Field Position Proves Crucial In Week 1 Tilts

The importance of field position in football is, despite its position as one of the indomitable "little things," still an understated facet of the game. Just take a look at last year's field position data from Football Outsiders. All of the top six teams in net drive start position -- average offensive initial field position minus average defensive initial field position -- won at least 10 games. In 2012, every two extra yards of net field position added an extra win, and net field position explains a whopping 31 percent of the team-by-team variation in winning percentage.

The implication of such a result is clear: as important as pure offensive and defensive talent can be, it can be negated by starting out in poor situations. And, conversely, a team can mask issues on either side of the ball by consistently beginning drives in good position (or, of course, forcing the opponent to begin in poor position). This result was in stark display throughout Week 1, as at least seven games tilted heavily on the field position battle. A focus on three of these contests shows the multiple ways in which the field position game can tilt a contest, as well as the ways a team can influence the field position battle -- of course it's a battle -- as the game goes on.

The following charts show starting field position by drive. The symbol placement shows starting field position and the symbol type shows the result of the drive. The bars show the expected points resulting from the starting field position.

DEN 49 - 27 BAL

Click for enlarged chart.

Begin with the season opener, in which Peyton Manning's Broncos demolished Joe Flacco's Ravens in a playoff rematch. Obviously, the big story in this game was Manning's performance. Given the ease with which Manning compiled 462 passing yards and seven touchdowns, the Broncos likely would have won this game without the field position gifts from Baltimore.

Still, the game was tight early on. Baltimore put together a pair of long touchdown drives early, but a Joe Flacco interception on the Broncos' side of the field plus a blocked punt inside Baltimore's own 20-yard-line early in the second half negated those early drives. The Broncos started two drives within 35 yards of the end zone and started outside the 20 six times.

But the real story here is how the Broncos consistent, explosive offense served as a fantastic defense as well. Baltimore started eight drives off kickoffs, and in part thanks to the recently-shortened field on kickoffs, the Ravens never took one outside their own 20. In fact, the Ravens didn't get the ball beyond the 20-yard-line until their 14th drive -- after the game was well in hand for Denver -- and five times they were forced to start inside the 20.

Overall, the Broncos were expected to score 19.0 points based on their starting field position as opposed to a meager 3.1 for Baltimore. Combine 25 expected points added (EPA) from Manning alone and it's no wonder the Broncos won in a rout. When an offense is as explosive as it was Thursday night in Denver, it feeds into itself and can become nearly impossible to stop, as the Ravens found out.

NE 23-21 BUF

Click for enlarged chart.

New England's offense was hardly as explosive as expected in their scrape victory against Buffalo. Still, despite the close win, the Patriots outgained the Bills by nearly 150 yards. More importantly, the Patriots picked up 11 more first downs. The Tom Brady-led offense averaged just 4.8 yards per play and 5.3 yards per pass -- not enough for multiple sustained, long touchdown drives. It did, however, extend drives just long enough to keep the field position consistently in their favor.

Take a look at the yards gained on drives in which the Patriots punted: 42, 28, 7, 36, 4 and 29 (average: 24.3). Now consider the yards gained on drives in which the Bills punted: 18, 2, 28, 7, 8, 26, -2, 37 and 4 (average: 14.2). As such, nearly every time the two teams traded punts, the Patriots inched closer and closer as the Bills were pushed farther and farther away.

Both teams had turnover issues -- two fumbles and an interception for New England, a pair of fumbles for Buffalo. New England's ability to pick up first downs with regularity and force multiple three-and-outs from the Bills put the Patriots in a position to win. The Patriots started with 15.1 expected points against just 7.5 for the Bills, a significant difference in a game that came down to a last-minute field goal.

SF 34-28 GB

Click for enlarged chart.

Nobody should be surprised the matchup between Colin Kaepernick and Aaron Rodgers produced by far the most thrilling quarterback duel of the week. Rodgers threw for 333 yards, a trio of touchdowns and just one interception in 37 pass attempts, good for 17 EPA and 0.61 WPA. Kaepernick, however, was impeccable, as he threw for 412 yards and three touchdowns without an interception in 39 attempts. His 0.76 WPA led the league in Week 1 and his 23.3 EPA trailed only Manning.

Despite Kaepernick's stupendous performance, however, the Packers had more than enough chances to win. Self-sabotage was a nemesis in the field position game, beginning when an Eddie Lacy fumble inside the 20 gave the 49ers a drive 14 yards away from the end zone, which eventually turned into a somewhat controversial touchdown.

The 49ers had some good field position chances -- five drives outside the 30 -- but only turned them into three points. Conversely, the Packers reached the end zone both times the 49ers handed them the ball outside the 30-yard-line (including once on a missed Phil Dawson field goal).

The Packers started inside the 20 four times. Two of those four cases were the direct result of kick returner Jeremy Ross's poor decision to take the ball out of the end zone, resulting in drives beginning on the four-yard-line and the nine-yard-line respectively. Such poor field position was a rarity for the Packers last season, who ranked third in the league with an average +4.6 yard field position differential.

The 49ers were the top dogs, as they started drives +6.3 yards closer to the end zone on average than the opponent. This is how the 49ers were able to win consistently even with a mediocre Alex Smith under center -- the defense would produce three-and-outs, the special teams would avoid costly mistakes and pick up yards on punts and kicks, and Smith and Frank Gore would pick up just enough first downs to inch closer every time they got the ball. With Kaepernick under the center, look for the 49ers offense to do even more damage with the short fields their other two units present them with.

Overall, San Francisco had 13.9 initial drive expected points to just 4.4 for Green Bay -- yet another difference (9.5) larger than the final margin of victory (6).

At least four other games -- Minnesota-Detroit, Pittsburgh-Tennessee, Philadelphia-Washington and Chicago-Cincinnati -- hinged heavily on field position. You can see those graphs here. If you would like to see the same charts for other games, go to the tab marked "DIY" (Do It Yourself), and check the two teams you'd like to look at from the list on the right.

Whether it's dodging turnovers, avoiding poor decisions on special teams, or merely keeping an offense chugging as opposed to suffering three-and-outs on the regular, teams have ways of winning the ever-important field position battle. An explosive offense or brick-wall defense can reverse the tides, but every yard an offense has to march -- or a defense has to protect -- can be critical, as it was early and often in Week 1.

  • Spread The Love
  • Digg This Post
  • Tweet This Post
  • Stumble This Post
  • Submit This Post To Delicious
  • Submit This Post To Reddit
  • Submit This Post To Mixx

9 Responses to “Field Position Proves Crucial In Week 1 Tilts”

  1. Anonymous says:


    I'm curious .... to what extent is the the effect of field position on wins independent from the effect of turnovers on wins?

    It seems like getting a turnover in the other team's territory is going to have big effect on average field position. Hence, I'd be curious if, controlling for turnovers, field position still predicts anywhere near 30% of the variance in wins.

  2. Anonymous says:

    Field position is the product of offense, defense and special teams. So I don't think this statement makes much sense:

    "The implication of such a result is clear: as important as pure offensive and defensive talent can be, it can be negated by starting out in poor situations. And, conversely, a team can mask issues on either side of the ball by consistently beginning drives in good position (or, of course, forcing the opponent to begin in poor position)."

    There is no starting out in good or poor situation component of a team's game. There is an offense that can avoid three-and-outs, score and thus kickoff, avoid turnovers (etc.), a defense that can force three-and-outs, prevent scores, force turnovers (etc.), and a special teams that can pin opponents within their 20/avoid being pinned within their 20 (etc.)

    Instead of poor field position negating offensive or defensive talent, it is instead largely the product of offensive and defensive talent. Naturally won-loss record is going to be impacted by the component performances of a team. I think you are misinterpreting a result as a cause and double counting offense and defense.

  3. Matt M says:

    Agree with the previous comment. Scoring touchdowns is going to lead to poor field position for opponents because the majority of kickoffs result in touchbacks (especially since the kickoff spot was moved up). Conversely, punting and turning the ball over is obviously going to lead to good field position for the opponent.

  4. Nate says:

    The fundamental notion of this article - looking at games through the lens field position - is sound, and has been done by a number of insightful football analysts. That said, the notion got a bit confused in the prose - at the start of the article field position seems like an external force that can overcome the teams' best efforts, but by the end "teams have ways of winning the ever-important field position battle". (In practice, of course, it is a little of both, with teams occasionally turning the ball over or winning the game by making an extra five yards on each punt.)

    The charts are a bit hard to read. Is there a specific point or context that they're supposed to illustrate? For example, if the goal was to illustrate field position's relation to score, it might make sense to have
    a chart show the 'expected points' and 'actual points' on top of each other as lines on a chart.

  5. Brian says:

    "Overall, the Broncos were expected to score 19.0 points based on their starting field position as opposed to a meager 3.1 for Baltimore."

    Yet the score was 49-27. What's going on here? Are Denver & Baltimore's offenses too strong? Is it time to refresh the Expected Points formula?

  6. Andrew M. says:

    When the kickoff was moved 5 yards forward (and presumably the starting field position was moved back a ways), did the scoring change as the Expected Points formula indicate? If not, I would think that the formula should be updated.

  7. Unknown says:

    Brian is far more qualified to discuss this, but yes, moving the kickoff to the 35-yard line does have an effect on EPA. I believe Brian's numbers use the 27-yard line as the average start for the offense after kickoff (my apologies if this is not the case), which is 0.7 EP. I believe now the average start line is the 22 or 23-yard line which puts the EP at about 0.5 or 0.6 EP. Obviously, this has an effect on the numbers; mainly, offense EPA numbers will be slightly higher due to the higher value of a touchdown, and the kicking team's EPA points will be a bit higher as well.
    The differences aren't that big, but they are there.

  8. Brian Burke says:

    Andrew-Yes. The model was updated.

    Brian-Possession itself has positive EP value, so "starting field position" will almost always represent positive EP (anywhere forward of a team's own 15, to be precise).

  9. Jeff S says:

    I'm curious if "average drive start" is a consistent statistic. In other words, do we see teams with good average drive starts continue to have good average drive starts. Or is it effectively random? Similar to the idea that defensive turnovers are largely random (my recollection is that there is very little to zero correlation between turnovers "forced" in 1st half of season to turnovers "forced" in 2nd half of season), are drive starts random? Or is their some consistency? And if so, could it potentially be used as a predictive input for a win probability model?

Leave a Reply

Note: Only a member of this blog may post a comment.