Stat Analysis
Advanced analytics on player and team performance

Going Long in Kansas City

by Mike Tanier
Stats compiled by Jim Armstrong

Kansas City hasn't had an NBA franchise since the Kings left town in the mid-1980's. Of course, they haven't needed one: They have the Chiefs.

With an exciting, balanced offense and an often-incompetent defense, Chiefs games often have final scores that would look right at home on the basketball section of the sports page. They are always winning 45-25 or 49-38, except when they are losing 38-31 or 34-31.

There are many stats that exemplify both the quality of the Chiefs offense and the futility of their defense, but the most telling figures may be the team's performance on long scoring drives, both for and against. Jim Armstrong's study of drives during the 2004 season reveals some interesting facts.

The Chiefs had more scoring drives of 80 or more yards than any other team in the league. When they got the ball at or inside their own 20-yard line and went at least 80 yards, they scored 22 times in 67 attempts (32.8%). Twenty-one of those drives resulted in touchdowns (31.3%). Only the Colts had a scoring rate over 30%; no team had a touchdown rate over 30%.

Meanwhile, on defense, the Chiefs allowed 11 scoring drives of 80+ yards. Several teams allowed more, but with only 43 opposing drives starting at or inside the 20-yard line, the Chiefs allowed the worst percentage of long drives in the NFL (25.6%). Ten of those drives resulted in touchdowns, again yielding a league-worst 23.3% success rate on long drives.

To continue the basketball analogy, an 80-yard drive is like taking an inbounds pass, dribbling the length of the floor, driving and scoring on a lay-up. It's offensive dominance and defensive frustration in its purest form: starting at the 20 and methodically moving down the field. And it was a common occurrence in most Chiefs games.

Take the team's 34-31 loss to the Buccaneers in Week 9. The Chiefs took the opening kickoff and marched 77 yards in 10 plays for a touchdown (77-yard drives aren't part of this study, but they certainly fit the "long drive" description). The Bucs responded with a 73-yard touchdown drive. Later in the quarter, the Bucs recovered a fumble and traveled 71 yards for a touchdown. At the start of the third quarter, the Bucs went 80 yards in two plays: one of them a 78-yard catch and run by Michael Pittman for a touchdown. The Chiefs answered later in the quarter with an eight-play, 80-yard touchdown drive. Unfortunately, the Bucs took the next kickoff and drove 80 yards for a touchdown that ultimately won the game.

Why were the Chiefs so successful at long drives and so inept at stopping them? The prosaic answer is that their offense was good and their defense was bad. A more detailed answer can be found in the specific strengths and weakness of the Chiefs. To be able to consistently sustain long drives, an offense has to:

1) Avoid third downs. Only the Colts and Giants attempted fewer third down conversions than the Chiefs last season. In their 49-38 win over the Titans, the Chiefs attempted just nine third down conversions, despite the fact that they had the ball on 13 separate drives (not counting a "kneel to end the half" drive).

2) Convert the third downs you do attempt. The Chiefs were also third in the NFL in third down conversions (47.2%). In their Week 13 win over the Raiders, the Chiefs executed touchdown drives 86, 80, and 88 yards in the second half. The Chiefs were 8-of-12 on third down conversions in that game.

3) Gain yards in bunches. The Chiefs were fourth in the NFL in 20+ yard pass plays (60), tied for seventh in 40+ yard pass plays (10), and tied for tenth in 20+ yard running plays (10). It's hard to drive 80 yards in six-yard increments. Someone has to make a big play.

Let's use a drive from that Raiders game to illustrate these points. The Chiefs started the third quarter on their own 16-yard line. Their first two plays netted first downs (avoiding third down). One of those plays was a 21-yard pass from Trent Green to Eddie Kennison (yards in bunches). On 2nd-and-7 later in the drive, Green and Larry Johnson connected on a 29-yard catch-and-run (yards in bunches, avoiding third downs). When the Chiefs faced 3rd-and-4 at the Raiders 11, Green hit Kennison for six yards (converting the third downs you do attempt). Johnson punched it in on the next play. Repeat ad infinitum.

Of course, turn these principles around and you have a blueprint for a defense that can't keep opponents from crisscrossing the field:

1) Rarely force third downs: Chiefs opponents attempted just 185 third downs, the lowest figure in the league. The teams at the bottom of the list are a mixture of poor defenses (Titans, Chiefs, Vikings) and great defenses that allowed very few drives (Steelers, Falcons). The Buccaneers only attempted nine third downs in their 34-31 win.

2) Allow opponents to convert the third downs they do attempt. The Chiefs were below average in third down conversions allowed at 38.4% (20th in the NFL).

3) Give up yards in bunches. Of course, this is the biggest problem for the Chiefs defense. The Chiefs allowed 77 pass plays of 20 or more yards, with 22 of those netting 40 or more yards. Both figures were worst in the league by far: no other team gave up more than 13 passing plays of over 40 yards.

Again, let's see these principles in action. In the season opener against the Broncos, the Chiefs trailed by three with 9:22 to play. The Broncos started a drive at their own 14-yard line. Quentin Griffin gained 11 yards on the first play. The Broncos would convert 2nd-and-8 and 2nd-and-4 situations (avoiding third down). A pass and another Griffin run would yield 18 and 19 yards (yards in bunches, though not an extreme example for the Chiefs). The Broncos would reach the Chiefs 25 yard line before they have to convert a third down, and they do so on a Jake Plummer draw (converting the third downs they do attempt). The drive takes so much time that the Chiefs are forced to burn timeouts, but it's all for naught when the Broncos score a game-icing touchdown to cap an improbable 86-yard drive.

There are other lessons to be learned from the Chiefs. Their running game, and their inability to stop the run, contributed to many of those long drives. The Chiefs are very good at using running backs in the passing game, and they were very bad at stopping opposing running backs from catching the ball (opposing RBs gained 836 receiving yards against the Chiefs, the third highest total in the NFL).

But the real moral of the story: When you are going to Arrowhead Stadium, try to get a seat courtside.

2004 Long Drive/Short Drive Stats

Top-ranked team in the league in bold, last-ranked team in italics. LONG drives must go 80 yards to count as a "long scoring drive."


Team LONG (Off. starts 20 or farther) SHORT (Off. starts 50 or closer) Avg.
Drives Scores Scor% TD TD% Drives Scores Scor% TD TD%
ARI 70 8 11% 7 10% 27 18 67% 6 22% 49.2 62.8
ATL 61 5 8% 5 8% 31 14 45% 9 29% 58.2 64.3
BAL 51 4 8% 4 8% 32 22 69% 10 31% 43.4 50.6
BUF 55 7 13% 6 11% 42 26 62% 16 38% 44.4 45.5
CAR 53 7 13% 6 11% 24 16 67% 8 33% 53.8 59.6
CHI 60 2 3% 2 3% 31 12 39% 4 13% 50.0 64.6
CIN 62 3 5% 3 5% 30 23 77% 11 37% 49.1 58.7
CLE 60 7 12% 6 10% 27 12 44% 5 19% 50.7 56.7
DAL 51 7 14% 7 14% 25 12 48% 8 32% 56.3 61.7
DEN 78 13 17% 12 15% 23 12 52% 2 9% 59.1 71.5
DET 66 7 11% 7 11% 23 11 48% 5 22% 54.8 59.6
GB 59 13 22% 13 22% 17 9 53% 4 24% 63.1 71.6
HOU 50 5 10% 4 8% 22 10 46% 9 41% 57.4 58.3
IND 51 16 31% 15 29% 21 12 57% 8 38% 62.8 66.3
JAC 61 3 5% 2 3% 13 8 62% 3 23% 58.6 67.0
KC 67 22 33% 21 31% 21 11 52% 8 38% 61.6 65.2
Team LONG (Off. starts 20 or farther) SHORT (Off. starts 50 or closer) Avg.
Drives Scores Scor% TD TD% Drives Scores Scor% TD TD%
MIA 63 5 8% 5 8% 30 19 63% 13 43% 46.6 52.4
MIN 63 18 29% 16 25% 14 8 57% 5 36% 65.7 69.5
NE 56 10 18% 7 13% 28 15 54% 11 39% 56.9 60.1
NO 60 9 15% 9 15% 20 14 70% 10 50% 52.7 57.5
NYG 52 8 15% 7 14% 23 13 57% 7 30% 50.8 56.5
NYJ 64 10 16% 10 16% 19 13 68% 9 47% 54.8 60.1
OAK 55 5 9% 5 9% 20 11 55% 6 30% 55.6 62.4
PHI 65 8 12% 8 12% 28 17 61% 9 32% 54.9 62.9
PIT 54 8 15% 8 15% 34 16 47% 7 21% 54.7 62.9
SD 42 10 24% 10 24% 30 21 70% 13 43% 55.1 60.8
SEA 61 6 10% 6 10% 22 18 82% 12 55% 56.5 61.8
SF 66 4 6% 3 5% 16 8 50% 4 25% 56.9 64.3
STL 67 12 18% 9 13% 8 6 75% 5 63% 63.6 67.6
TB 64 8 13% 7 11% 21 11 52% 9 43% 59.4 61.4
TEN 70 11 16% 11 16% 22 13 59% 7 32% 56.7 65.2
WAS 50 5 10% 3 6% 24 13 54% 7 29% 53.4 57.6
TOTAL 1907 266 14% 244 13% 768 444 58% 250 33% 55.6 61.9


Team LONG (Off. starts 20 or farther) SHORT (Off. starts 50 or closer) Avg.
Drives Scores Scor% TD TD% Drives Scores Scor% TD TD%
ARI 98 9 9% 8 8% 25 19 76% 12 48% 50.9 53.6
ATL 71 13 18% 12 17% 22 14 64% 10 46% 56.6 61.5
BAL 78 8 10% 7 9% 21 13 62% 6 29% 51.6 55.3
BUF 58 8 14% 6 10% 31 15 48% 7 23% 54.6 62.2
CAR 56 7 13% 7 13% 24 11 46% 8 33% 61.0 67.8
CHI 56 7 13% 6 11% 33 16 49% 8 24% 55.2 60.7
CIN 57 8 14% 8 14% 22 14 64% 5 23% 53.5 62.4
CLE 57 8 14% 6 11% 38 19 50% 10 26% 53.0 60.1
DAL 56 9 16% 8 14% 19 15 79% 8 42% 55.7 64.2
DEN 64 10 16% 10 16% 26 23 89% 12 46% 45.8 55.0
DET 75 16 21% 16 21% 24 10 42% 5 21% 64.3 71.5
GB 47 5 11% 5 11% 21 12 57% 10 48% 58.6 61.3
HOU 68 12 18% 12 18% 24 14 58% 11 46% 55.1 59.9
IND 43 7 16% 5 12% 19 13 68% 8 42% 55.8 63.6
JAC 65 7 11% 7 11% 18 8 44% 3 17% 58.6 68.9
KC 43 11 26% 10 23% 16 7 44% 3 19% 62.1 69.2
Team LONG (Off. starts 20 or farther) SHORT (Off. starts 50 or closer) Avg.
Drives Scores Scor% TD TD% Drives Scores Scor% TD TD%
MIA 76 5 7% 5 7% 34 15 44% 10 29% 53.2 58.0
MIN 49 9 18% 9 18% 23 16 70% 8 35% 55.1 62.5
NE 48 9 19% 9 19% 19 11 58% 8 42% 55.4 60.2
NO 62 4 7% 3 5% 22 17 77% 9 41% 54.1 61.4
NYG 52 10 19% 10 19% 30 18 60% 10 33% 53.1 61.8
NYJ 56 8 14% 7 13% 17 10 59% 5 29% 58.0 62.8
OAK 74 15 20% 15 20% 24 12 50% 7 29% 61.9 66.4
PHI 58 3 5% 2 3% 19 8 42% 4 21% 56.8 60.6
PIT 60 5 8% 4 7% 14 9 64% 4 29% 55.0 60.6
SD 61 9 15% 8 13% 19 10 53% 4 21% 58.3 63.5
SEA 41 3 7% 1 2% 23 17 74% 13 57% 53.5 55.9
SF 57 11 19% 10 18% 36 25 69% 16 44% 50.6 56.1
STL 53 8 15% 8 15% 28 19 68% 7 25% 50.9 61.3
TB 60 6 10% 5 8% 32 16 50% 8 25% 50.7 58.0
TEN 49 10 20% 9 18% 22 7 32% 5 23% 62.6 66.2
WAS 59 6 10% 6 10% 23 11 48% 6 26% 55.0 60.6
TOTAL 1907 266 14% 244 13% 768 444 58% 250 33% 55.6 61.9


32 comments, Last at 08 Jul 2005, 3:15am

1 Re: Going Long in Kansas City

I wonder if chiefs fans prefer the 90's chiefs with a killer d and no offense or the millenium chiefs with the exact opposite. Neither version won a playoff game. Well except for the year they had joe montana. So i guess the solution is when new england cuts tom brady, the chiefs pick him up and they'll have a shot at going to the superbowl.

2 Re: Going Long in Kansas City

What I love about the chart is that it shows how you can't really tell how good an offense is just by how many points it scores. When you've got a defense and special teams as good as the Bills have, you're going to score on a lot of drives even if your offense isn't very good.

3 Re: Going Long in Kansas City

Re 1:

Unless you're asking the question to force one of the two exclusive choices... I think we want a team that gets to the playoffs, and does so without having to deal with the devil, paying for all eternity as the Bucs are now doing...

4 Re: Going Long in Kansas City

Good stuff.

Obviously a team is less likely to score on long drives (80+ yards, 14% scoring) than short drives (50- yards, 58% scoring). Interestingly, if a team were to score on a long drive it is more likely a touchdown (14% scoring, 13% TD, 244 total TDs) than a short drive (58% scoring, 33% TD, 250 total TDs). 244 TDs scored with drives longer than 80 yards, and 250 TDs scored with drives shorter than 50 yards!! WOW!

Is this true for other years?

5 Re: Going Long in Kansas City

Re #4: I can think of two explantions for that. One is after a long, sustained drive of 10+ plays, the defense is both physically and mentally drained and less able to stop a drive in field-goal range. The second is with more field to cover, it's easier for an offense to get behind the safetys for a big play that results in a touchdown. With a shortened field, the safties playing deep can keep the offense out of the end-zone.

6 Re: Going Long in Kansas City


The average Buffalo TD drive was only 45 yards, shortest in the league. That kinda suggests the defense and ST helped out a lot :)

7 Re: Going Long in Kansas City

Great article!

Any chance of posting the medium drive stats (drives started between the 20 and the 50)?

What's interesting here is that it's not just the percentages that vary so much. Some team defenses (e.g. MIA, ARZ) are having to face far more overall drives than other teams, presumably due to inability of their offense to take time off the clock. Without the medium drive stats though, it's hard to be sure if this disparity is just that some teams mostly face all their drives as medium drives.

8 Re: Going Long in Kansas City

Mike, another fine job. I wonder how much the long drive stats are affected by penalties and specifically, if the emphasis on defensive holding last season increaesd the number of long drives significantly over previous years?
I assume that more holding penalties kept more drives alive and to a lesser extent increased the number of 'big plays' defenses gave up.

9 Re: Going Long in Kansas City

#4 + #5,
Also, the teams that are able to sustain those long drives from inside the 20 to get into FG range are more than likely having their way with the defense (that is, a team that can drive down the field on a defense to get into scoring position isn't likely to lose that ability once there). Scoring drives from inside the 20 are self-selecting for TDs because they probably correlate with a superiority over the defensive unit.

10 Re: Going Long in Kansas City


It turns out that the "you couldn't stop us this far, why should you be able to stop us now" theory is more accurate than the "we've worn them out, so now we can score" theory from #5. At the risk of sounding like Aaron, there's an article in the book about that, based on much the same data from Jim Armstrong as this article is.

11 Re: Going Long in Kansas City

Is this a single article or the first of a round of thirty-two?

12 Re: Going Long in Kansas City

Just wanted to clarify something about the numbers above. A drive was considered a long scoring drive only if it gained at least 80 yards. The Chiefs actually had three FG drives starting at or inside their own 20 that didn't go at least 80 yards (60, 68, and 70 yards), but those aren't reflected in the 22 times scored. It turns out that the average FG drive starting at or inside the 20 is only about 67 yards. We debated whether to lump those in with long 80+ yard TD drives for this study and we chose not to.

13 Re: Going Long in Kansas City

As a life-long Chiefs fan, the answer to "which version of the Chiefs do you prefer?" is pretty easy. It was a lot more fun when Derrick Thomas, Neil Smith, etc were wreaking havoc all over the defensive side of the ball. As much as I love Priest Holmes and the Al Saunders offense, bring me back somewhere close to the middle...please! clarification, we did win a playoff game without Montana...1991 over the Raiders in a wild card matchup. After last year, San Diego fans now know the true meaning of pain via "Martyball".

15 Re: Going Long in Kansas City

A couple things jumped out at me in this chart. First, no defense appears to be more field-position dependent than Seattle. Best at preventing long scores, worst at preventing short. Maybe they were especially susceptible after turnovers or big ST plays, the defense didn't get set quickly enough or whatever. Maybe they're the poster children for the "go for the throat after a turnover" theory. Anyone who watched them have any observations or data either way?

Also, the Rams really didn't have a ton of possessions that led to short scoring drives. If I recall, they didn't force a ton of turnovers last year - some of their decline in record was attributed to not recovering hardly any fumbles, I think. Is it safe to assume that this seemingly random luck should reverse/normalize next year, and if so should we assume they score a few more points just from having a short field more often? Or was there something else at work, like they had an unusually high number of 3-and-outs following turnovers or starting inside the 50?

16 Re: Going Long in Kansas City

What causes this success of KC's? Illuminating the factors that contribute to long, effective drives is step one, but how do teams effectively convert on 2nd more often? Is it worth going for the 10++ yard strike, given the increased chances of an interception or simple miss? Do we value high completion percentages or more yards per catch?

17 Re: Going Long in Kansas City

Re #13 - the Chiefs have won playoff games when they had HoF qbs (Montana or Dawson) or when their opponents' had a complete brain lockup and started a rookie qb who played one regular season game (Todd Marinovich, 1991) who threw 4 INTs in the playoff game. That plus 2 other turnovers and home field advantage won the week prior by beating the Raiders in the regular season finale resulted in a 10-6 romp. Marty-ball's finest moment.

18 Re: Going Long in Kansas City

Just a point:

The teams at the bottom of the list are a mixture of poor defenses (Titans, Chiefs, Vikings) and great defenses that allowed very few drives (Steelers, Falcons).

Defenses cannot prevent drives. And they especially can't prevent drives by stopping them quickly. That would correlate to MORE drives, not fewer drives, since the game has constant length and the shorter each drive is, the more there are. I suspect the point is that those teams' offenses ran a lot and therefore reduced the number of drives. That shouldn't be attributed to the defense at all.

19 Re: Going Long in Kansas City


If you define "drive" as "having multiple first downs," then it makes sense.

20 Re: Going Long in Kansas City

Fun fact: The Denver Broncos were the only team that was more likely to score a TD if you gave them the ball inside their 20 than if you gave them the ball across midfield. I guess that's the power of Ashley Lelie.

Maybe it's a good thing the defense doesn't get more turnovers.

Wait a minute, they also give up the most scores when the other team gets the ball on the wrong side of midfield. Let's amend that. The defense should focus on getting turnovers... but only when the other team is in the red zone.

21 Re: Going Long in Kansas City

Yeah, I was looking at the Broncos lines with some surprise too. The most long drives by the offense is no surprise, considering our ST last year, and something of a credit to the offense. But for the defense, every score at that range a TD? Jim's clarification makes that make more sense: if the opponents scored a FG with a drive <80 yards from behind their 20 yard line, it won't show in this line.

The big surprise is on both the lines for short drives. Lowest TD %age for offensive drives beginning from the 50, AND greatest scoring %age (nearly greatest TD %age too) for opponents driving from the 50! This sounds like something that can only be attributed to coaching... or am I making too much of it?

The Buffalo line really points out how important ST is in football.

Great article, thanks guys :-)

22 Re: Going Long in Kansas City

18: I think what I was getting at was that the Steelers and Falcons had very good defenses and ball control offenses that held onto the ball. You are right, of course, that a very successful defense will have lots and lots of three and out "drives", while an awful defense could only allow five drives per game, all of them for 80 yards and a TD.

23 Re: Going Long in Kansas City

What I love about the chart is that it shows how you can’t really tell how good an offense is just by how many points it scores. When you’ve got a defense and special teams as good as the Bills have, you’re going to score on a lot of drives even if your offense isn’t very good.

Actually, checking the numbers under Football Outsiders' drive stats, field position actually has a tiny NEGATIVE correlation (-0.143) with points per drive. In plain English, this means that good offenses tended to overcome bad field position and bad offenses tended to waste good field position. So I'm looking at Buffalo as an outlier and assuming that offenses that score a lot of points are pretty good.

24 Re: Going Long in Kansas City

Taking Vince's keen observation one step further, field position doesn't make much difference for most teams, just the outliers. The difference last season between the best average starting field position on offense (Buffalo) and the worst (St. Louis) was over 10 yards. But half of the teams ranked in the middle were within just two yards.

25 Re: Going Long in Kansas City

Nice observations. I stand corrected.

Anyone have a guess as to why that is? Does it have something to do with the salary cap creating something of a zero-sum game? Are the teams with good offenses the ones that spend a lot of money on their offenses and therefore don't have much to spend and have bad defenses? And vice versa?

And does anyone have thoughts on the best ways to judge the quality of an offense or defense in the days before we had play-by-play data?

26 Re: Going Long in Kansas City


Click my name to find OFF DVOA and DEF DVOA sorted from the best to the worst along with the playing salary (cap value) of each unit.

See more here..

Out of the top ten OFF DVOA values from 2000-2004, 4 paid alot of money relative to the norm, 1 paid a good bit more, 4 paid less than the norm....including the 2004 New England Patriots...OFF DVOA was 26.3, and the OFF playing salary was only $25.231 mil, which was more than 26% LESS than the norm for that year.

27 Re: Going Long in Kansas City

Anyone have a guess as to why that is? Does it have something to do with the salary cap creating something of a zero-sum game? Are the teams with good offenses the ones that spend a lot of money on their offenses and therefore don’t have much to spend and have bad defenses? And vice versa?

What do you mean? Are you asking why the correlation is negative? Well, for all practical purposes, it would be accurate to say there is NO correlation between field position and points per drive, at least over the length of the season. I'd guess that 2004 was a whacky year where the best offenses happened to have the worst punt and kick return teams or something. I only looked at offenses, by the way, not defenses.

I think the biggest factor is that, as Jim A pointed out, the vast majority of teams were very, very close to each other in starting field position, at least over the course of a 16-game season. If you looked at each game individually, I bet you'd see a much higher correlation with winning. But over 16 games, I think the returns and turnovers more or less even out.

28 Re: Going Long in Kansas City

Thanks for the link, J. If I'm reading your stats correctly, it looks like you're saying the 49ers last year had one of the best offensive bargains of recent seasons -- that on a per dollar basis, the 49ers' offense was actually pretty good. Is that correct? What do you think it means?

29 Re: Going Long in Kansas City

I do not have much time now, so this will be quick; for now.

Last year SF's OFF playing salary, cap value, was half that of the norm...only $17m...the average OFF last year was around $34m.

SF offense was terrible -19 DVOA; however, with the little money they did spend on their OFF playing salary, they probably should have been even worse.

Basically, they got more than what they paid for, BUT they paid so little it was still terrible.

Compared to Chicago, their 2004 OFF playing salary was just about average, $34m, but their OFF DVOA was -39. They paid average, but got far less. Injuries and perhaps poor coaching would explain this.

I will try to write more later...

Also, at this time, my "stats" are not stats, just comparisons.

31 Re: Going Long in Kansas City

Well, I don't want to start an argument about what a "stat" is, but I think generally any time you use numbers to organize or interpret information, it's a stat.

Anyway, I much admire what you're doing, J. I think there's a lot of good stuff to be found if you study the payrolls of NFL teams. For that matter, I think a smart analyst could take a look at the payrolls of a lot of organizations and determine whether companies were really getting their money's worth. I know I've worked in a few places where the highest-paid employees and the most productive employees were not the same people.

32 Re: Going Long in Kansas City

Pretty incredible that GB's AVERAGE TD drive last year was 71.6 yards.