Broncos and Turnovers: It's time to play takeaway, not giveaway

In part 1, we saw that Broncos have been below average at holding onto the ball in recent history, but we also saw that Broncos have not been very good at taking the ball away from the other team either. The table below shows how the Broncos have fared in the past 20 seasons at forcing turnovers and where we have ranked relative to the rest of the league.

Broncos	Takeaways
	Total	Lg Rank	INT	Lg Rank	FUMBLES	Lg Rank
2013	26	16	17	12	9	21
2012	24	16	16	13	8	22
2011	18	28	9	28	9	19
2010	18	32	10	30	8	24
2009	30	9	17	13	13	6
2008	13	32	6	31	7	26
2007	30	11	14	23	16	5
2006	30	10	17	13	13	10
2005	36	4	20	6	16	4
2004	20	29	12	27	8	28
2003	20	28	9	32	11	19
2002	22	29	9	30	13	12
2001	37	5	22	6	15	5
2000	44	2	27	2	17	5
1999	26	25	15	16	11	18
1998	30	13	19	10	11	15
1997	31	9	18	8	13	13
1996	32	13	23	4	9	26
1995	21	29	8	30	13	15
1994	26	20	12	22	14	7

As you can see from the above table, the Broncos have been average or below average at forcing turnovers for most of the past 20 seasons. We were in the top 3^rd of the league in only 7 of the past 20. We were in the top quartile in only 3 of the past 20 season in forced turnovers (2000, 2001 and 2005). You have to go back to 2009 to find a year in which we ranked in the top 10 in either INTs or defensive fumbles recovered. The 13 turnovers forced by the Broncos in 2008 were only one more than the fewest ever forced in a 14 or 16 game season (the 82 Colts only forced 11 in 9 games due to the strike; the 06 Skins only forced 12 in 16 games). The best takeaway total, 61, for an NFL season was the 83 Skins (Sandy Eggo forced 66 in the AFL in 1961).

Can teams "learn" to force more turnovers?

How do we address this? One way is to look at year over year changes in the numbers of takeaways. I tracked this for the entire NFL going back to 2000 and here is what I found: First the bad – change relative to previous season

Worst Year-to-Year Takeaway Change 2000-present

Now the good

Best Year-to-Year Takeaway Change 2000-present

Year	Change	Team
2013	23	KC
2004	21	BUF
2007	20	STL
2003	20	SD
2013	18	PHI
2009	17	DEN
2000	17	NO
2006	17	NE
2001	17	CHI
2001	17	CLE

FWIW - the 2005 Broncos just missed the top 10 by forcing 16 more turnovers than the year prior. Looking at the whole set of data what you find is that while a team can change dramatically from one year to the next (see the 2013 Chiefs), in almost every instance that team then regressed to the mean (or worse) in the next season. What goes up has to come back down (or does it?). Look at some of those teams who made the biggest jumps:

• 2005 Bills had 9 fewer takeaways than the 04 team
• 2008 Chargers forced 24 fewer than the 07 team
• 2004 Rams had 31 fewer takeaways than the 03 Rams
• 2010 Broncos forced 12 fewer TOs than the 09 team
• 2010 Saints forced 14 fewer than in the previous season

The team on the good list who had the least regression in the nest season was the 2006 Pats who only had 4 fewer takeaways in 2007. Based upon this, the Chiefs and the Eagles will most likely regress to the mean next season which would mean a drop of 10 for the Chiefs and a drop of 5 for the Eagles. This is what history tells us: it is almost impossible to make a huge jump in one year and then sustain that level of takeaways. Since 2000 no team has been able to do that. See the chart below for how often big changes in takeaways happen (total occurrences in the NFL since 2000).

Change in Takeaways	# of Occurences	Frequency
more than or equal to +10	55	12.4%
less than or equal to -10	64	14.4%
more than or equal to +15	17	3.8%
less than or equal to -15	22	4.9%
more than or equal to +20	4	0.9%
less than or equal to -20	6	1.3%

One out of eight teams will force 10 or more turnovers next year than they did in 2013 (so, 4 teams). One out of seven teams will force 10 fewer TOs (or worse) than they did in 2013. The instances of teams changing by 20 or more (for good or bad) are even more rare and it is slightly more likely for a team to go in the bad direction than in the good direction. The really drastic swings usually involve a team moving from the top to the bottom (or the bottom to the top) of the league in one season. The 2013 Chiefs and Eagle are great examples of this – they were tied for the fewest takeaways in 2012 at 13 (one more than the worst ever). Both teams improved dramatically such that the Chiefs were 2^nd in the league in 2013 and the Eagles were tied for 5^th (with TB and CIN).

So what does this mean for the Broncos?

The odds of the Broncos forcing 36 TOs next season (after forcing 26 in 2013) are worse than the odds of forcing 16 next season. Interestingly, last season 19 teams improved in takeaways relative to 2012 while only 12 regressed (this bucks the historical trend). So let’s talk about teams from the recent past who have improved gradually and then stayed near the top of the league in forced TOs once they got there. The Broncos have improved in takeaways over the past two seasons (+2 and +6). Quick gains in takeaways appear to not be sustainable, but gradual improvements can be sustained. So the Broncos might be able to follow the example of other teams that have gradually improved to the point of being at the top of the league. The best five examples of gradual improvement are shown below (with DEN’s best shown at the end). Teams who regressed at any point during a time-frame are not shown.

Years	Team	Magnitude of Change
2003-06	CHI	24
2002-05	CIN	24
2010-13	SEA	17
2000-04	IND	14
2008-11	DET	14
2004-07	DEN	10

Seattle is included even though they had a net zero between two of the seasons (they didn't improve, but they didn't regress). The Colts are the only team to have shown an improvement over 5 consecutive seasons. Again this is only since 2000, I did not go back any farther. The Broncos improved from bottom of the league (20 forced TOs) to top third (30 forced turnovers) in 2004-2005 then they stayed at that level until 2008 when the defense hit rock bottom. Let’s look at three of these teams in depth (CHI, CIN and SEA) to see if we can learn anything from their gradual improvement. How did they do it?

Bears:

In 2003 the Bears were in the near bottom of the league in takeaways (20, T-31st). In 2004 they moved up to 16^th. In 2005 they moved up to 6^th and in 2006 they lead the league in takeaways. They did this by gradually improving their INT numbers and then having a big jump in recovered fumbles.

Year	Takeaways	INT	Fumbles Recovered
2006	44	24	20
2005	34	24	10
2004	29	17	12
2003	20	15	5

In terms of defensive personnel, the Bears started getting better in 2004 with a MLB in Urlacher (perennial Pro-Bowler), a pro-bowl CB (Jerry Azumah) that had a great season and an injured CB that was an emerging star, Charles Tillman. R.W. McQuarters was obtained from the 49ers and started when Tillman went down. The biggest change from 03-04 for the Bears was their ability to pressure the QB. They only notched 18 sacks (worst in the league) in 03 and they had 35 in 04 (22^nd in the league, still not good, but better than worst). Interestingly, not a single Bears defender made the 04 Pro-Bowl. In 2005 the Bears defense really came into its own. They had 5 defensive starters make the Pro-bowl (and 2 All-Pros – Urlacher and Briggs). Their defensive line was a dominant force although there wasn’t a single dominant pass rusher (Ogunleye led the team in sacks with 10). Defensively, they forced 18 fumbles. Their starting defensive backfield combined for 19 INTs. They were tied for 9^th in sacks (41) in 2005. Where they really shined in 2005 was in QB rating against; they were the best in the league with a 61.2 QB rating against them. That 61.2 makes them one of the best passing defenses in the past 20 years. They only got better in 2006, particularly on the forced fumbles ledger where they notched 25 forced fumbles – recovering a remarkable 20 of those. Their 44 takeaways led the league and they were second in INTs and first in recovered fumbles.

So this is what the Bears did, but we’ve yet to understand HOW they did it. The how can be traced largely back to the Lovie Smith. He is known for teaching his players how to generate takeaways. He also had a defensive coordinator who is known for teaching players how to generate takeaways, Ron Rivera. So the Bears had a combination of better defensive players and directed coaching to move from last to first in takeaways as a team in the space of four seasons. Here’s a direct quote from Smith about teaching turnovers

"… but with our system of single-gap football, passing downs being able to mix in a little bit of Cover-Two, but for us the thing that dictates whether you win or lose each week the most will be turnover ratio. As you look at our time in Chicago from 2004 and on, on the defensive side, we took the ball away more than any team around – that didn’t just happen by chance. It’ll be the way that we practiced it and the mindset that we’ll have for our players."

Bengals:

In 2002 the Bungals were 31^st in the league in takeaways – dead last in INTs. In 2003, new head coach Marvin Lewis brought a different attitude and a defensive mindset to the team. The team improved from 2-14 to 8-8, but that improvement could not really be tied to the slight improvement in takeaways (+4).

Year	Takeaways	INT	Fumbles Recovered
2005	44	31	13
2004	36	20	16
2003	24	14	10
2002	20	9	11

The biggest factor in that improvement was a dramatic change in the defensive players. Only 3 of the starting defenders from 2002 started for the Bungals in 2003 (Justin Smith, Tony Williams and Artrell Hawkins).

The 2002 Bungals were dead last in points allowed. The new defensive starters allowed them to move up to 25^th in points allowed (28.5 to 24.0 ppg). The real jump occurred in 2004 when the Bungals forced 12 more turnovers than the previous season. The 2004 Bungals were still in the bottom third of the league in points allowed (21^st), but somehow they generated more takeaways. They had 3 "new" starters on D in 2004, two were rookies (FS, Madieu Williams, and MLB, Landon Johnson) and one was a former Bronco (Deltha O’Neal) who joined the other former Bronco on the Bungals, Tory James, as the starting CB duo. In total the 2004 Bungals D only had 4 players who were drafted by CIN starting. So how do we explain the big jump in takeaways? Well, it probably wasn’t a result of increased pressure on the QB as the 2004 Bengals only had 7 more sacks than the 2003 Bengals (they were from average to slightly above average). The biggest thing that I can see different between the two teams was an improved ability to stop the run (in 03 they gave up 4.8 ypc, in 04 they improved to 4.4 ypc allowed). Teams were forced to throw a little more and with O’Neal on one side, Tory James took advantage of being targeted more to turn in his career year (8 INTs, 13 PD, 2 FF). The Bungals continued to improve in 2005, in which the team finished 11-5 after two straight 8-8 years. The defense was still in the bottom third of the league in points allowed (22^nd), but now the D lead the league in takeaways with 44 (+8 from the previous season). Additionally the O was the fifth best team in the league at NOT giving the ball away.

So we have another team, like the Bears who went from worst to first in four seasons in terms of takeaways. What was the big change from 04 to 05? It was a dramatic increase in INTs (+11). This was done with 2 new starting safeties (Ohalete and Kaesviharn – both veterans who were undrafted) and another rookie starter (LB, Odell Thurmon; Landon Jonhston started as a rookie at LB in 04). Thurmon, Ohalete and Kaesviharn combined for 9 of the team’s league-best 31 INTs. 31 interceptions was 7 more than the second best team in the league (CHI and MIN both had 24). 2005 was O’Neal’s career year (All-Pro, 10 INTs, 20 PDs), but Thurmon was a huge contributor as a rookie (104 tackles, 5 INTs, 5 FFs, 9 PDs, 1.5 sacks, 1 TD, 1 FR). Thurmon was busted for PEDs and recreational drugs and never again played in the NFL. What’s really interesting is that the Bengals had this huge jump in INTs while regressing in terms of pass rush. The 04 Bengals tallied 37 sacks. The 05 Bengals only notched 28. By many measures, the 05 defense was worse than the 04 defense. Looking more closely at the 05 Bengals, they get most of the takeaways in 5 games: 7 vs MIN, 6 vs CHI, 5 vs GB, 4 vs DET and 4 vs PIT: 26 of 44 takeaways in those 5 games. The NFC North really sucked that year (at least against CIN). Part of that jump in takeaways can be chalked up to the QBs they were facing in those games

Dante Culpepper – 5 INTs (the worst game of his career)

Kyle Orton – 5 INTs (the single worst game of his career)

Brett Favre – 1 TD, 5 INTs (one of the worst games of his career, if not the worst)

Jeff Garcia – 1 TD, 3 INTs (one of the worst games of his career)

Was this luck or was the good defense? I don’t know, but I do know that the Bungals weren’t able to stay on top once they got there. In 2006 the Bengals were still good in takeaways (31, 8^th in the league), but that was -13 relative to 2005. This drop was due to 12 fewer INTs in 06 relative to 05 – which could be attributed to having 3 new starters in the secondary (from the 05 secondary only James started more 11 games for the Bengals).

Did Cincy’s improvement have more to do with coaching (Marvin Lewis as HC; Leslie Frazier, Chuck Bresnahan, Mike Zimmer as DCs) or with players? Well we can look at what CIN has done in terms of takeaways during the rest of Lewis’ tenure as HC (Dick LaBeau was the HC in 2002)

Year	Takeaways	NFL Rank
2013	31	3
2012	30	8
2011	22	22
2010	26	17
2009	25	19
2008	24	17
2007	35	3
2006	31	8
2005	44	1
2004	36	3
2003	24	23
2002	20	31

Six out of 11 seasons in top third of the NFL in takeaways with 4 season in the top 3 and only two seasons in the bottom third (with 3 different DCs). That would say to me that this is more about the coaching than the players, since there is only one defensive starter on the 2013 Bungals who started for the 2007 Bungals (NT - Peko). Marvin Lewis is also on the record as being a coach that actively teaches "takeaways" to his defense.

Seahawks:

The 2010 Seachickens had only 22 takeaways in Carroll’s first season as HC after getting run off from SoCal with his tail between his legs for cheating. 22 takeaways was 25^th in the league. So unlike CHI and CIN, SEA did not start from dead last for their improvement.

Year	Takeaways	INT	Fumbles Recovered
2013	39	28	11
2012	31	18	13
2011	31	22	9
2010	22	12	10

FWIW, the 2009 Seachickens with Jim Mora as the HC, actually forced 23 turnovers, one more than in Pete Carroll’s first season. Pete Carroll is known for "teaching" takeaways. The 2011 team team that got 9 more takeaways featured 7 new defensive starters relative to 2010: Bryant, Branch, Hill, Wright, Browner, Sherman and Chancellor. Three of those guys were rookies in 2011. With the exception of Browner who was suspended for a good portion of the season, all of those guys played big roles in the 2013 SEA defense which led the league in turnovers. So the big change in players came in 2011. The +10 in INTs can’t be attributed to an improved pass rush (in 2011 SEA had 4 fewer sacks than in 2010) unless that pass rush was getting more pressure WITHOUT getting more sacks. The +10 in INTs is completely on the new starters who accounted for 16 of the team’s 22 INTs. The only returning starters who got INTs were Hawthorne (3), Thomas (2) and Trufant (1). The 2011 SEA defense had a 71.8 passer rating against compared to a 74.8 value for the 2010 team. That’s fairly comparable. The 2013 team that led the league in takeaways really only had two new starters relative to the 2012 team – DT McDaniel (replaced Branch) and WLB Irvin (replaced Hill). The 2013 team was +10 in terms of INTs relative to the previous year (they were also +7 in sacks). In terms of INTs, Maxwell and Chancellor combined for 7 regular season INTs after 0 the year before. Along with the dramatic increase in INTs, SEA also saw a marked improvement in PDs (77 in 12 to 111 in 13, +34). The 2013 passer rating against continued its improvement to 64.8 (down from 71.8 in 2012).

Seattle’s improvement in takeaways appears to be both a function of coaching (Carroll’s "Turnover Tuesdays") and players. The general theme amongst all of the new starters on the 2011 Seachickens is that they are all fast. They get to the ball quickly and they are not afraid to miss a tackle while attempting to separate the ball from the ball-carrier.

Back to Denver

Of the three teams that we looked at in depth, only the Bears had a big jump in fumble recoveries to make their leap to the top in takeaways. With two players rushing the passer who are adept at the strip-sack (Miller and Ware) we should have an improvement in fumble recoveries in 2014 (with any luck, see below). Additionally we added a Safety in Ward who has as many forced fumbles in career (5) as interceptions (admittedly, he really isn't known for generating turnovers). The other big defensive signee, Talib has a career high of 6 INTs (in 11 games in 2010) so his presence might help increase our INTs next season, but more than his presence at CB, our improved pass rush should generate more opportunities for INTS among all of our DBs. I would not be surprised to see Webster or Carter end up with significantly more INTs this year based solely on being in the right place at the right time on forced or errant throws due to QB pressure.

On the fumble side, I fully expect a healthy VonWare to force a host of fumbles. The fewest forced fumbles that Ware has had in his career when healthy (he had 0 last year) is 2. He has played 9 NFL seasons (141 games) and has 32 career forced fumbles (3.6 per season, 1 every 4.4 games). Von Miller has 12 career forced fumbles in 40 games (4 per season, 1 every 3.3 games). VonWare will get some stripsacks, but the Broncos have to be lucky enough to fall on the ball after we force the fumbles. Much like the offensive side of the ball, we have had some lucky years and some unlucky years in terms of gaining possession after forcing fumble on defense

Broncos Defense - Fumbles

Year	Forced	Recovered
2013	16	56%
2012	19	42%
2011	12	75%
2010	13	62%
2009	20	65%
2008	9	78%
2007	20	80%
2006	20	65%
2005	19	53%
2004	16	50%
2003	17	65%

In 2012 we forced 19 fumbles but only recovered 8 of those. Conversely, in 2007 we forced 20 and recovered 16 of them. When a single turnover can mean the difference between a win and a loss, the luck involved in recovering a forced fumble is something to consider.

Conclusion

So you've seem the data and the history. You know what players we have coming in on D and who we have coaching. Players matter in generating takeaways, but so does coaching. Do you think that the Broncos can improve in 2014 in takeaways and if so how much? Do we have the right players? Do we have the right coaching? Sound off in the comments.

Go Broncos!