Writers of Pro Football Prospectus 2008

14 Jun 2011

Broken Tackles 2010 Part II

by Aaron Schatz

Today, more of our series presenting various 2010 stats from the multitude of Football Outsiders spreadsheets. Last week, we looked at broken tackles from the point of view of the defense. Today, let's look at broken tackles from the point of view of the offense.

As I noted last week, broken tackles are a stat from game charting, not from the standard play-by-play. We defined a "broken tackle" as one of two events: either the ballcarrier escapes from the grasp of the defender, or the defender is in good position for a tackle but the ballcarrier jukes him out of his shoes. If the ballcarrier sped by a slow defender who dived and missed, that didn't count as a broken tackle. We only measured broken tackles for standard plays; volunteers didn't have the time to track them for all special teams plays.

The resulting numbers are subjective, obviously, but there were over two dozen charters involved, so no team's numbers could be overly slanted because of the bias of a single specific charter. We know that there are a other groups on the Web who track broken tackles, and because of the subjectivity, their numbers won't be exactly the same as ours. Given the mistakes that are easy to make when marking players off of television tape, a difference of one or two broken tackles isn't a big deal. But looking at the players with the most and fewest broken tackles does a good job of showing us which ballcarriers are able to power through defenders -- or avoid them with agility -- and which ballcarriers go down quickly when there's contact.

Here is a list of all running backs with at least 20 broken tackles in 2010:

Most Broken Tackles, 2010 RB
Player Team BT Runs Rec Touches BT/Touch
23-A.Foster HOU 45 325 68 393 11.5%
28-C.Johnson TEN 41 316 44 360 11.4%
25-L.McCoy PHI 38 207 78 285 13.3%
44-A.Bradshaw NYG 38 275 48 323 11.8%
33-M.Turner ATL 38 334 12 346 11.0%
20-D.McFadden OAK 37 223 47 270 13.7%
24-M.Lynch BUF/SEA 36 201 22 223 16.1%
28-A.Peterson MIN 35 284 36 320 10.9%
40-P.Hillis CLE 35 270 61 331 10.6%
22-F.Jackson BUF 34 222 31 253 13.4%
27-L.Blount TB 33 201 5 206 16.0%
34-R.Mendenhall PIT 33 324 23 347 9.5%
Player Team BT Runs Rec Touches BT/Touch
28-F.Jones DAL 30 184 49 233 12.9%
32-M.Jones-Drew JAC 30 300 34 334 9.0%
32-C.Benson CIN 30 321 28 349 8.6%
39-S.Jackson STL 28 330 46 376 7.4%
29-C.Ivory NO 27 137 1 138 19.6%
25-J.Charles KC 27 231 45 276 9.8%
28-J.Stewart CAR 25 178 8 186 13.4%
27-K.Moreno DEN 25 183 37 220 11.4%
46-R.Torain WAS 22 164 18 182 12.1%
27-R.Rice BAL 22 307 63 370 5.9%
20-J.Forsett SEA 21 118 33 151 13.9%
24-C.Williams TB 21 125 46 171 12.3%
35-M.Tolbert SD 21 181 26 207 10.1%
39-D.Woodhead NE 20 97 34 131 15.3%

Now, leaguewide we marked five percent fewer broken tackles this year compared to 2009. That's probably not tackling getting better, its just a natural year-to-year swing when we're still tuning the proper definition of a subjective metric. Still, even taking that into consideration, there were some surprising drops from 2009 to 2010. We marked Ray Rice with 57 broken tackles in 2009, second in the league behind Chris Johnson. That dropped by more than 60 percent, to just 22 broken tackles last year. Maurice Jones-Drew dropped from 52 to 30. Steven Jackson dropped from 49 to 28. Further down our list, Joseph Addai dropped from 29 broken tackles to just nine, and Beanie Wells dropped from 20 broken tackles to just four.

There's really no equivalent player who had the same amount of touches in both 2009 and 2010 but had many more broken tackles in 2010. The closest is probably LeSean McCoy, who went from 24 broken tackles to 38.

(If you want to see 2009 numbers, click here. The individual numbers are an ESPN Insider piece, the team numbers were free on Football Outsiders.)

Here's another way to look at things, the highest and lowest rates of broken tackles per play. We're adding together catches and carries to get the total number of touches for each player. This is just running backs, with a minimum of 80 touches:

Highest Broken Tackle Rate, 2010 RB
Player Team BT Runs Rec Touches BT/Touch
29-C.Ivory NO 27 137 1 138 19.6%
24-M.Lynch BUF/SEA 36 201 22 223 16.1%
27-L.Blount TB 33 201 5 206 16.0%
39-D.Woodhead NE 20 97 34 131 15.3%
23-P.Thomas NO 17 82 30 112 15.2%
20-J.Forsett SEA 21 118 33 151 13.9%
20-D.McFadden OAK 37 223 47 270 13.7%
28-J.Stewart CAR 25 178 8 186 13.4%
22-F.Jackson BUF 34 222 31 253 13.4%
25-L.McCoy PHI 38 207 78 285 13.3%
Lowest Broken Tackle Rate, 2010 RB
Player Team BT Runs Rec Touches BT/Touch
23-T.Choice DAL 2 66 17 83 2.4%
26-C.Wells ARI 4 116 5 121 3.3%
21-L.Tomlinson NYJ 10 220 52 272 3.7%
30-J.Kuhn GB 4 84 15 99 4.0%
42-B.Green-Ellis NE 10 229 12 241 4.1%
21-F.Gore SF 11 203 46 249 4.4%
29-C.Taylor CHI 6 112 20 132 4.5%
23-S.Greene NYJ 10 185 16 201 5.0%
20-T.Jones KC 13 246 14 260 5.0%
22-M.Forte CHI 15 236 51 287 5.2%

Yes, that's two undrafted rookies in our top three, and four undrafted players in our top five. That group of undrafted players is broken only by Marshawn Lynch, who apparently was already in Beast Mode before the playoffs even started. Remember, these are regular season numbers only, and don't include the big run against New Orleans. By the way, Lynch is a great example of how the ability to break tackles is very different from the ability to push through the line in short-yardage situations. As Doug Farrar pointed out on Yahoo! a few days ago, Lynch was one of the worst running backs in the NFL on third downs last season, despite all these broken tackles.

I am surprised by how few broken tackles we marked down for Frank Gore. He wasn't particularly high in our 2009 charting either, with 24 broken tackles on 281 touches (8.5 percent).

The charters did a better job of marking broken tackles by quarterbacks this year, so we were able to split out quarterbacks evading sacks from receivers evading tackles after the reception. Michael Vick, no surprise, led all quarterbacks in broken tackles with 20; that includes both missed sacks and missed tackles after he scrambled past the line of scrimmage. Only three other quarterbacks were marked with more than five broken tackles: David Garrard (11), Tim Tebow (11), and Josh Freeman (10).

Percy Harvin led all wide receivers in broken tackles for the second year in a row, with 13. Here's a look at the top ten wide receivers and tight ends in broken tackles:

Most Broken Tackles, 2010 WR/TE
Player Team BT Runs Rec Touches BT/Touch
12-P.Harvin MIN 13 18 71 89 14.6%
85-P.Garcon IND 12 2 67 69 17.4%
13-N.Burleson DET 12 7 55 62 19.4%
11-B.Gibson STL 11 3 53 56 19.6%
84-J.Morgan SF 11 2 44 46 23.9%
84-J.Gresham CIN 10 0 52 52 19.2%
81-T.Owens CIN 10 0 72 72 13.9%
82-K.Winslow TB 9 0 66 66 13.6%
19-B.Marshall MIA 9 2 86 88 10.2%
82-M.Manningham NYG 9 1 60 61 14.8%
80-E.Bennett CHI 9 3 46 49 18.4%

A list with the lowest rate of broken tackles for wide receivers would be kind of silly, since there were a lot of wide receivers and tight ends with only one or two broken tackles. Of note, we had only two receivers who had more than 30 catches with zero broken tackles according to our charters, and both were Houston Texans: Andre Johnson and Kevin Walter. And it isn't like the people charting Houston were particularly stingy about marking broken tackles -- they marked plenty for Arian Foster, and remember that Houston led the league in broken tackle rate on defense.

Finally, here's the overall list for broken tackles on offense:

Broken Tackles by Offenses, 2010
Offense Plays Plays w BT Total BT Pct Plays w BT
TB 951 77 94 8.1%
PHI 1022 77 92 7.5%
JAC 1014 72 80 7.1%
BUF 950 67 73 7.1%
NYG 1022 71 80 6.9%
CLE 914 61 68 6.7%
CAR 962 63 76 6.5%
OAK 1024 67 75 6.5%
SEA 947 61 71 6.4%
ATL 1090 68 78 6.2%
DAL 1026 63 73 6.1%
NO 1053 64 75 6.1%
MIN 983 59 65 6.0%
NE 970 57 69 5.9%
TEN 902 53 61 5.9%
CIN 1039 61 71 5.9%
Offense Plays Plays w BT Total BT Pct Plays w BT
DEN 1017 58 67 5.7%
PIT 975 55 68 5.6%
HOU 1023 57 66 5.6%
KC 1052 58 62 5.5%
STL 1046 57 65 5.4%
WAS 993 54 62 5.4%
MIA 1031 55 57 5.3%
SD 1020 54 61 5.3%
DET 1057 53 56 5.0%
SF 937 46 58 4.9%
IND 1081 52 60 4.8%
CHI 924 41 47 4.4%
ARI 933 40 45 4.3%
BAL 1009 43 48 4.3%
GB 988 42 46 4.3%
NYJ 1078 43 44 4.0%

Posted by: Aaron Schatz on 14 Jun 2011

41 comments, Last at 18 Jul 2011, 5:54pm by Andy

Comments

1
by Russell (not verified) :: Tue, 06/14/2011 - 5:54pm

I suspect that Matt Schaub has alot to do with Houston wide receiver's lack of broken tackles. It was almost painful to watch all the times Johnson would have to slow down or even come to a complete stop waiting for Schaub's weak throws to get to him. Pretty easy to tackle a guy standing still.

33
by Mr Shush :: Wed, 06/15/2011 - 7:38pm

Schaub doesn't have a great deep ball, it's certainly true, but Johnson also played almost the whole season with an ankle injury and may well have been going down on initial contact to avoid exacerbating it. Walter just plain isn't powerful or elusive enough to be likely to break a lot of tackles.

2
by tunesmith :: Tue, 06/14/2011 - 5:54pm

Tebow with 11 broken tackles after playing less than 1/4 the season... yeesh.

3
by Aaron Brooks' Good Twin (not verified) :: Tue, 06/14/2011 - 6:59pm

How did Nate Burleson get to 12 BTs?

4
by Karl Cuba :: Tue, 06/14/2011 - 7:09pm

Do you have a figure for the average league-wide rate for breaking tackles as a point of comparison?

Also, I'm really looking forwards to Harbaugh's offense, get the ball in the hands of Davis, Crabtree and Morgan (with his break rate of over 20%) and let tham make a play. I just hope these current rumours about progress in the labour talks have some truth to them because the niners need training camp more than almost every team, their players haven't been coached in the scheme (and arguably haven't really been coached well for about ten years).

5
by tunesmith :: Tue, 06/14/2011 - 7:11pm

By the way, I think there's a bug with the site in that the audio for the ads starts playing on page load.

6
by Intropy :: Tue, 06/14/2011 - 7:26pm

"Touches" is a bit confusing. I originally thought it was the number of times a defender had touched the carrier. Doesn't that make more sense for the denominator than number times the player possessed the ball?

Anyone have the numbers for Roethlisberger?

18
by Karl Cuba :: Wed, 06/15/2011 - 8:08am

With all the brouhaha surrounding Roethlisberger I think Florio would have reported if he'd broken his tackle;)

32
by MJK :: Wed, 06/15/2011 - 6:29pm

Doesn't that make more sense for the denominator than number times the player possessed the ball

I thought the same thing at first, but I don't think so upon reflection. I imagine counting times a defender touched the carrier is highly subjective...and in the cases where the carrier jukes past a defender in good position (or high-jumps over him) and never gets touched, that would give you one broken tackle divided by zero touches...so you can skew the ratios really easily.

On the other hand, using "ball touches" (receptions plus rushes) makes a lot of sense...it allows you to compensate for the fact the different running backs are utilized differently.

For a great example, consider Danny Woodhead. He barely makes the list of RB's with at least 20 broken tackles, and is 26th on the list. So we might conclude that he isn't that great at tackle-breaking. However, the Patriots use a committee approach at RB, and even when Woodhead or BJGE were heavily utilized in a game, each got only about half the carried. On the other hand, LeSean McCoy and Chris Johnson were getting something like 95% of their team's RB touches, so they had a lot more opportunities to break tackles. Ball touches generally directly reflects a player's snap count and therefore allows you to compensate for utilization differences. When you factor in ball touches, Woodhead is 4th on the list, implying that he is indeed a fairly shifty back.

35
by Shattenjager :: Wed, 06/15/2011 - 7:50pm

Chris Johnson had 82.9% of his teams RB touches (he had 360, Javon Ringer had 58, and Ahmard Hall had 16).
LeSean McCoy had 72.7% (he had 285, Jerome Harrison had 48, Eldra Buckley had 22, Mike Bell had 17, Leonard Weaver had 1, and Owen Schmitt had 19).
BenJarvus Green-Ellis had 51.9% (he had 241) and Danny Woodhead had 28.2% (he had 131, Fred Taylor had 45, Sammy Morris had 27, Kevin Faulk had 14, and Thomas Clayton had 6).

I just added these up because I was curious about what their percentages actually were.

7
by Sequence (not verified) :: Tue, 06/14/2011 - 7:38pm

"...five undrafted players in our top six."

Justin Forsett was actually a 7th-round draft pick in 2008.

9
by Aaron Schatz :: Tue, 06/14/2011 - 8:47pm

Whoops! I'll go fix that.

8
by Reinhard (not verified) :: Tue, 06/14/2011 - 8:44pm

Broken tackles don't take place in isolation. In a slightly bigger hole it's easier to break the tackle, and it's harder to make it.
I'm thinking of Arian Foster. I actually really like him and don't get why he is so disliked, but the players voting his FB one of the top players in the game, along with Andre Johnson the best WR in the NFL out there, it does open things up for him nicely.

34
by Mr Shush :: Wed, 06/15/2011 - 7:42pm

That may be true, but subjectively I would say that Foster has outstanding agility in the open field coupled with pretty good power. He made a lot of significant gains on plays that could easily have been ended in the backfield.

36
by Shattenjager :: Wed, 06/15/2011 - 7:51pm

Arian Foster is disliked? (This is not arguing--I'm seriously asking.)

37
by Sander :: Wed, 06/15/2011 - 9:05pm

Actually the vast majority of broken tackles tracked by game charters occur in the open field, at least in my experience. Tackles around the line of scrimmage often don't factor in because there's almost always a blocker involved somewhere.

10
by Billy B (not verified) :: Tue, 06/14/2011 - 8:51pm

Yes, it seems like Roethlisberger evades a sack on every single play. Or at least on every single big pass play. So it's surprising to see that game charters didn't see more than 10.

23
by Aaron Brooks' Good Twin (not verified) :: Wed, 06/15/2011 - 10:40am

Usually though he never actually broke the tackle -- he just threw the ball with three guys hanging off of him.

26
by Intropy :: Wed, 06/15/2011 - 2:46pm

By the definition given that would be a broken tackle since it's an escape from the grasp of a defender.

27
by tuluse :: Wed, 06/15/2011 - 3:23pm

No, it's the ball leaving the player, not the player escaping the grasp of a defender.

I would assume fumbling wouldn't count as a broken tackle either.

30
by Intropy :: Wed, 06/15/2011 - 5:47pm

A player forcing a fumble is credited with a tackle, so a player fumbling the ball did not break a tackle. A player forcing another out of bounds is credited a tackle. A player hanging on to a quarterback who throws a forward pass is not credited with a tackle. Going by the language above, "the ballcarrier escapes from the grasp of the defender," I can see that you may be right depending on interpretation of the language (good old de dicto vs de re making an appearance).

But in light of the way tackles are credited statistically and an understanding of the relative desirability of outcomes it seems that a once a defender has a carrier grasped a "tackle" is a play where the defender wins by making the tackle in the traditional sense, forcing a fumble, halting forward progress resulting in a dead play, or forcing the carrier out of bounds. A "broken tackle" is a play where the ball-carrier wins by eluding the tackle whether through a juke, shaking off a tackler, or executing a legal forward pass.

I don't think a thrown ball is a particularly borderline case since the ball-carrier has definitively "won" the encounter in which a grasp occurred with no subsequent tackle. More borderline scenarios would be a player who never escapes the tackler but drags him forward for several yards or until scoring or until another defender arrives to assist.

31
by tuluse :: Wed, 06/15/2011 - 6:28pm

No question that the ball carrier won the encounter, but not every victory is a broken tackle.

11
by Kent (not verified) :: Tue, 06/14/2011 - 9:43pm

I wonder how much a knee injury and then allergic reaction to treatment played into his poor performance. You could notice just watching games that he looked a little timid and was going down on first contact for to easily and far to often for a player of his size.

12
by Sciz (not verified) :: Tue, 06/14/2011 - 10:22pm

Watching Aaron Hernandez, it seemed like he broke an awful lot of tackles for a TE. Do the numbers confirm this?

13
by tuluse :: Tue, 06/14/2011 - 10:24pm

The extremely low rate of broken tackles by Jet's running backs is interesting.

That is a beast of an offensive line they've assembled.

14
by Dan :: Tue, 06/14/2011 - 11:41pm

Five the of 6 worst tackle-breaking teams made the playoffs. Coincidence? Probably.

15
by ebongreen (not verified) :: Wed, 06/15/2011 - 12:56am

Oh, I don't know about that. GB, for example, is much more interested in YAC and passing than it is in broken tackles through ground-pounding. The same strikes me as true for IND.

CHI's offense seemed to change radically during the course of the year, so I'm not sure how their philosophy is particularly measurable. OTOH, from the little I saw of both neither BAL nor the NYJ seemed to have an awesomely dynamic passing game, so I'm not sure how they'd count.

16
by tuluse :: Wed, 06/15/2011 - 1:05am

Don't you think an offense designed to create YAC would lead to broken tackles from the receivers?

17
by Ununanonymous (not verified) :: Wed, 06/15/2011 - 3:40am

The Packers stress holding onto the football at all costs. Fumblers get sent into McCarthy's dog house for a long, long time. James Jones would've played a much larger role had McCarthy trusted him.

21
by Flounder :: Wed, 06/15/2011 - 9:39am

To me, the GB offense has a couple different strengths. Fitting balls into small windows, where the receiver if often almost immediately tackled by multiple defenders (see Aaron Rodgers being really good), and also hitting receivers in stride who have beaten their defender by a step, while designing the play so there are no other defenders within close proximity (see Greg Jennings and Aaron Rodgers being really good).

28
by LionFanInAZ (not verified) :: Wed, 06/15/2011 - 4:44pm

I would think that an offense that emphasizes throwing the ball into small windows is courting with disaster.

40
by Flounder :: Thu, 06/16/2011 - 12:03pm

I didn't say they emphasize it. I'm sure GB doesn't. Nevertheless, they are good at it (see Aaron Rodgers).

38
by Sander :: Wed, 06/15/2011 - 9:07pm

A receiver making a defender miss after he's caught the ball is also a broken tackle.

19
by RichC (not verified) :: Wed, 06/15/2011 - 8:29am

"Lynch is a great example of how the ability to break tackles is very different from the ability to push through the line in short-yardage situations"

When you include dodging a defender as a broken tackle, this happens.

25
by Roadspike73 (not verified) :: Wed, 06/15/2011 - 12:32pm

Anecdotally, Lynch was always getting hit 2-3 yards behind the line, breaking a tackle, and then getting hit again (while already slowed) at the line of scrimmage. That tends to lead to a lot of broken tackles, but not much yardage. He's not (usually) the type of runner who jukes a defender out of their shoes, that's more Forsett's style (or just letting the defender slip off of him). Lynch tends to shrug off a defender's hit, but then be left so high and slow that he has no chance against the second hit--and with Seattle's makeshift offensive line last year, that second hit came almost as quickly as the first hit did.

20
by Ranccor (not verified) :: Wed, 06/15/2011 - 9:32am

Couple of observations from a Colts fan perspective:

-All of the RB is in the AFC South without a horseshoe on their helmet are always in BEAST MODE. We already knew that, but this adds to the fire. And Addai falls over when he is hit with a stiff breeze.

-Not surprised at all to see Garcon at the top of the WR chart. He was awesome once he got his hands on the ball, but made way too many easy drops to be consistently good (at least that is what my memory tells me).

22
by andrew :: Wed, 06/15/2011 - 10:05am

Is that broken tackles # of broken tackles or # of plays in which they broke a tackle?

i.e., if a running back breaks 3 tackles on one play, does it count as 1 or 3?

24
by Aaron Schatz :: Wed, 06/15/2011 - 11:17am

Aha. It is number of broken tackles, so it counts as three.

29
by LionFanInAZ (not verified) :: Wed, 06/15/2011 - 4:49pm

This confirmed my suspicion that Earl Bennett was one tough-to-bring-down WR. I was surprised to see Burleson up there as well. One more reason to be happy the Lions picked him up!

39
by Howdy (not verified) :: Thu, 06/16/2011 - 8:41am

Does down/distance have an impact here? I'm just thinking that in New England, when BJGE gets a carry he's supposed to run into the line and get 3 yards, while they often try to get Woodhead the ball in space. I agree that Woodhead's very shifty (see his TDs against Buffalo and Indianapolis) but I feel like BJGE shed's a lot first contact, or at least always falls forward.

41
by Andy (not verified) :: Mon, 07/18/2011 - 5:54pm

I've seen Andre Johnson finish very low in this (& similar) metric the past 3 seasons. The odd thing is, he's a big, strong, fast guy -- who you'd think would excel at such a thing. It makes me wonder if it's because of the way HOU uses him most of the time (as a possession receiver, not taking advantage of his athletic ability), and not his actual ability to make defenders miss (ie: does he just not have enough space based on the routes he runs? not far enough down field, or back far enough from the LOS?). I know everyone can probably picture a play or do where he just muscles guys off (vs ARI I think, going in for the TD), so it would seem he DOES have the ability, he's just not put into the position to do so.