Fantasy Defense Scoring: Coaching Changes

This is the third in a series on fantasy defenses. I started with an overview of trends in defensive fantasy scoring and last week looked at whether a defense’s “base” alignment affected its fantasy scoring. This week, I want to examine the impact on fantasy scoring of changing coaches and alignments.

A quick review of my methods: I’m studying from 2002 to the present and only looking at Tm Defensive (Tm D) fantasy scoring, NOT defense AND special teams (D/ST) scoring. Here’s the scoring system I’m using:

• Interceptions = 2 FP
• Fumble Recoveries by the Defense = 2 FP
• Sacks = 1 FP
• Safeties = 2 FP
• Interception and Fumble Recovery Returns TDs = 6 FP
• 0 Points Allowed (PA) = 12 FP
• 2-7 PA = 8 FP
• 8-12 PA = 4 FP
• 13-17 PA = 2 FP (note NFFC scoring for PA may change for 2019)

All statistics in this study come from profootballreference.com’s (PFR) invaluable Play Index.

And by alignments, I mean whether a defense uses a 4-3 or 3-4 as its base defense as categorized by PFR (see last week’s article for discussion on this topic).

There are two types of coaching changes that could affect fantasy scoring: head coaching (HC) changes and defensive coordinator (DC) changes. I’m only going to look at DC changes today, I’ll get into HC changes in another analysis.

Teams change DCs for two main reasons: the DC does a good job and gets hired as a HC elsewhere, or the DC does a bad job and gets fired. Coordinators retire or leave for other reasons, but those are the main two.

In the first several seasons after the league expanded to 32 teams in 2002, the number of teams changing defensive coordinators ranged from 7 to 15 and averaged around 9. Then in 2008, there was an explosion of turnover, with 22 franchises hiring new defensive honchos after that season. Naturally, 2009 was a down year for DC turnover, but starting in 2010, there have been double-digit DC changes each year, with an average of 12 per year, a significant increase over the 2002-2007 period.

Only about 15% of DCs who left their job in this period went on the HC jobs – after 2018, it was just Vic Fangio and Brian Flores:

Not surprisingly, the FP of the Team Defenses (Tm Ds) that saw their DC get promoted performed better:

Fantasy Defense Scoring of Teams Keeping or Losing Their DCs, 2003-2018

DC Got HC Job			DC Returns			DC Fired
FP	FP Next Year	PCT Change	FP	FP Next Year	PCT Change	FP	FP Next Year	PCT Change
145	133	-8%	132	128	-3%	115	123	7%

Those Tm Ds averaged 145 FP compared to 132 FP for DCs who kept their jobs (DC Returns, FP column). DCs who were fired or left for other reasons only average 115 FP (FP is of course an imperfect measure of actual NFL performance).

In the next season, Tm Ds that replaced a promoted coordinator still averaged more FP (133) than other teams, even though their own FP dropped about 8%. They were still better fantasy teams (indicating that their players were part of why they were good, it wasn’t just the coaching) but if you draft NE or CHI this year, you can probably expect a decrease in fantasy scoring.

Tm Ds that returned their DCs saw a slight slip in fantasy scoring, from 132 to 128 FP, a 2% fall. Since 127 FP was the average in this period, regression to the mean appears to affect these teams (and this effect was probably also a factor in why TM Ds with promoted DCs declined). They were good enough to keep their coordinator but not so good he got promoted.

As we’d expect, teams with the worst defenses (115 FP) fired their DCs (or they left for other but probably related reasons). Those defenses improved the next year to 123 FP(up 7%), but were still below average, indicating it wasn’t just the coach’s fault.

Left unmentioned in this table is the effect of personnel changes – naturally, teams with worse defenses probably draft or sign free agents to improved their defenses and better defenses are more likely to see their teams deal with problems on the offensive side. The effect of those decisions as well as the coaching moves are reflected in the outcomes shown in the table, even though only the coaching changes as described.

The next table lists the fantasy performance of each team in 2018, what happened to their DC after the season ended, and depending on that status, applies the appropriate percentage change to create an “Expected 2019 FP:”

Team Defense	2018 FP	Status of 2018 DC	Expected 2019 FP	Expected 2019 Tm D Rank
Arizona Cardinals	101	Old DC Fired	108	22
Atlanta Falcons	95	Old DC Fired	102	24
Baltimore Ravens	135	Old DC Fired	145	5
Buffalo Bills	112	Returning	109	21
Carolina Panthers	99	Returning	96	27
Chicago Bears	196	Old DC Got HC Job	180	1
Cincinnati Bengals	106	Old DC Fired	114	17
Cleveland Browns	111	Old DC Fired	119	14
Dallas Cowboys	101	Returning	98	25
Denver Broncos	124	Old DC Fired	133	6
Detroit Lions	109	Returning	106	23
Green Bay Packers	96	Returning	93	28
Houston Texans	151	Returning	147	4
Indianapolis Colts	128	Returning	124	11
Jacksonville Jaguars	119	Returning	116	16
Kansas City Chiefs	146	Old DC Fired	157	2
Los Angeles Chargers	114	Returning	111	19
Los Angeles Rams	155	Returning	151	3
Miami Dolphins	121	Old DC Fired	130	7
Minnesota Vikings	126	Returning	122	12
New England Patriots	136	Old DC Got HC Job	125	10
New Orleans Saints	123	Returning	120	13
New York Giants	94	Returning	91	30
New York Jets	103	Old DC Fired	110	20
Oakland Raiders	55	Returning	53	32
Philadelphia Eagles	100	Returning	97	26
Pittsburgh Steelers	116	Returning	113	18
San Francisco 49ers	71	Returning	69	31
Seattle Seahawks	131	Returning	127	8
Tampa Bay Bucs	86	Old DC Fired	92	29
Tennessee Titans	121	Returning	118	15
Washington Redskins	130	Returning	126	9

For example, the Cardinals Tm D scored 101 FP in 2018 and fired Al Holcomb, replacing him with Vance Joseph, Increasing their 2018 FP by 7%, the average for Tm Ds who fired their DC, means we’d “expect” 108 FP in 2019, which would be an “expected” rank of 22^nd.

This method doesn’t really change a lot of the Tm D rankings. For example, the Bears would still be the #1 defense for fantasy. But instead of beating the next best Tm D by 41 FP (about 2.5 FP/G), they’d be “only” 23 total FP better than the #2 Tm D in 2019. That’s a loss of about 1 FP/G in their advantage over everyone else.

Truth in analysis: in reality, the correlations (and R-squared values of regressions) between one year’s TM D FP and the next year’s are very low. This is obvious to anyone who has played fantasy football for very long: fantasy Tm Ds are very inconsistent from year-to-year. Even playing the averages as implied by the “Expected FP” above is pretty shaky.

What about teams that change DCs and change their defensive alignments?

Defense Alignment Changes of Teams Keeping or Losing Their DCs, 2003-2018

DC Got HC Job			DC Returns			DC Fired
#	Changed Alignment	PCT Change	#	Changed Alignment	PCT Change	#	Changed Alignment	PCT Change
28	3	11%	330	12	4%	154	32	21%

This table shows how often teams change or keep their DCs and what happens to their defensive alignments in those cases. For example, of the 28 teams from 2003-2018 that saw their DC get promoted after the previous season, only 3 (11%) changed their alignment. This is not surprising, those were generally successful defenses, so they didn’t have a lot of incentive to change.

And in just 4% of the 330 cases where the DC returned did teams change their base defense alignment. I suspect most of these are instances where it’s something of a judgment call what alignment those teams used.

Base defense alignment is most likely to change when a DC leaves for negative reasons. 21% of the time a DC is fired, the franchise also makes a decision to change how they play defense, at least in term of formation.

In fact, unless teams are changing DCs for negative reasons, they are very unlikely to also change alignment. Those sample s sizes (3 and 12 cases) are so small as to make any averages meaningless statistically. And the number of defenses who kept their base defense when a DC was promoted or retained was so relatively large that basically the average FP change is the same as the overall percentage three tables back.

The interesting case is when teams fire a DC (or he leaves for “other” reasons). Remember they typically see an increase of 7% in Tm D FP. But it turns out that when they keep the same alignment, the increase is only 5%. When teams both change DCs and schemes from 4-3 to 3-4 or vice versa, they see a 16% improvement on average in Tm D fantasy scoring.

Sadly, even the improvement in this narrow category is still not statistically significant. But for what it’s worth, these teams have both fired their old DC and brought in new coordinators who may be likely to change the base defense: Arizona Cardinals, Kansas City Chiefs, Miami Dolphins, New York Jets, and Tampa Bay Buccaneers.

It bears watching how these situations play out. For example, Brian Flores (HC not DC, but still) has said the Dolphins will be “multiple,” whatever that entails. So they may or may not be a candidate for a boost in fantasy scoring based on a scheme change. The Chiefs were actually a pretty good fantasy defense in 2018 even though they were bad on the field. Steve Spagnuolo is a long-time 4-3 guy replacing the 3-4 of Bob Sutton. Some Chiefs’ beat writers think their personnel might be better suited for a 4-3 anyhow, so this could be a strong place to look for good Tm D scoring. Arizona, the Jets, and Tampa were all poor defenses in 2018; even a 16% improvement would leave the Cards and Jets out of the Top 12 and the Bucs still in the lowest tier. In best ball formats where you’d draft more than one Tm D, you might look at the Cards and Jets as sleeper picks for your 2^nd or 3^rd D. I’d still avoid TB unless there are also dramatic personnel changes.