punt. Once again, the rushing play emerged as a clear winner.
New England vs. Kansas City
A tricky 4th down decision for the Chiefs with 3:59 remaining in the game
The situation: KC trails 33-37, 3:59 remaining in the 4th quarter, 4th and 7 at KC 28 – Chiefs punt
Conventional wisdom would suggest a team should almost never attempt a first down on a 4th and 7 in their own territory. But as we often see in our EdjFootball analysis, it is all about context. While trailing 37-33 to the Patriots on Sunday night, the Chiefs faced a long 4th down that resulted in the very typical choice of punting. However, this was not a typical situation. By punting, the Chiefs are setting out to stop a high-powered Patriots offense, regain possession, and score a touchdown with under four minutes remaining in the game. Attempting a first down is a volatile choice, but an average NFL team, historically, would be expected to succeed about 44% of the time. Furthermore, the Chiefs’ offense has been clearly above average so far this season. It turns out that a pass play produces approximately 9% more wins on average by boosting the Chiefs’ GWC from 9.8% to 18.8%. In the context of winning the game, the more volatile and apparently riskier choice of a first down attempt literally doubles the Chiefs’ winning prospects.
Miami vs. Chicago
Miami fumbles and Chicago recovers in the end zone for a touchback
The situation: Game tied 28-28, 5:06 remaining in overtime, 3rd and 1 from the CHI 1 – Miami fumbles
On a Sunday full of memorable plays around the NFL, this particular play stands out. In an extremely exciting, back-and-forth game between Miami and Chicago, Miami appeared a near certainty to clinch the victory by punching it in from one yard out. Moving beyond the outcome for a moment, Miami head coach Adam Gase would have faced a very difficult decision if Drake held onto the ball but was stuffed at the goal line. Miami would have faced a fourth and goal from the one. Of course, converting this fourth down means winning the game, but failing means your opponent can beat you with a field goal on their subsequent possession. A field goal would require a defensive stop to win the game. We suspect that most play callers would choose to kick a field goal in this hypothetical situation. The model would favor going for it, but only by a very slim margin. The GWC (Game-Winning Chance) in going for it is 78.5 percent versus 77.7 percent for attempting the field goal.
Atlanta vs. Tampa Bay
Dan Quinn decides to send Matt Bryant to attempt a 59-yard field goal
The situation: ATL leads 31-29, 1:16 remaining in the 4th quarter, 4th and 2 from the TB 39 – Field goal attempt
This was a very interesting decision by Dan Quinn. The field goal in question was no gimme for Matt Bryant, one of the top kickers in the league. We estimate that he is successful 64 percent of the time with the favorable kicking conditions (indoor on artificial turf). Historically, NFL teams convert third-and-two situations from midfield approximately 60 percent of the time, dating back to 2000.
If they gain a first down, Atlanta can easily run out the clock. If Atlanta were to surrender the ball to Tampa Bay, either by missing the field goal or turning the ball over on downs, the model estimates Tampa Bay’s GWC at 12 percent. To simplify, let’s assume that Tampa Bay takes over from their own 25, trailing by 5 points in the event that Bryant makes the kick. Their GWC from this game state would be 2.5 percent. With the relevant numbers in hand, we can “peek under the hood” so to speak to understand how the model may evaluate this decision.
– Bryant makes the kick: 64%
– Atlanta converts the fourth down: 60%
– ATL holds on to win, up 5 when TB starts from their own 25: 97.5% GWC
– ATL holds on to win, up 2 when TB starts from their own 39: 88% GWC
From Atlanta’s perspective:
– Going for it means they win 60% * 100% + 40% * 88% = 95.2%
– Attempting the field goal means they win 64% * 97.5% + 36% * 88% = 94.1%
It’s worth noting the possibility that Atlanta’s highly proficient offense converts the fourth down at a rate higher than 60 percent. A nicety of breaking down this decision in this way is that we can perform a “sensitivity analysis” on our assumptions. We can try to tilt the situation in favor of the field goal for instance by lowering the chance Atlanta wins when they miss the field goal or fail the fourth down attempt. If we lower this from 88 percent to 70 percent while keeping everything else constant, the decision becomes much closer; 88 percent when they go for it, but 87.6 percent when they attempt the field goal. Even with these more charitable assumptions, it appears that Atlanta would have been better off going for it.
Cincinnati vs. Pittsburgh
Bengals pass from own 32-yard line at 5:42 of 2nd quarter
The situation: Game tied 7-7, 5:42 remaining in 2nd quarter, 3rd and 6 at CIN 31
While the conditions were less than ideal for catching passes in Cincinnati on Sunday, Cody Core had no excuses when Andy Dalton threw him a perfect unobstructed pass for what would have been a very important first down. With the score tied at 5:42 of the 2nd quarter it was not to be, as Core let it slip through his hands and off his chest. Instead of continuing the possession with a first down at midfield, the Bengals were forced to punt from their own 32-yard line. This seemingly simple mishap cost the Bengals approximately 9 percent of GWC as it would have moved their winning prospects from 66.6 percent to 75.6 percent GWC.
Official review of Steelers’ completed pass
The situation: Game tied 7-7, 2:00 remaining in 2nd quarter, 3rd and 6 at CIN 30
One of the more acrobatic and controversial plays of the week occurred at the 2:00 mark of the first half between the Steelers and Bengals. Tied at 7-7, Ben Roethlisberger threw a pass to JuJu Smith-Schuster resulting in a confusing entanglement with the Bengals’ Darqueze Dennard. At first look, it was difficult to distinguish between an incompletion, an interception and a completion at the one-yard line. After a challenge and an extensive review, it was ruled a completion. Let’s look at the three possible resulting scenarios and how they could have impacted the outcome of the game.
– Interception: If the Bengals take over on their one-yard line, the Steelers have 43.6% GWC.
– Incompletion: Results in a 4th and 6 for the Steelers and placing them in a possible field goal situation with 38.7% GWC.
– Completion (actual result): Puts the Steelers in excellent position for a touchdown and boosts their GWC to 53.9%
Most interestingly, an interception is a far more favorable result than an incompletion as it is effectively the same as a perfectly executed punt.
EdjSports empowers smarter decision-making with proven predictive and prescriptive analytical models and custom software solutions in the sports industry. Our game-based approach analyzes pools of data, helping teams make critical call decisions based on the highest probability of achieving their goals. At EdjSports, we assist decision makers and enhance their ability to gain the competitive edge that ultimately impacts the bottom line – winning. For more information, visit www.edjsports.com.