Are High Tackle Numbers for Safeties a Bad Thing?

[GTripp0012]

It's a pretty common analysis refrain: if the safety is making a lot of tackles, your defense probably isn't performing well. Is this really the case, or are high tackle numbers more indicative of a player with a lot of range, which is a mark of a good player?

I looked at the 2007 numbers of all safeties and how many tackles they registered per game. To qualify for this study, a player had to appear in at least twelve games, and average 5.5 tackles per game or more. Here is the complete list of the qualifiers:

Gibril Wilson (NYG)-7.1
Michael Lewis (SF)-6.5
Chris Harris (Car)-6.4
Bob Sanders (Ind)-6.4
Roy Williams (Dal)-6.1
Sean Jones (Cle)-6.0
Laron Landry (Wash)-5.9
Donte Whitner (Buf)-5.9
Von Hutchins (Hou)-5.8
Sammy Knight (Jax)-5.8
Madieu Williams (Cin)-5.7
Roman Harper (NO)-5.6
Lawyer Milloy (Atl)-5.6
Bernard Pollard (KC)-5.6
Jermaine Phillips (TB)-5.5
Kinnoy Kennedy (Det)-5.5
Josh Bullocks (NO)-5.5

A few interesting things stand out there. Though exactly half the league has one safety on this list, only New Orleans has more than one, and Josh Bullocks barely made the cut. What is clear here is that most teams do tend to have a strong safety that gets more tackles than the other, even if his designation to the team is "free" safety. Only St. Louis' safety tandem--Corey Chavous and O.J. Atogwe--finished with the same number of tackles (75), and neither made this list.

Of those 16 teams, here are their Defensive DVOA rankings (with Rush Def DVOA in parenthesis) to see how they stack up against teams who did not have their safeties making as many tackles:

3rd (8th)
4th (7th)
6th (11th)-Washington Redskins
10th (14th)
12th (22nd)
14th (10th)-New York Giants
16th (24th)
17th (16th)
19th (17th)
22nd (21st)
24th (13th)-Cincinnati Bengals
27th (9th)-New Orleans Saints
28th (23rd)-San Francisco 49ers
29th (28th)
30th (29th)
31st (27th)-Detroit Lions

Just looking at those ranks, 5 out of the 6 worst defenses in 2007 had at least one safety make more than 5.5 tackles in a single game. The teams with safeties on that list did produce a statistically significant amount of DVOA worse than those without a team on that list.

The rush defenses were even worse for the teams who had a lot of safeties making tackles. The Bengals and the Saints totally bucked this trend, but on the whole, a team who had a safety make a lot of tackles was worse against the run than the pass.

I went back a few years to see if this trend held:

2006
Chris Hope (Ten)-7.6
Sean Jones (Cle)-6.9
Sean Taylor (Wash)-6.9
Donte Whitner (Buf)-6.9
Gibril Wilson (NYG)-6.7
Stuart Schweigert (Oak)-6.7
Antwan Bethea (Ind)-6.4
Erik Coleman (NYJ)-6.3
Will Demps (NYG)-6.3
Lawyer Milloy (Atl)-6.1
Kerry Rhodes (NYJ)-6.1
Ken Hamlin (Sea)-6.0

9th (16th)
13th (11th)-New York Giants
15th (29th)
17th (17th)
20th (23rd)
21st (25th)
24th (28th)
26th (32nd)
27th (31st)
32nd (16th)-Washington Redskins

2005
Erik Coleman (NYJ)-7.6
Gibril Wilson (NYG)-7.0
Adrian Wilson (Ari)-6.8
Michael Lewis (Phi)-6.7
Lawyer Milloy (Buf)-6.6
Kerry Rhodes (NYJ)-6.5
Bob Sanders (Ind)-6.5
Kenoy Kennedy (Det)-6.1
Chris Hope (Pit)-6.0

2nd (1st)-SB Champion Pittsburgh Steelers
8th (17th)
11th (2nd)-New York Giants
14th (7th)-Philadelphia Eagles
17th (20th)
18th (28th)
21st (23rd)
26th (31st)

I bolded the teams who had a better run defense than a total defense. A few things stand out:

The NFC East seems to be the exception to the rule here. The Giants are consistently above average in defense, and Gibril Wilson average more than 6 tackles a game every season. The Eagles were above average the year Michael Lewis made the list, the Cowboys were above average when Roy Williams made it, and the Redskins were above it when Landry made it. The only exception is when Sean Taylor was on the tackles list, the Redskins were 32nd in total D, but even still, the Run D was pretty decent that year.

Outside of the NFC East, the results are pretty conclusive. If your safeties make a bunch of tackles, your defense is going to be pretty bad, and your run defense is likely to be downright awful. Conventional logic holds here. If the safety is making 6 tackles a game, with a few key exceptions, the team is playing losing football.

[EARTHQUAKE2689]

I think there is a fine line when talking about tackles for safties. Because it boils down to three things in my opinion: The range of the player in question, the penetration of the defensive lineman, and the particular style of the safety. Certain players play closer to the line such as Roy Williams, or a Gibril Wilson and therefore their tackle count will be higher without necesarrily hurting the defense or the defense playing well. But on the flip side for a safety such as Ed Reed, or Bob Sanders who garner alot of tackles but their defenses reamin solid and they dont play as close to the line as Roy Williams does on a consistent basis then that shows how well those particular guys read and react to a developing play. But as was shown in 2006 with Sean Taylor leading our team in tackles, that was mainly to the fact that no one on the team could make a tackle and the defensive line wasnt getting any pressure up front so I feel that it is a combination of all three of those things.


Good Topic Tripp

[GTripp0012]

Originally Posted by EARTHQUAKE2689

I think there is a fine line when talking about tackles for safties. Because it boils down to three things in my opinion: The range of the player in question, the penetration of the defensive lineman, and the particular style of the safety. Certain players play closer to the line such as Roy Williams, or a Gibril Wilson and therefore their tackle count will be higher without necesarrily hurting the defense or the defense playing well. But on the flip side for a safety such as Ed Reed, or Bob Sanders who garner alot of tackles but their defenses reamin solid and they dont play as close to the line as Roy Williams does on a consistent basis then that shows how well those particular guys read and react to a developing play. But as was shown in 2006 with Sean Taylor leading our team in tackles, that was mainly to the fact that no one on the team could make a tackle and the defensive line wasnt getting any pressure up front so I feel that it is a combination of all three of those things.


Good Topic Tripp

I was a bit shocked because I thought the safeties that played closer to the line and registered more tackles would result in worse pass D, like ST in 2006. In reality, these teams were bad on defense, but worse against the run.

You learn something new everyday. Since it's after midnight, I guess I don't have to go to class tomorrow.

[flashalexb]

Well, didn't Sean Taylor lead the league in missed tackels in 2006. Most likely because he was always going for the big hit, but it shows that he had the range to play all over the field. This was even more so when he had some assistance from the SS position i.e. "Suicide Mission" Landry.

[That Guy]

is it the offseason?

i don't think there's really enough there to determine causation and i don't think one stat that can be attained in many vague ways (and is calculated differently by different scorers) can be universally used as a single point of failure.

ie, if a ball passes 20 yards over a WR's head in philly, the scorer there counts it as a pass defended, but that's not standard practice anywhere else.

is this X tackles pure game behind the line? after catches? on broken plays where the LBs over-pursue? some guys are faster and have a better nose for the ball, so they get more tackles cause they're actually good... other time the front seven gets blown up and they're just the clean up crew.

judging just on tackles alone without sorting them out is fairly meaningless, since you've got the 3rd best D as well as the 31st best, that pretty much screams nevermind to me.

If the safety is making 6 tackles a game, with a few key exceptions, the team is playing losing football.

except that half the teams are in the top 16 in defense, as would be expected in a RANDOM sampling in two of the three years (3/6, 3/10, 4/8), which means your conclusion is wrong twice as often as it's right.

[dmek25]

too me, a high number of tackles will mean one of two things. either your defense isn't very good. or your scheme has the safety around the line of scrimmage. the skins are a good point of argument for the safety hanging around the line. Landry makes tons of tackles. and then towards the end of the season, when Doughty took over, he was making alot of tackles

[GTripp0012]

Originally Posted by That Guy

except that half the teams are in the top 16 in defense, as would be expected in a RANDOM sampling in two of the three years (3/6, 3/10, 4/8), which means your conclusion is wrong twice as often as it's right.

Except that was close to true in 2005 only, not even remotely close in 06 or 07. In 2005, it appeared to not be a huge factor for total defense, but in 2006 and 2007, it was a statistically significant trend

[QBall]

Landry started at SS and then moved to Free.

Think SS is supposed to have more tackles than Free. Doughty should have more tackles than Landry if he's SS.

I liked how Landry was used towards the end, more for his ball hawking skills just like Taylor. The big hits are nice but awesome pass defense from a FS with his range is even better.

[BeastsoftheNFCeast]

Originally Posted by That Guy

is it the offseason?

i don't think there's really enough there to determine causation and i don't think one stat that can be attained in many vague ways (and is calculated differently by different scorers) can be universally used as a single point of failure.

ie, if a ball passes 20 yards over a WR's head in philly, the scorer there counts it as a pass defended, but that's not standard practice anywhere else.

is this X tackles pure game behind the line? after catches? on broken plays where the LBs over-pursue? some guys are faster and have a better nose for the ball, so they get more tackles cause they're actually good... other time the front seven gets blown up and they're just the clean up crew.

judging just on tackles alone without sorting them out is fairly meaningless, since you've got the 3rd best D as well as the 31st best, that pretty much screams nevermind to me.



except that half the teams are in the top 16 in defense, as would be expected in a RANDOM sampling in two of the three years (3/6, 3/10, 4/8), which means your conclusion is wrong twice as often as it's right.

You are right in saying that it is not a causation, but there is a definite correlation. I dont think there is any single stat that is a causation, it is a collection of the correlations that helps build a winning team, and even that is not always true.

[SouperMeister]

For a strong safety deployed in the box on running downs, you would expect high tackle totals if he's a good player. I never want to see the free safety among the leaders in tackles as Sean Taylor was in 2006. That was the biggest sign that our defense was broken that season.

[Skins4Life101]

It depends on the production from the D-Line. If those tackle numbers comes with 4+ sacks from the front 4 than these stats would not mean TOO much. But if some 2nd string running back racked up 140+ yards on us...

Than this means we have problems.

This could also show that the player in particular may be HIGHLY aggressive and a ball hawk. So that could reflect positives as well. For instace, if Sean Taylor (RIP) were to average 10 tackles a game. We would all know that there were at least 4 hitstick moments..

[Big C]

ive always figured safeties with huge tackle numbers means that the linebackers and dline arent making plays, and they are tackling after a 10 yard gain a lot. look at archuleta with us, he had tons of tackles when he started but most were way downfield after the guy caught it on him

[hooskins]

Another interesting point to follow this argument is if a S has alot of tackles should that be viewed positively for Probowl selection?

A similar method is used for kickers, and usual the "best" kickers are those on teams that kick a lot, which may not be a good thing.

[DC52]

A S making a lot of tackles isn't necessarily a bad thing. It all depends on how the safety is used. If he is brought down into the box constantly because the opposing team runs a lot than he should have a lot of takles or if the D Coordinator like stunts like bringing Safety blitzes. The bad time may be when you safety is told to play deep coverage and is still making a load of tackles. That either means the Safety isn't following the play or you defense is getting owned.

[That Guy]

Originally Posted by GTripp0012

Except that was close to true in 2005 only, not even remotely close in 06 or 07. In 2005, it appeared to not be a huge factor for total defense, but in 2006 and 2007, it was a statistically significant trend

what are you talking about? 6.0+ tackles were 3/6 and 4/8 teams in the top 16 which is 50%, which is a random sampling, and not statistically significant in any way at all.

that means your own conclusion only covers one of the three years you've included in your first post.