Each week, we’ll highlight a few of the most important coaching decisions that contributed heavily to the final result. While many of these high leverage situations occur on fourth down, we’ll introduce some other situations where a coaching decision swung his team’s chances of winning in a major way as the season goes on.
There are a few main tools for doing this analysis: The most obvious is win probability, which takes into account the situational components of the game along with the strength of each team in each facet to predict the chance of winning. Another data point that is useful and in most cases aligned with win probability is expected points, which takes into account the field position, down and distance to predict the number of points that we expect the offense to score on the next offensive play. We can also compare the expected points for a team before and after a play to measure how successful a play was. We’ll also make use of conversion probability, which leverages the strength of each team in each facet to predict the chance a team picks up a first down on a given play.
During the season, more analysis may work its way into this column, but this gives us plenty to start with. While we leverage machine learning to make sound predictions for each of the above, we will do our best to make this article accessible to all and we’ll do our best to answer any questions.
Dan Quinn a Beautiful Redemption Story
[2:10 left 4th quarter, 4th and 3 @ ATL 46, PHI 20 ATL 17]
Sunday night of Week 2 seemed to be going exactly the way you’d expect if you’ve watched the Falcons over the past three seasons. The Falcons had outplayed the Eagles for the majority of the game, but a key fourth-down decision (see honorable mention below) by Decision King Doug Pederson and a herculean play by Carson Wentz propelled the Eagles back into the lead with just a few minutes left in regulation. The Falcons' inability to get over the hump late against the Eagles would be nothing new and neither would Dan Quinn deciding to kick instead of leaving his high-powered offense on the field on a fourth and short; over the last two seasons, the Falcons have made the correct go for it decision on fourth down decision at the seventh-lowest rate.
Quinn had two timeouts plus the two-minute warning, which actually made this a tough decision in the mind of some. The tough decision according to math, however, was not whether to keep the offense on the field but rather what play to call after so many of these pivotal play calls have gone horribly wrong for the Falcons. The decision to go for it was clear cut; the Falcons stood to increase their chance of winning to 43% with a three-yard conversion. A punt would give them just a 25% chance to win, and a turnover on downs would cut their chances to 15%. The Falcons had plenty of room to work with and a top tier offense that had torched the Eagles secondary on multiple occasions giving them a 54% chance of converting the 4th-and-3 according to our machine learning model. Last week, we discussed a simple rule that can be used here to determine if the decision to go for it is sound: If the difference between a conversion and a kick (43% minus 25%) is larger than the difference between a kick and a turnover on downs (25% minus 15%) as it is in this situation, then converting at a 50% rate ensures that going for it is the right decision. That is exactly the situation we have here, a no-doubter.
We can put an exact value on the chance of winning if you go for it by multiplying the probabilities of each outcome by the chance of those outcomes (converting or getting stopped), and we’ll see that the Falcons gave themselves a 30% chance to win by lining up to go for it compared to a 25% chance by lining up to punt.
Dan Quinn exercised many demons by making the correct decision to go for it and the perfect play-call was equally as exquisite, allowing the two best Falcons players to make a play to finally beat the Eagles.
Doug Pederson, because of course
[4:57 left 3rd quarter, 4th and 4 @ ATL 4, PHI 6 ATL 17]
It wouldn’t be a real week in the NFL if we didn’t have Doug Pederson making a crucial decision to give his team a better chance of winning the football game. Down by 11, two-thirds of the way through the 3rd quarter, Pederson understood that you win games with touchdowns, not field goals. A conversion gave his Eagles a 35% chance to win, a field goal just a 19% chance and a turnover on downs just 17% (notice that your opponent being backed up after a turnover isn’t the end of the world). We would give the Eagles' offense a 46% chance of converting this 4th down. Multiplying outcomes of going for it by the chances of each outcome gives the Eagles a 25% chance to win if they go for it, compared to a 19% chance if they kick a field goal. This decision turned the game around and ultimately would have keyed a comeback if not for an incredible decision and play by the Falcons and a few extra inches that were not to be for the Eagles.
Frank Reich, not here to lose without Luck
[2:23 left 4th quarter, 4th and 1 @ IND 35, IND 19 TEN17]
Reich went for this 4th-and-1 on his own end despite leading by less than a field goal, the ultimate test of whether you are playing to win or not to lose. With a conversion, Reich would secure an 85% chance of victory, a punt would give them a 67% chance and a turnover would drop them below 50% to 46%. There is a huge potential swing here, but the Colts also had a 69% chance of converting and the quarterback sneak, which they employed has a 90% success rate historically. The fact that a conversion was far greater than a 50/50 proposition made this decision far and away the right call. Going for it gave the Colts a 73% chance to win, compared to the 67% chance of winning with a punt. Frank Reich is a legend in the making.
Byron Leftwich establishes the run
On Thursday night, the Buccaneers earned a higher passing grade (81.0 to 51.2), receiving grade and coverage grade than the Carolina Panthers. They passed for more 1.8 more yards per attempt, took the same amount of sacks, rushed for 1.5 more yards per carry and had a turnover differential of +1. In what sounds like an easy road win for the Bucs, Cam Newton and company came, in fact, fell two yards and a heroic fourth-down stop short of winning the game. How is that possible? Enter Simpson’s paradox and play selection.
Tampa’s pass offense struggled in their season opener and produced the league-worst EPA/pass play and passing grade. We can never be sure about the primary reasons behind a coaches’ actions, but it feels like this caused a major overreaction and mistake from Byron Leftwich.
In Week 2, he decided to establish the run, ignoring league-wide evidence and team-specific data telling him that he should let Jameis Winston pass the ball on early downs (1st and 2nd down with 7 or more yards to go). Generally, every team fares better on early-down passes than on early-down runs. The core of their current passing offense has been installed since at least 2017, and the Buccaneers rank eighth in EPA/pass play and third in percentage of pass plays with positive EPA (success rate) in that span. Leftwich had other plans: He chose to fade the pass.
On 44 early downs, Jameis Winston dropped back to pass only 17 times even though the offense was highly effective when he did just that. Winston posted a 90.1 passing grade along with 0.56 EPA/play and a success rate of 65% on early downs. They handed it off 27 times, gaining -0.07 EPA/play with a 41% success rate. While this is bad enough, what turns it into an avoidable, bad process is that the Bucs faced base defense on 43 of these snaps. Consequently, they ran into a box of eight or more defenders – something that is considered a cardinal sin by most analysts – 16 of 27 times.
Too many runs on early downs generally leave your quarterback with unfavorable third-down situations, and this game was no exception. Multiple drives stalled after the Bucs couldn’t move the ball on early-down runs because Winston’s early-down success didn’t translate to third-and-long situations when the defense expects the pass and can cover the area beyond the sticks. Four of seven of his negatively graded throws came on difficult third-and-long plays, and three of them were caused by conservative play-calling on early downs.
As a particularly egregious example, we consider their first drive of the second half. Tampa comes out with three tight ends and rushes it into a stacked box for a one-yard gain. They get caught for a false start on second down, making it 2nd-and-13. They run again out of a heavy formation, this time for no gain. The ensuing 3rd-and13 is, of course, a low percentage play, and Winston can’t connect with Mike Evans on a pass that wouldn’t have had the chance of a first down anyway.
With decisions like these, Leftwich repeatedly put his quarterback in a bad spot and almost laid to waste what was a fairly dominant game by his pass offense as compared to the opposing pass offense. The game ended with eating a win, but that won't stop us from addressing the elephant in the room: Pass the ball on early downs.