The 2009 Cowboys had the most prolific offense in franchise history with 6,390 yards, handily outpacing the 1978 Cowboys (5,916) and ranking second in the NFL, just 71 yards shy of the New Orleans Saints. The Cowboys also ranked No.1 in yards per first down play (6.52) and total yards per play on offense (6.3).
Many were the offensive highlights we saw last year, from Miles Austin's franchise record breaking 250 receiving yards against Kansas City, Felix Jones' league leading 5.9 yards per rush attempt, Tony Romo's career high and franchise record 4,483 passing yards, his league leading 2,154 yards after catch, and so on and so forth.
So how come the Cowboys ended up with only the 14th best scoring offense? Not enough takeaways to give the offense a short field position; a not very effective kick return game, resulting in unfavorable field position for the offense; perhaps our red zone offense is part of the answer, as are the placekicker issues the Cowboys had - take your pick. Another part of the answer could well be the Cowboys third down offense.
The official stats show that last season the Cowboys managed to convert 40.5% of their third downs (82/202), and were ranked a middling 14th in the NFL. That 40.5% conversion rate is the lowest for a Cowboys team since 2005. The one piece of good news in this stat is that the 202 third downs were actually the sixth fewest in the NFL (the NFL average was 215) - looks like the Cowboys offense was pretty good at avoiding third down situations all together.
To better understand what was or wasn't happening on third downs, we'll need to look at some stat breakdowns.
The stats in this post are broken down by 'distance to go' on third down, and follow the NFL definitions for short and medium situations. I've broken up third and long situations into 8-10 yards and 11+ yards as they will be relevant for the analysis.
Distance required
As you look at distance to go on third down, you would naturally assume that the longer the distance, the less likely a first down conversion becomes. And you would be right in that assumption.
Advancednflstats.com have run an excellent analysis detailing first down probability, which they've summarized in the graph below. Notice the almost linear decrease in conversion probability, both in second and third down situations.
courtesy of advancednflstats.com
The Cowboys have faced roughly the same number of down situations for all distance to go situations, from short, medium, long and very long third down situations. And it's no surprise that the conversion rate drops as the distance to go increases.
What does come as a surprise is that our conversion rate on third and medium is well below the norm. According to the graph, we should expect a conversion rate in the mid forties, but the Cowboys come in at 32%. The Cowboys have a better conversion rate on third and long than they do on third and medium!
Cowboys third down conversions by distance required, 2009
| Short (1-3) | Medium (4-7) | Long (8-10) | Very Long* (11+) | |
| Attempts |
54 | 53 | 46 | 46 |
| Conversions |
36 | 17 | 20 | 9 |
| in % |
67% | 32% | 44% | 20% |
We'll look at these numbers in more detail below as we break down the passing and rushing game, but two notes* on the 46 very long situations: I've excluded three knees Romo took on third and very long. Also, 21 of those 46 third and very long situations were the result of one or more penalties on the previous plays and 14 resulted from sacks on previous plays.
Passing
Romo's completion rate on third downs is 56%. Not good by any stretch of the imagination, but if you were to take out Roy Williams' numbers, it would improve to 63%. Now before I get blasted again for fiddling with the numbers, take it as food for thought only, perhaps there is a lesson in there somewhere ...
Pass completion percentages by down and distance, 2009
| Short (1-3) |
Medium (4-7) |
Long (8-10) |
Very Long (11+) |
|
| Passes |
25 | 43 | 39 | 37 |
| Pass compl. % | 72% | 40% | 64% | 57% |
| 1st dn% |
64% | 33% | 51% | 24% |
| Sacks |
1 | 2 |
4 |
3 |
Of the 43 pass plays in third and medium situations the Cowboys converted only 14. It should come as no surprise that only Austin converted for a first down with any regularity, converting seven of nine passes thrown his way for a first down. Even Jason Witten (4-11) did not shine at this distance. Roy Williams (2-10) and Patrick Crayton (1-6) at least caught a pass, Tashard Choice failed twice (0-2), all others were 0-1 (Marion Barber, Martellus Bennett, Sam Hurd, John Phillips). The question here is, what are the one or two pass plays on medium yardage that the Cowboys could and should use with a fairly high success rate?
Nowhere has Barber's fractured thumb and torn quad affected us more than on these medium distance passes. In 2008, Barber caught 52 of 62 passes overall for a league leading 85.2% reception rate. This year, he's only 26 for 36, and that's critical production quantity and quality the Cowboys sorely missed.
No other quarterback in the NFL has had to pass on more third and long/very long situations ithan Tony Romo (76 pass attempts on third an 8+). That is one big, ugly stat to hold the lead in. Remarkably, Stats LLC shows Romo with the fourth best completion percentage at that distance in the league at 38.2. It would stand to reason that the longer the distance required on third down, the lower the conversion rate would be, but Romo actually completes more passes in third and long/very long situations than in third and medium situations.
Target Distribution
Every week, NFL quarterbacks line up on third down knowing that if they can connect with their clutch receiver - that go-to guy who rarely lets them down - they can keep a drive alive rather than head back to the sidelines in frustration. So who's that guy for Tony Romo?
Passing target distribution, third down:
| Name | Targets | Cmp | Cmp % | 1st down | 1st dn % |
|---|---|---|---|---|---|
| Witten | 34 | 24 | 71% | 15 | 44% |
| Austin | 32 | 21 | 66% | 17 | 53% |
| Crayton | 24 | 15 | 63% | 13 | 54% |
| Williams | 26 | 7 | 27% | 6 | 23% |
| Choice | 10 | 7 | 70% | 3 | 33% |
| Ogletree | 4 | 4 | 100% | 3 | 75% |
| Bennett | 5 | 1 | 20% | 1 | 20% |
| Hurd | 3 | 1 | 33% | 1 | 33% |
| Jones | 1 | 1 | 100% | 0 | 0% |
| Barber | 3 | 0 | 0% | 0 | 0% |
| Phillips | 1 | 0 | 0% | 0 |
0% |
| Total | 144 | 81 | 56% | 59 | 41% |
Stat geek alert: Click on the column headers to sort the table to your liking
While Witten continued to impress with a high reception rate, this has not translated into above average third down conversions. Miles Austin and, much to my surprise, Patrick Crayton, are now heavily involved in the battle for 'Romo's security blanket' - a battle that is sure to extend well into training camp and will most likely also feature Kevin Ogletree.
Williams on the other hand will have to do a lot of, excuse the pun, 'catching up' to stay in that battle. 19 of 26 passes on third down not caught simply doesn't cut it.
Rushing
Running backs first down conversions by down and distance, 2009
| Short (1-3) |
Medium (4-7) |
Long (8-10) |
Very Long (11+) |
|
| Attempts |
26 | 7 | 3 | 2 |
| 1st dn% | 69% | 29% | 0% | 0% |
The numbers above are the rushing stats for the running backs only. Romo's numbers: 2/2 on third and short, 1/1 on third and medium, 0/4 on some desperate third and very long scrambles (18, 18, 20 and 24 yards to go).
On third and medium, the running backs ran the ball seven times and converted two times.
Rushing target distribution, third down
| Name | Att | 1st down | 1st dn % |
|---|---|---|---|
| Barber | 21 | 12 | 57% |
| Choice | 12 | 6 | 50% |
| Romo | 7 | 3 | 43% |
| Jones | 5 | 2 | 40% |
| Total | 45 | 23 | 51% |
Stats LLC rank Barber's 57% conversion rate on third down as the 16th best in the league. Noteworthy for a rushing ranking is that four quarterbacks make it into the top six (Rodgers, Sanchez, Garrard, Campbell) with high third down rushing conversion percentages.
The rushing stats also give an indication of the roles each of the three running backs on the Cowboys squad has: Only 4% of Felix Jones' rushes are on third down (5/116), Marion Barber: 10% (21/214) and Tashard Choice: 19% (12/64).
Conversely, 62% of Jones' rushes are on first downs (72/116), and his 6.0 YPC on first is ranked third in the NFL behind Kansas City's Jamaal Charles and Baltimore's Willis McGahee, both with 6.4. Barber rushed on first downs on 57% of his carries (122/214), Choice on 56% (36/64).