24 Jul 2008
by Aaron Schatz
In honor of our fifth anniversary, we're running a series of articles looking at the best and worst players in the history of our advanced stats, DVOA and DYAR. If you are unfamiliar with our advanced stats -- perhaps you are a new reader visiting our website for the first time after picking up a copy of Pro Football Prospectus 2008 -- you can read all about them here. The series so far:
First up today, the top 20 total DYAR games by running backs since 1995:
| Player | Team | Year | Week | Opp | Total DYAR |
Rush DYAR |
Rec DYAR |
Runs | RuYd | RuTD | Pass | Rec | RecYd | RecTD |
| Priest Holmes | KC | 2002 | 12 | SEA | 157 | 104 | 53 | 23 | 197 | 2 | 7 | 7 | 110 | 1 |
| Corey Dillon | CIN | 1997 | 15 | TEN | 133 | 122 | 11 | 39 | 246 | 4 | 2 | 2 | 30 | 0 |
| Marshall Faulk | IND | 1998 | 13 | BAL | 126 | 85 | 41 | 17 | 192 | 1 | 8 | 7 | 75 | 1 |
| Joseph Addai | IND | 2006 | 12 | PHI | 126 | 121 | 5 | 24 | 171 | 4 | 3 | 2 | 37 | 0 |
| Brian Westbrook | PHI | 2007 | 3 | DET | 121 | 60 | 62 | 14 | 110 | 2 | 5 | 5 | 111 | 1 |
| Marshall Faulk | STL | 2000 | 17 | NO | 117 | 92 | 24 | 32 | 220 | 2 | 8 | 7 | 41 | 1 |
| Marshall Faulk | STL | 2000 | 7 | ATL | 116 | 80 | 36 | 25 | 208 | 1 | 10 | 7 | 78 | 0 |
| Edgerrin James | IND | 2004 | 11 | CHI | 110 | 103 | 7 | 23 | 204 | 1 | 1 | 1 | 11 | 0 |
| Brian Westbrook | PHI | 2007 | 10 | WAS | 109 | 47 | 62 | 20 | 99 | 1 | 6 | 5 | 83 | 2 |
| Marshall Faulk | STL | 1999 | 16 | CHI | 108 | 6 | 101 | 10 | 54 | 0 | 13 | 12 | 204 | 1 |
| Player | Team | Year | Week | Opp | Total DYAR |
Rush DYAR |
Rec DYAR |
Runs | RuYd | RuTD | Pass | Rec | RecYd | RecTD |
| Fred Taylor | JAC | 2000 | 12 | PIT | 105 | 109 | -4 | 30 | 234 | 3 | 7 | 3 | 14 | 1 |
| Larry Johnson | KC | 2004 | 13 | OAK | 105 | 67 | 39 | 20 | 118 | 1 | 4 | 3 | 56 | 1 |
| Clinton Portis | DEN | 2003 | 14 | KC | 104 | 88 | 16 | 22 | 218 | 5 | 3 | 2 | 36 | 0 |
| Pierre Thomas | NO | 2007 | 17 | CHI | 103 | 45 | 57 | 20 | 105 | 0 | 15 | 12 | 121 | 1 |
| Marshall Faulk | STL | 2002 | 7 | SEA | 102 | 84 | 17 | 32 | 183 | 3 | 11 | 7 | 52 | 1 |
| Marshall Faulk | STL | 2000 | 5 | SD | 99 | 22 | 78 | 7 | 55 | 0 | 7 | 6 | 116 | 2 |
| Marshall Faulk | STL | 2000 | 3 | SF | 99 | 80 | 19 | 25 | 133 | 3 | 12 | 7 | 65 | 0 |
| Dorsey Levens | GB | 1999 | 17 | ARI | 99 | 86 | 13 | 24 | 146 | 4 | 4 | 3 | 23 | 0 |
| Priest Holmes | KC | 2002 | 5 | NYJ | 99 | 54 | 44 | 23 | 152 | 1 | 10 | 9 | 81 | 1 |
| LaDainian Tomlinson | SD | 2005 | 3 | NYG | 99 | 116 | -18 | 21 | 192 | 3 | 7 | 6 | 28 | 0 |
There has never been a rushing and receiving combination quite like Marshall Faulk, and he dominates the list of the top running back games, with seven of the top 20. For good measure, he also has the games ranked 24th and 28th. However, Faulk does not have the top game; that belongs to Priest Holmes in a wild 39-32 shootout loss to Seattle back in Week 12 of 2002. (On the other side, Shaun Alexander had a pretty good game too -- 145 yards on 23 carries with two touchdowns and 49 DYAR.)
Just behind Holmes is Corey Dillon, who ran for 246 yards against a Titans defense that ranked third in DVOA against the run back in 1997. The Titans only gave one other 100-yard rushing game all season -- oddly enough, to backup Seattle running back Steve Broussard, who had 138 rushing yards on just six carries in Week 6 (with touchdown runs of 43 and 77 yards). In nine out of 16 games that year, the Titans kept their opponent's leading rusher to 51 or fewer yards.
The always underrated Brian Westbrook shows up twice in the top ten with games from last year; that appearance is no surprise, but Pierre Thomas's appearance certainly is. His Week 17 game against Chicago was a testament to efficiency. Ten of his 20 carries gained six or more yards. Eleven of his 12 receptions gained eight or more yards. Between rushing and receiving, Thomas had 15 first downs and a touchdown, and he would rank even higher if he had not been stuffed twice at the goal line.
So if Thomas is here, where the heck is Adrian Peterson? Peterson's NFL-record rushing performance from Week 9 is worth 98 total DYAR (91.3 rushing, 6.3 receiving), making it the 22nd best game since 1995. As I've noted in the past, Peterson suffers a bit because of a fumble and three carries that lost yardage. Peterson's game still does a lot better in our stats than the game that held the record before 2007, Jamal Lewis's 295-yard day against Cleveland in Week 2 of 2003. That game doesn't even small the top 50, worth only 55 DYAR (59 rushing, -4 receiving). Playing against the run defense ranked 25th in DVOA, Lewis got 193 of his yards on just three runs, and he had ten runs for a yard or less (including a four-yard loss on first-and-goal from the Cleveland two-yard line).
A surprising name leads the list of the worst total DYAR games since 1995:
| Player | Team | Year | Week | Opp | Total DYAR |
Rush DYAR |
Rec DYAR |
Runs | RuYd | RuTD | Pass | Rec | RecYd | RecTD |
| Curtis Martin | NYJ | 1998 | 9 | KC | -95 | -94 | -1 | 30 | 42 | 0 | 3 | 2 | 14 | 0 |
| Ricky Watters | SEA | 1998 | 5 | KC | -95 | -80 | -15 | 22 | 59 | 0 | 2 | 2 | -6 | 0 |
| Ricky Williams | NO | 2001 | 17 | SF | -90 | -58 | -33 | 11 | 33 | 0 | 5 | 4 | -8 | 0 |
| Ricky Williams | NO | 1999 | 17 | CAR | -86 | -46 | -40 | 14 | 7 | 0 | 7 | 5 | 5 | 0 |
| Mike Alstott | TB | 1999 | 10 | KC | -80 | -81 | 1 | 16 | 33 | 0 | 1 | 1 | 7 | 0 |
| Duce Staley | PHI | 2000 | 4 | NO | -72 | -33 | -39 | 18 | 44 | 0 | 5 | 3 | 16 | 0 |
| Harvey Williams | OAK | 1995 | 7 | DEN | -71 | -48 | -22 | 6 | 8 | 0 | 6 | 4 | 9 | 0 |
| Joe Aska | OAK | 1996 | 7 | DET | -71 | -43 | -28 | 16 | 44 | 0 | 4 | 3 | 10 | 0 |
| Ray Zellars | NO | 1996 | 12 | ATL | -71 | -61 | -10 | 10 | -4 | 0 | 2 | 0 | 0 | 0 |
| Travis Henry | BUF | 2002 | 17 | CIN | -69 | -64 | -5 | 30 | 81 | 0 | 2 | 2 | 0 | 0 |
Going through these games, you can just feel the frustration of running backs running into a brick wall over and over. The highlight of Curtis Martin's day was getting drilled in the backfield for an eight-yard loss and a fumble on second-and-13. That was one of two fumbles and one of six carries that lost yardage, to go with five carries of zero yards, five carries of one yard, and six carries of two yards. Against the same Kansas City defense four weeks earlier, Ricky Watters fumbled three times, with five carries that lost yardage and a minus-seven-yard reception. And boy, did Ricky Williams know how to finish up a season, or what? In the last week of 1999, the Panthers stopped Williams for lost yardage on half his carries, a seven-yard carry was his only run over four yards, and he had no first downs. In the last week of 2001, he had catches for minus-six and minus-seven yards, proving that it may be a good idea to drop the ball whenever Aaron Brooks attempts to dump off to you.
And yes, that is the same Joe Aska who later resurfaced in the XFL.
Now, looking at this list of worst games, you probably are wondering why there aren't any games from the last five years. This exposes a general problem that I will need to fix with the individual stats. As I've noted in the past, he baselines are derived from the past five years, but the offensive environment of the league changes over time. This is particularly true when it comes to the running game; the average running back simply wasn't as effective between 1996 and 2001 compared to between 2002 and 2007. At some point, it makes sense to create "era-adjusted" stats that will translate all players into today's environment, but I want to try to do it in such a way that the translation of "success points" to yardage is still fairly accurate. Overall, it's a future project, and for now we'll run with what we have. In the meantime, here are the five worst running back games since 2003:
| Player | Team | Year | Week | Opp | Total DYAR |
Rush DYAR |
Rec DYAR |
Runs | RuYd | RuTD | Pass | Rec | RecYd | RecTD |
| Edgerrin James | ARI | 2006 | 6 | CHI | -64 | -66 | 2 | 36 | 57 | 0 | 2 | 1 | 7 | 0 |
| Clinton Portis | WAS | 2004 | 15 | SF | -64 | -63 | -1 | 35 | 110 | 0 | 4 | 3 | 20 | 0 |
| Tony Hollings | HOU | 2003 | 14 | JAC | -64 | -28 | -36 | 18 | 19 | 0 | 5 | 1 | 6 | 0 |
| Steven Jackson | STL | 2007 | 1 | CAR | -61 | -28 | -33 | 18 | 59 | 0 | 7 | 1 | 3 | 0 |
| Julius Jones | DAL | 2005 | 17 | STL | -59 | -37 | -22 | 15 | 37 | 0 | 2 | 2 | -11 | 0 |
Yes, that's correct... the worst running back game since 2003 was the same night as the worst quarterback game since 1995. In case you were wondering how the Bears could win despite Rex Grossman having the worst quarterback game of the DVOA Era, well, now you know. I wrote a special Every Play Counts that year about just how awful the Arizona run blocking was. James is actually worth -79 YAR before opponent adjustments, but of course the 2006 Bears had an excellent defense. Meanwhile, I said it at the start of last season and I'll say it again -- I have no idea how a running back possibly manages to catch only one of seven intended passes. The nice thing about a game from 2007 is that we have game charting. Only one of Jackson's seven passes that day was actually thrown behind the line of scrimmage, and none of them were dropped -- we have them listed as Thrown Behind, Overthrown, Thrown Ahead, Defensed (by Dan Morgan), Bulger hit in motion by Mike Rucker blowing through Adam Goldberg, and Defensed (by Thomas Davis). Of course, just because the passes aren't dropped doesn't mean Jackson doesn't share responsibility for some of the incompletes.
Now let's look at the best and worst games if we isolate just rushing DYAR or receiving DYAR. First, the best rushing days, extended to 12 games to get in Adrian Peterson's record-setting day from last year:
| Player | Team | Year | Week | Opp | Total DYAR |
Rush DYAR |
Rec DYAR |
Runs | RuYd | RuTD | Pass | Rec | RecYd | RecTD |
| Joseph Addai | IND | 2006 | 12 | PHI | 126 | 121 | 5 | 24 | 171 | 4 | 3 | 2 | 37 | 0 |
| Corey Dillon | CIN | 1997 | 15 | TEN | 133 | 122 | 11 | 39 | 246 | 4 | 2 | 2 | 30 | 0 |
| LaDainian Tomlinson | SD | 2005 | 3 | NYG | 99 | 116 | -18 | 21 | 192 | 3 | 7 | 6 | 28 | 0 |
| Fred Taylor | JAC | 2000 | 12 | PIT | 105 | 109 | -4 | 30 | 234 | 3 | 7 | 3 | 14 | 1 |
| Priest Holmes | KC | 2002 | 12 | SEA | 157 | 104 | 53 | 23 | 197 | 2 | 7 | 7 | 110 | 1 |
| Edgerrin James | IND | 2004 | 11 | CHI | 110 | 103 | 7 | 23 | 204 | 1 | 1 | 1 | 11 | 0 |
| Tiki Barber | NYG | 2006 | 17 | WAS | 93 | 96 | -3 | 23 | 234 | 3 | 4 | 2 | 24 | 0 |
| Marshall Faulk | STL | 2000 | 17 | NO | 117 | 92 | 24 | 32 | 220 | 2 | 8 | 7 | 41 | 1 |
| Priest Holmes | KC | 2001 | 8 | SD | 93 | 92 | 1 | 30 | 181 | 1 | 1 | 1 | 9 | 0 |
| Mike Anderson | DEN | 2000 | 14 | NO | 98 | 92 | 6 | 37 | 251 | 4 | 1 | 1 | 5 | 0 |
| Adrian Peterson | MIN | 2007 | 9 | SD | 98 | 91 | 6 | 30 | 296 | 3 | 2 | 1 | 19 | 0 |
| Shaun Alexander | SEA | 2001 | 9 | OAK | 93 | 90 | 3 | 35 | 266 | 3 | 1 | 1 | 7 | 0 |
One game here that is particularly interesting is the day Priest Holmes went nuts on the 2001 Chargers. That gets a huge opponent adjustment because San Diego ranked second in run defense DVOA. Holmes's total of 181 yards was 50 percent higher than the total rushing yardage for the entire team combined in the 15 other games against the Chargers defense.
The problem of lower rushing productivity before the turn of the century rears its head on this table too, so I ran another table with the top rushing games from 1995 through 1999:
| Player | Team | Year | Week | Opp | Total DYAR |
Rush DYAR |
Rec DYAR |
Runs | RuYd | RuTD | Pass | Rec | RecYd | RecTD |
| Corey Dillon | CIN | 1997 | 15 | TEN | 133 | 122 | 11 | 39 | 246 | 4 | 2 | 2 | 30 | 0 |
| Terrell Davis | DEN | 1997 | 14 | SD | 90 | 89 | 2 | 26 | 178 | 1 | 5 | 4 | 36 | 0 |
| Dorsey Levens | GB | 1999 | 17 | ARI | 99 | 86 | 13 | 24 | 146 | 4 | 4 | 3 | 23 | 0 |
| Marshall Faulk | IND | 1998 | 13 | BAL | 126 | 85 | 41 | 17 | 192 | 1 | 8 | 7 | 75 | 1 |
| Garrison Hearst | SF | 1998 | 15 | DET | 74 | 85 | -11 | 24 | 198 | 1 | 2 | 0 | 0 | 0 |
| Emmitt Smith | DAL | 1995 | 1 | NYG | 75 | 80 | -5 | 21 | 163 | 4 | 1 | 1 | 0 | 0 |
| James Stewart | JAC | 1997 | 7 | PHI | 71 | 80 | -9 | 15 | 102 | 5 | 3 | 2 | 7 | 0 |
| LeShon Johnson | ARI | 1996 | 4 | NO | 79 | 79 | 0 | 21 | 214 | 2 | 0 | 0 | 0 | 0 |
Can you imagine the roto stampede when backup running back LaShon Johnson, with 101 career rushing yards up to that point, came off the bench to run for 214 yards against New Orleans in early 1996? Johnson had 97 yards on 12 carries the following week... then managed just 319 yards the rest of the season combined.
The list of the least valuable rushing games has a better mix of the 13 years of DVOA:
| Player | Team | Year | Week | Opp | Total DYAR |
Rush DYAR |
Rec DYAR |
Runs | RuYd | RuTD | Pass | Rec | RecYd | RecTD |
| Curtis Martin | NYJ | 1998 | 9 | KC | -95 | -94 | -1 | 30 | 42 | 0 | 3 | 2 | 14 | 0 |
| Mike Alstott | TB | 1999 | 10 | KC | -80 | -81 | 1 | 16 | 33 | 0 | 1 | 1 | 7 | 0 |
| Ricky Watters | SEA | 1998 | 5 | KC | -95 | -80 | -15 | 22 | 59 | 0 | 2 | 2 | -6 | 0 |
| Jerald Moore | STL | 1998 | 1 | NO | -67 | -68 | 2 | 15 | 31 | 0 | 1 | 1 | 6 | 0 |
| Edgerrin James | ARI | 2006 | 6 | CHI | -64 | -66 | 2 | 36 | 57 | 0 | 2 | 1 | 7 | 0 |
| Travis Henry | BUF | 2002 | 17 | CIN | -69 | -64 | -5 | 30 | 81 | 0 | 2 | 2 | 0 | 0 |
| LaMont Jordan | OAK | 2005 | 11 | WAS | -50 | -63 | 14 | 27 | 52 | 0 | 7 | 5 | 37 | 0 |
| Garrison Hearst | ARI | 1995 | 16 | PHI | -51 | -63 | 12 | 26 | 95 | 0 | 1 | 1 | 18 | 0 |
| Clinton Portis | WAS | 2004 | 15 | SF | -64 | -63 | -1 | 35 | 110 | 0 | 4 | 3 | 20 | 0 |
| Ricky Williams | NO | 1999 | 5 | ATL | -68 | -63 | -5 | 19 | 53 | 0 | 4 | 3 | 16 | 0 |
Clinton Portis's game against San Francisco is the worst 100-yard rushing game of the DVOA Era. 35 carries for 110 yards looks good, right? Then you realize it is barely three yards per carry against the defense that ranked 31st in DVOA against the run. Portis was stopped for zero or lost yardage ten times.
Weird that the bottom three games all came against the same Kansas City defense, which ranked fifth against the run in 1998 and 15th in 1999.
Here are the best receiving games by running backs since 1995, extended a bit to get a few games in there from the early years.
| Player | Team | Year | Week | Opp | Total DYAR |
Rush DYAR |
Rec DYAR |
Runs | RuYd | RuTD | Pass | Rec | RecYd | RecTD |
| Marshall Faulk | STL | 1999 | 16 | CHI | 108 | 6 | 101 | 10 | 54 | 0 | 13 | 12 | 204 | 1 |
| Brian Westbrook | PHI | 2004 | 13 | GB | 79 | -15 | 93 | 12 | 37 | 0 | 11 | 11 | 156 | 3 |
| Marshall Faulk | STL | 2001 | 12 | ATL | 88 | 9 | 79 | 12 | 70 | 0 | 7 | 6 | 128 | 3 |
| Moe Williams | MIN | 2003 | 10 | SD | 86 | 8 | 79 | 12 | 45 | 0 | 12 | 11 | 113 | 2 |
| Marshall Faulk | STL | 2000 | 5 | SD | 99 | 22 | 78 | 7 | 55 | 0 | 7 | 6 | 116 | 2 |
| Brian Westbrook | PHI | 2003 | 11 | NYG | 92 | 19 | 73 | 9 | 47 | 1 | 6 | 5 | 60 | 2 |
| LaDainian Tomlinson | SD | 2003 | 14 | DET | 85 | 15 | 70 | 25 | 88 | 0 | 10 | 9 | 148 | 2 |
| Larry Johnson | KC | 2004 | 17 | SD | 68 | 4 | 65 | 17 | 46 | 1 | 10 | 8 | 115 | 1 |
| Richard Huntley | PIT | 1999 | 1 | CLE | 78 | 14 | 64 | 10 | 41 | 1 | 6 | 5 | 67 | 2 |
| Charlie Garner | OAK | 2002 | 16 | DEN | 88 | 26 | 62 | 7 | 61 | 0 | 9 | 8 | 91 | 1 |
| Brian Westbrook | PHI | 2007 | 3 | DET | 121 | 60 | 62 | 14 | 110 | 2 | 5 | 5 | 111 | 1 |
| Brian Westbrook | PHI | 2007 | 10 | WAS | 109 | 47 | 62 | 20 | 99 | 1 | 6 | 5 | 83 | 2 |
| Larry Centers | ARI | 1995 | 13 | ATL | 80 | 18 | 61 | 13 | 62 | 0 | 8 | 7 | 101 | 0 |
| Raymont Harris | CHI | 1996 | 1 | DAL | 50 | -12 | 61 | 10 | 21 | 0 | 4 | 3 | 103 | 1 |
| Barry Sanders | DET | 1995 | 14 | CHI | 53 | -8 | 61 | 23 | 90 | 0 | 8 | 6 | 93 | 1 |
That Marshall Faulk game against Chicago in 1999 is just ridiculous, with six different receptions for 15 or more yards. Moe Williams gets a high rating for his 2003 game against San Diego because he had four different receptions that converted on third down, including two third-and-long situations.
Finally, the worst receiving days by running backs:
| Player | Team | Year | Week | Opp | Total DYAR |
Rush DYAR |
Rec DYAR |
Runs | RuYd | RuTD | Pass | Rec | RecYd | RecTD |
| Edgerrin James | IND | 2002 | 1 | JAC | -43 | 6 | -49 | 26 | 99 | 0 | 9 | 2 | 10 | 0 |
| Ricky Williams | NO | 1999 | 17 | CAR | -86 | -46 | -40 | 14 | 7 | 0 | 7 | 5 | 5 | 0 |
| Duce Staley | PHI | 2000 | 4 | NO | -72 | -33 | -39 | 18 | 44 | 0 | 5 | 3 | 16 | 0 |
| Charles Way | NYG | 1998 | 3 | DAL | -48 | -9 | -39 | 9 | 26 | 0 | 7 | 5 | 0 | 0 |
| Ricky A. Williams | IND | 2003 | 6 | CAR | -57 | -21 | -37 | 17 | 57 | 0 | 5 | 2 | 7 | 0 |
| Tony Hollings | HOU | 2003 | 14 | JAC | -64 | -28 | -36 | 18 | 19 | 0 | 5 | 1 | 6 | 0 |
| Adrian Murrell | NYJ | 1995 | 17 | NO | -38 | -2 | -35 | 10 | 43 | 0 | 10 | 9 | 12 | 0 |
| Terrell Davis | DEN | 1995 | 7 | OAK | -65 | -30 | -35 | 18 | 34 | 0 | 11 | 5 | 25 | 0 |
| James Stewart | DET | 2002 | 4 | NO | -49 | -15 | -34 | 18 | 65 | 0 | 5 | 3 | -5 | 0 |
| Curtis Martin | NYJ | 1999 | 9 | ARI | -58 | -24 | -34 | 38 | 131 | 0 | 5 | 4 | 2 | 0 |
| Willis McGahee | BUF | 2004 | 15 | CIN | -53 | -19 | -33 | 11 | 25 | 0 | 3 | 2 | -1 | 0 |
| Steven Jackson | STL | 2007 | 1 | CAR | -61 | -28 | -33 | 18 | 59 | 0 | 7 | 1 | 3 | 0 |
| Ricky Williams | NO | 2001 | 17 | SF | -90 | -58 | -33 | 11 | 33 | 0 | 5 | 4 | -8 | 0 |
| Edgar Bennett | GB | 1995 | 4 | JAC | -50 | -17 | -32 | 21 | 66 | 0 | 4 | 3 | 16 | 0 |
| Michael Bennett | KC | 2007 | 2 | CHI | -38 | -6 | -32 | 7 | 14 | 0 | 4 | 3 | 24 | 0 |
It was fun going through the game charting to figure out how Steven Jackson managed to catch just one of seven passes against Carolina last year, but I really wish we had 2002 game charting so I could figure out just how Edgerrin James managed to catch just two of nine passes from Peyton Manning (?!?!) including three incompletes in the red zone. Oh, and James fumbled away one of the two balls he did manage to catch.
A lot of games end up with negative receiving DYAR because of fumbles -- for example, Michael Bennett's double-fumble day against Chicago last year. When you don't have a lot of opportunities, fumbling once or twice will have a big impact on your rating for the day. Adrian Murrell, however, manages to make this list despite no fumbles and -- even more astounding -- only one incomplete pass. How on earth do you catch nine of ten passes and end up with -35 DYAR? Murrell had three receptions that lost yardage. They all lost at least four yards, and two of them came on fourth down. The one incomplete came on third-and-3. Murrell also had a zero-yard reception on second-and-10 and a three-yard reception on second-and-17.
Next week: The best running back seasons and career totals, and we add all running back DYAR from 1995-2007 to the website.
"The always underrated Brian Westbrook shows up twice in the top ten with games from last year; the Week 10 game against Washington is the game where he sat down on the one-yard line to ice the clock instead of scoring his fourth touchdown of the day."
No, he did that in Dallas in Week 15 (link in name). Nitpicking aside, great article!
Whoops! My mistake. Fixed.
Very interesting article.
Amazing that not only did someone have over 250 yard total offense and not score a TD - it happened 3 times? Wow! Marshall Faulk at least has the Rams offense to blame, what was the deal with Corey Dillon? Oh, and btw thanks for calling the 90's "before the turn of the century". I'd forgotten that I'm freakin' old.
I love the guy, but maybe Edge figures no publicity is bad publicity, so long as they spell his name right. His consistency and success rate were both so high, but there was some boom and bust there: I'll remember him for a long time, the 200+ yards I saw live against Seattle along with the 13 yards on 9 carries against the Jets in that 41-0 playoff loss. The OL must have been a seive in the game he caught 2 of 7 from Manning.
oh, and phenomenal article, a few typos aside.
That Fred Taylor game against Pittsburgh was (at the time - not sure if it still is) the single-game rushing record against them, and he did it in Pittsburgh. That was an impressive game to watch and it's good to see the advanced stats back that up.
Looking at this article prompted me to look at the season stats for the Saints' RBs, which prompted me to ask the question: Is Pierre Thomas destined to be the next Marion Barber/Jerious Norwood, languishing on the bench behind clearly inferior players?
I am utterly amazed that the post-Curtis Martin, pre-Corey Dillon Patriots don't show up on any of the "worst rushing game" lists. Am I mis-remembering how abysmal their rushing attack was for about five years?
The Ultra Back!
Re 9:
This is single player performance, not team. I presume the Patriots aren't up there because no single player sucked enough on their own.
It's interesting how the QB performances were (seemingly) much more intuitive than the RB performances. In other words, the games that QBs were ranked highly on seemed to also be intuitively good-- at least compared to RB games.
Does this say something about the way that FO rates running backs, or the way the public rates them? Or something in between?
12: FO stats are rating RB production, not the RBs themselves. I have long suspected that rushing DVOA is dramatically affected by the quality of the run blocking and the offensive system - much more so than yards per carry.
Re: 12
I think some of it also has to do with the fact that we hear more often "QBEagles went 25-40 for 300 yards", but we don't hear "Bo Jackson ran for 100 yards on 43 carries"--we just get the raw yardage, so that sounds quite impressive.
FO stats are rating RB production, not the RBs themselves
A semantic, surely... most people do rate RBs on their production, as opposed to their.. um... attractiveness or some such.
14. You may be on to something here. After all, it's hard for a QB to rack up a bunch of yards without completing a lot of passes... and MOST of the time (or at least more so than a 4 yard run), a completed pass will be a success, in the FO- version of success. But we do see RBs who rack up a ton of yards, and can even average a ton of yards per carry without consistently providing value.
I think the thing is that the public often is very impressed by, for example, 25 carries for 100 yards, because they place too much emphasis on things like ball control/killing the clock/wearing down the defense/etc.
On the other hand, I think FO underrates some backs by undervaluing long runs. When a back has a reputation as a "homerun hitter", it often causes defenses to devote extra attention to him, since they don't want to get burned for a long one. That, in turn, opens up things for the passing game, especially the play action. Of course, this effect doesn't show up in the RB's stats.
1. Great article, as always.
2. Brian Westbrook ROCKS. I really hope he and the Eagles can come to an agreement on his contract. I see him as a Tiki Barber-type of back, only getting better in the next couple of years. There really isn't anything he can't do.
how long can we call Brian Westbrook underrated until he stops being underrated? People are finally starting to look at him in the same light as guys like LT. He'll never have the career numbers...but there isn't a single thing you can ask of a RB that Westbrook can't do.
A semantic, surely… most people do rate RBs on their production, as opposed to their.. um… attractiveness or some such.
Production is dramatically affected by situational stuff. FO stats give Kenton Keith a higher DVOA than Frank Gore. That is a pretty poor measure for RB ability. (It's true that running plays involving Kenton Keith were, on average, more effective than running plays involving Frank Gore. It just doesn't mean Keith is a better running back.)
Yards per carry is actually probably a better proxy for RB ability. As mentioned, Kenton Keith has a high DVOA, comparable to Adrian Peterson's, despite a much lower yards per carry. That's because people play safeties deep against Manning, and he gets his 4.5 yards, and then someone tackles him, and that's that. Against Peterson, people stuff the box fearlessly, sometimes literally playing a 5-3 defense, and sometimes he never has a chance. Nonetheless, he hits enough homeruns to have a much higher YPC than Keith.
I think this sort of thing happens a lot. Given two backs on different teams with equally high DVOA, where one averages 4.5 yards per carry and the other averages 5.5 yards per carry, I'd guess that the first is a decent back on a great team, but the second is a great back on a weaker team.
I know there's an issue with quarterbacks' rushing DYAR, but what about RBs' passing? It's probably impossible to figure in, but in that 20th best DYAR game, Tomlinson threw for a TD too.
@18: I've been stealing Westbrook late first round/early second every year for the last few years, because my friends trust those "fantasy draft rankings" that never seemed to have him in the top five. We don't even give points for receptions, either. Unfortunately it seems those rankings finally realized he always outperforms them and now he's top 3 everywhere :(
I think some of it also has to do with the fact that we hear more often "QBEagles went 25-40 for 300 yards", but we don’t hear "Bo Jackson ran for 100 yards on 43 carries"–we just get the raw yardage, so that sounds quite impressive.
Actually, you'll hear total carries, fairly often, but you never hear about number of SUCCESSFUL carries. That's the real key.
How great was Westbrook against Detroit to finish that high in DYAR despite facing a defense that finished 30th in the league?
Is Pierre Thomas destined to be the next Marion Barber/Jerious Norwood, languishing on the bench behind clearly inferior players?
I don't know about Marion Barber, but I'd say Jerious Norwood is getting all the carries he can handle. Actually, maybe a few more than he can handle, given that he has yet to go through an entire season uninjured. If he can only carry the ball about 100 times without losing significant time to injury, then he can still be a very valuable player (and he is), but he shouldn't be given a full workload of carries. Remember, DVOA doesn't account for durability at all. This isn't a problem for most positions, but with RB, it can become an issue. Just because his DVOA is much higher than his teammates' when he gets 100 carries doesn't mean he should get more carries than he currently does. Remember, if he's on the bench injured, his DVOA doesn't change at all, but his value to his team plummets.
How great was Westbrook against Detroit to finish that high in DYAR despite facing a defense that finished 30th in the league?
How about this for crazy:
Receiving:
Westbrook against Detroit - 111 yards, 1 TD.
Terrell Owens against Detroit - 21 yards, 0 TDs.
Rushing:
Westbrook against Detroit - 110 yards, 2 TDs.
Entire Cowboys Team against Detroit - 87 yards, 2 TDs.
Yeah, that Westbrook guy, he's kinda good, I guess.
#23:
You've got to be kidding. Norwood has missed a total of 3 games in 2 years, which is a trivial number for a RB. It's absurd to affix the "injury-prone" label to a guy who has suffered a couple of minor injuries, without his production suffering at all.
It's also extremely ironic, given that you spend the rest of your post praising Westbrook, who has NEVER played a "full season" (by your definition) in any of his SIX years in the league! In fact, early in his career, the knock against Westbrook was exactly the one you're making against Norwood: that he was too small and too injury-prone to be able to handle a full load. Pretty silly, huh?
The fact is the Falcons will never know whether Norwood is capable of carrying the load unless they try giving it to him, which their recent signing of Turner seems to suggest that they remain unwilling to do. It may end up working out OK for them, as Turner certainly represents a significant upgrade over the broken-down shell of Warrick Dunn, but I still say Norwood deserves better.
Norwood has missed a total of 3 games in 2 years, which is a trivial number for a RB. It’s absurd to affix the “injury-prone†label to a guy who has suffered a couple of minor injuries, without his production suffering at all.
I didn't say he was a bad player. And maybe he's not that injury-prone, but still, I think it makes sense to be cautious with a young RB that's a little on the light side, so that you don't overwork him and get him really badly injured. So, sure, increasing his carries isn't out of the question, but I don't think he should be getting 15-20 carries a game. 8-12 carries a game (which would be a modest increase over his previous years) is probably a better idea. If he can handle that, great, keep increasing his carries gradually over the years, and see what happens. But a full load of carries is way too risky at this point in his career.
In fact, early in his career, the knock against Westbrook was exactly the one you’re making against Norwood: that he was too small and too injury-prone to be able to handle a full load.
Yes, and early in his career, Westbrook was too small and injury-prone to handle a full load of carries. That's why the Eagles' coaching staff wisely decided not to give him a full load of carries. You think it's some huge coincidence that this past year was the first time Westbrook has even gone remotely close to 300 carries in a season? That he only had 20+ carries in a game twice in his first 4 seasons? That he's had 30+ carries in a game a grand total of 1 time (in his sixth year)?
Westbrook is injury-prone (like seemingly every other player on the Eagles' roster, but that's neither here nor there). He's one of the best RBs, if not the best RB, in the NFL, in spite of that. And he didn't get that way by having a full load of carries in his third year in the league. In fact, it might be helpful to look at how many carries Westbrook had in his first few seasons, for comparison purposes:
2002 - 46 carries
2003 - 117 carries
2004 - 177 carries
2005 - 156 carries
Not exactly threatening the Curse of 370, was he? Hell, until his 5th season, he wasn't even threatening the Curse of 270. And I don't think it had anything to do with his team disrespecting his abilities. They just didn't want to lose one of their best players to a career-ending injury, so they played it safe until they were sure he could handle more and bounce back from whatever minor injuries he had along the way.
Westbrook didn't have 300+ carries in any of his first few seasons. He didn't last year, and he probably never will. But that didn't stop him from leading the league in rushing DPAR/DYAR in 2007.
Look, I'm not criticizing Norwood. I think he's done an excellent job so far, and he has a good chance of developing into an elite RB. But he's not going to do it in a hospital. If you asked Norwood to carry the ball 300 times next season, he'd probably run it very well for maybe 200-250 carries, easily break 1000 yards, and then break his legs. That doesn't help him, the Falcons, or anyone else involved. Caution is the way to go with undersized RBs. It can yield fantastic results.
And let's be honest, it's not like a little more Jerious Norwood is all the Falcons needed to be a playoff team last year. Winning 5 or 6 games instead of 4 isn't a good reason to risk a major injury to a young RB. In contrast, if Warrick Dunn suffers a career-ending injury, not such a big deal, since he should be retiring soon anyway (well, ok, I guess it is a big deal to Dunn, but not because it's going to rob him of his prime or something). Norwood will have plenty of time to get more carries when the Falcons are actually competitive, and his efforts can be put to good use. But there's no reason to rush things.
The fact is the Falcons will never know whether Norwood is capable of carrying the load unless they try giving it to him, which their recent signing of Turner seems to suggest that they remain unwilling to do.
I might be in the minority in this opinion, but I doubt the Falcons are going to expect either Turner (who has never had more than 80 carries) or Norwood to "carry the load", at least not next year. My guess is that Norwood and Turner will split carries roughly evenly, and the Falcons will use each of them in the situations where they are most effective, maximizing their value and minimizing their injury risk. Say each of them gets somewhere between 150-200 carries. I suspect they'll form an impressive two-headed monster, and the team's rushing attack will benefit greatly.
I guess I'm having trouble seeing that as unfair to Norwood. I mean, yeah, if all Norwood wanted to do was compile stats, then that won't help him, but presumably he's a bit more interested in helping his team win, and it will help him do that. But I could be wrong about the Turner thing, we'll have to wait and see.
#25:
You make a good point about how the Eagles have used Westbrook, although I think part of the reason he's gotten more carries the last couple of years is that Reid has been unable/unwilling to rely on McNabb quite as heavily as he once did.
Perhaps I misinterpreted your statement about 100 carries, which I took to mean that you didn't feel that Norwood should ever get much more than 100 carries. If, as you suggest, the Falcons employ a 50/50 split next year, that would seem to be a very reasonable way to employ their 2 backs. Very few backs are actually capable of thriving while getting 300+ carries per year.
However, I have serious doubts that there will be a 50/50 (or even 60/40) split. I don't think they gave all that money to Turner to watch him standing on the sidelines. I'm actually expecting Turner to get about 3/4 of the carries. I'm basing this on the fact that Dunn got about 2/3 of the carries, and Turner is younger, bigger, and figures to be much more productive than Dunn.
However, I have serious doubts that there will be a 50/50 (or even 60/40) split. I don’t think they gave all that money to Turner to watch him standing on the sidelines.
They don't have to put Turner on the sidelines to have Norwood on the field. There's no rule against using both of them at once. In fact, I think that's probably the best idea for the Falcons. And not just for running. They can have one split out wide, or in the slot, while the other stays in the backfield to block or sell a play action fake. But in any case, I think Norwood is going to be getting more carries this year than he did last year. I think the main reason they paid Turner so much is that he was a free agent, whereas Norwood was already on the team, so they don't need to offer him a huge contract.
I’m basing this on the fact that Dunn got about 2/3 of the carries, and Turner is younger, bigger, and figures to be much more productive than Dunn.
Maybe, but I don't think Dunn was viewed as a long-term investment the way Turner is, so the team didn't have to be as careful with Dunn. With Turner, I think they'll probably give him a reasonably light workload, but greater than his workload in previous seasons, before they count on him to take on a full workload.
Still, since my belief is based on the idea that Falcons' management has some small modicum of intelligence, it is obviously quite suspect.
Btw, I must say, Raiderjoe is right about Sierra Nevada Pale Ale. It's very good stuff. I only had two before I wrote this post, and I almost made some very obvious spelling/grammar mistakes. I can totally understand why that kind