San Francisco and Losing Defensive Players
by Aaron Schatz
Last year, the San Francisco 49ers fell to 8-8 after three straight seasons where they went 11-4-1 or better. Conventional wisdom says the 49ers are going to get even worse this year, and this is one place where Football Outsiders is in agreement with the conventional wisdom. In the new Football Outsiders Almanac 2015, our season simulation has the 49ers averaging just 6.8 wins and making the playoffs just 20 percent of the time.
The 49ers were one of five teams listed in this ESPN Insider piece I wrote yesterday about teams we believe are likely to decline in 2015. One of the elements I mentioned in the ESPN piece was a new variable that we're using in this year's improved team projection system. We used Pro Football Reference's Approximate Value stat to get the value of defensive players leaving and joining each team going back to 2003. We analyzed value over replacement, which we found worked best as each player's AV over 3. Obviously, there are ways to improve this metric by looking at more than just every player's AV from the previous season, and we'll be playing with those ideas in the future. But this is what we used for now, and it definitely improved our projection system by catching teams where major personnel overhaul helped teams to play better or worse than would be expected by simple regression towards the mean.
The New York Jets chapter of FOA 2015 notes that the Jets are tied with the 2009 Lions for the biggest positive change in this metric of any team going back to 2003. It has a table showing the teams with the biggest positive values, with a list of which players were involved. (Washington's 2015 offseason is also on this list, a bit of a surprise.)
I meant to put a similar table in the San Francisco chapter, and then forgot. (Part of the reason is that we finalized the four NFC West chapters first so that Vince Verhei, who wrote those teams, could concentrate on book layout from that point on.) So I wanted to share the table here for everyone to see one of the reasons we have the San Francisco defense likely to get worse in 2015.
The teams are listed here with defensive DVOA and rank, both before and after losing the offseason listed. We're also listing the players being counted with their AV over replacement. While these teams did not tend to decline quite as much as the teams which added talent tended to improve, they did decline more than we would expect from simple regression towards the mean.
|Highest Net AV over Replacement Loss of Defensive Talent, 2003-2015|
|Players Added||Players Lost|
|TB||2009||-32||-10.7%||6||8.0%||25||+18.6%||NONE||D.Brooks (8), G.Adams (6), P.Buchanon (5), K.Carter (5), J.Haye (4), C.June (4)|
|NYJ||2013||-24||-4.2%||9||-5.6%||12||-1.3%||D.Landry (3)||B.Scott (7), L.Landry (5), Y.Bell (4), S.Pouha (4), B.Thomas (4), M.Devito (3)|
|CAR||2010||-23||-12.8%||6||-1.1%||16||+11.7%||NONE||J.Peppers (12), D.Lewis (5), C.Harris (4), N.Diggs (2)|
|PHI||2003||-20||-11.2%||4||3.0%||17||+14.2%||M.Coleman (4), N.Wayne (6)||H.Douglas (14), S.Barber (8), L.Kirkland (6), B.Bishop (2)|
|SF||2015||-20||-10.1%||5||--||--||--||S.Wright (2)||J.Smith (5), P.Cox (4), C.Culliver (4), R.McDonald (4), C.Borland (3), D.Skuta (2)|
|CAR||2006||-18||-14.2%||2||-10.9%||4||+3.3%||M.Kemoeatu (5), R.Howard (1)||W.Witherspoon (7), B.Buckner (6), M.McCree (5), B.Short (5), R.Manning (1)|
|SD||2013||-17||2.0%||18||17.5%||32||+15.5%||D.Freeney (3), D.Cox (2)||Q.Jammer (5), S.Phillips (5), T.Spikes (4), A.Cason (3), V.Martin (2), A.Bigby (1), A.Franklin (1), D.Williams (1)|
|CLE||2011||-17||1.7%||18||4.2%||22||+2.6%||D.Patterson (1)||K.Coleman (5), A.Elam (4), M.Roth (4), E.Barton (3), D.Bowens (1), E.Wright (1)|
|CLE||2003||-17||-5.1%||10||-1.9%||14||+3.2%||NONE||E.Holmes (5), D.Hambrick (4), D.Rudd (4), D.Bush (2), C.Fuller (2)|
|HOU||2013||-16||-14.2%||4||2.5%||18||+16.7%||NONE||C.Barwin (5), B.James (5), G.Quin (4), S.Cody (2)|
|SEA||2011||-16||12.0%||29||-7.1%||10||-19.1%||NONE||L.Tatupu (4), A.Curry (3), L.Milloy (3), K.Balmer (2), C.Cole (2), K.Jennings (2)|
|NE||2009||-16||3.6%||17||-1.1%||14||-4.7%||L.Bodden (2), D.Burgess (1), S.Springs (1)||E.Hobbs (6), R.Seymour (5), M.Vrabel (4), T.Bruschi (3), D.O'Neal (2)|
|NYJ||2005||-16||-3.7%||14||1.0%||18||+4.7%||L.Legree (1)||D.Abraham (7), J.Ferguson (5), R.Tongue (5)|
|KC||2008||-15||-3.8%||14||13.3%||28||+17.1%||D.Williams (4), R.Boiman (2)||J.Allen (13), N.Harris (4), T.Law (4)|
|SD||2003||-15||2.4%||22||12.0%||30||+9.6%||K.Lassiter (2)||J.Seau (5), R.McNeil (4), A.Molden (4), R.Harrison (2), R.Beckett (1), L.Carson (1)|
|STL||2003||-15||-4.5%||11||-11.4%||6||-6.9%||J.Sehorn (1)||D.Bly (6), J.Zgonina (5), K.Herring (4), D.Davis (1)|
|AVERAGE (without SF)||-4.2%||12.3||1.5%||17.7||+5.7%|
A few further notes about this table.
1) Do you notice which names are missing? Because replacement value is 3 AV, neither Patrick Willis nor Aldon Smith counts as "lost talent" for the 49ers. Both players only had 3 AV because of playing partial seasons in 2014. The 49ers come out with one of the strongest net losses in defensive talent even without counting arguably the two most talented players they are losing. If we do a future version of this metric where we consider "expected AV" of players rather than just "last year's AV," the 2015 49ers might rank even lower.
2) On the other hand, the metric we used in building our projection system did not include the return of injured players. That's something we'll need to work on for next year's projection system. Nonetheless, I decided to try to account for these players anyway while doing 2015 projections. So, for example, I gave Dallas some credit based on Greg Hardy and Sean Lee's AV values from 2013. I gave Philadelphia some credit based on Kiko Alonso's AV value from 2013. And... I gave San Francisco some credit based on NaVorro Bowman and Darnell Dockett's AV values from 2013. If I had not done that, our San Francisco projection would have actually come out even worse.
3) Once I added in those values for players who were injured in 2014, the team that actually ended up last in this metric was New England at minus-12. However, the San Francisco defense is projected to decline more than the New England defense because of a number of other variables. The 49ers have a new head coach/defensive coordinator combination, which tends to lead to defensive decline. They also ranked second in the league by ending 17.1 percent of opponent drives with takeaways, a stat that tends to heavily regress towards the mean from year to year. This is another place where the 49ers and Jets represent opposites in 2015, as the Jets were dead last in that stat and thus are likely to have more takeaways this season.