Seattle Seahawks Statistical Performance by Personnel

by John Morgan on Aug 25, 2009 12:32 PM PDT

This is a trial. It's fruitless to scrutinize personnel in the preseason, because teams use a limited playbook and the depth chart is very fluid. But when the regular season arrives, I will post full tabulations of Seattle's performance by personnel, formation and play call. As the season progresses, I should be able to go into greater and greater detail. This, on the contrary, is simple. It's only two offensive drives and two defensive drives. It separates by personnel instead of package. It's a start and a chance for the community to give ideas and criticism so I can do this better during the regular season.

Offense	Plays	P. Success %	ANY/A	TO %	Sack %	RY/A	R. Success %
4 WR	2	0	-3.5	0	50%	n/a	n/a
3WR	3	100%	21	0	0	n/a	n/a
2 WR	6	16%	3.7	0	0	2.7	0%
Defense	Plays	P. Success %	ANY/A	TO %	Sack %	RY/A	R. Success %
4-3	8	60%	8.4	0	0	2.7	67%
4-2	6	33%	14	0	0	5.3	33%
Blitz	2	100%	19.5	0	0	n/a	n/a

Tweet Comment 32 comments | 0 recs |

Do you like this story?

Comments

Is there a primer on the metrics used in the chart?

Similar to the Lookout Landings? Oh and CHARTS!!!!

Child please...

by Airborne Hawk Guy on Aug 25, 2009 12:40 PM PDT reply actions

Good point

The advanced stats are from the Hidden Game of Football. ANY/A is adjusted yards per attempt. Which is net yards (passing yards – sack yards) adjusted (+10 for TDs, -45 for Int) divided by attempts. I use 10 yards for touchdowns, because I think that’s newly accepted standard. I’ll check that.

Success is based on Carroll, Palmer and Thorn’s work on how many yards is need in any given situation for a play to be a success. It splits 40% on first down, 60% on second down and 100% on third and fourth down. There’s since been some adjustments (Football Outsiders uses 45% for first, but I think that’s counter intuitive (i.e. there’s no such thing as a 4.5 yard rush )), but I like this standard because it’s still very accurate and better agrees with scouting.

I think everything else is self explanatory, but let me know.

by John Morgan on Aug 25, 2009 12:49 PM PDT up reply actions

Awesome, thanks.

Child please...

by Airborne Hawk Guy on Aug 25, 2009 12:53 PM PDT up reply actions

Question on success?

If the first down is an incomplete (obviously not a success), the 2nd down needs to get 6 yards for a success?

If the first down is a gain of 4, then 2 more yards are needed for success? or is it 60% of 6 yards more, so 3.6 yards need to be a success?

And then, if it’s a sack for 10 yards, if second down is a pass for 12 yards, that’s a success?

by LantermanC on Aug 25, 2009 2:01 PM PDT up reply actions

Yes, No, Yes, Yes.

by John Morgan on Aug 25, 2009 2:03 PM PDT up reply actions

I've known about these critera before

but hadn’t before thought it through very much except to understand their value. I don’t see any problem with the 40/60/100 breakdown by down. And I’m sure it holds water when applied to actual stats, results, and their impact on winning games.

But I wonder if they could be improved by establishing a trichotomy rather than a boolean success/failure, if not more than three categories, and also by separating runs from passes and having separate scales for each.

Second down seems the most difficult to me to quantify success. A 1-yard gain on 2nd & 3 wouldn’t meet the criteria here to be a success, but a 3rd & 2 result doesn’t feel like a failure to me, but somewhere in between.

Also, that’s assuming the play was a run. A 1-yard gain on 2nd & 3 on a pass, a hitch or bubble screen, would feel like a complete lack of success; I’d have less of a problem calling that a failure than the run. Just some thoughts off the cuff.

by jacobstevens on Aug 25, 2009 3:24 PM PDT up reply actions

When you say success for the defense.

Does it imply the defense was successful? Or the offense against it was successful?

by LantermanC on Aug 25, 2009 1:43 PM PDT reply actions

Offense.

by John Morgan on Aug 25, 2009 1:54 PM PDT up reply actions

So low number is better.

by John Morgan on Aug 25, 2009 1:54 PM PDT up reply actions

It would seem, by a very quick, almost pointless analysis, that conservative is good.

Both when we go 4 wide and when we blitz, we get burned.

by djafrot on Aug 25, 2009 2:40 PM PDT up reply actions

This is awesome JM.

It will really help in nailing down just how often certain plays are called and how successful they are. Something that’s all too easy to try and eyeball and miss horribly on.

by Fear on Aug 25, 2009 2:08 PM PDT reply actions

This is cool.

I expect the numbers to change at least in the blitz cat., since I suspect we are using blitzes and blitz packages now that will be more elaborate and creative during the reg. season. I’d hope our team is not allowing opponents to scout us now, but rather we’re holding back a good portion of what we’ll really bring come week one.

Early prospect watch: RB C.J. Spiller, QB Jevon Snead, OT Ciron Black, DT Gerald McCoy, S Eric Berry, DT Ndamukong Suh, CB Ras-I Dowling 6'2, 200, RB Jonathan Dwyer

by Misfit74 on Aug 25, 2009 2:36 PM PDT reply actions

what qualifies as a blitz?

I imagine it’s at least one non linemen being sent in to pressure the QB?

by Hancock.Brett on Aug 25, 2009 2:38 PM PDT reply actions

Five or more

or else a 3-4 would blitz on virtually every down. The level of complexity in modern defenses makes the simple “non-lineman rushing the passer” definition obsolete.

by John Morgan on Aug 25, 2009 3:32 PM PDT up reply actions

What about a zone "blitz"

when only one linebacker rushes and a DE drops back in coverage? Technically it’s considered a blitz, but there are only four rushers, so does it not count?

by B.B.Finnegan on Aug 25, 2009 5:15 PM PDT up reply actions

Most fire zones

Which is very close to what you describe (DE in cover/linebackers rush) are five-man blitzes with the linebackers overloading one side.

by John Morgan on Aug 25, 2009 5:49 PM PDT up reply actions

To better answer your question

I will try to mark zone blitzes, but I don’t want to bite off more than I can chew. I have a problem with that.

by John Morgan on Aug 25, 2009 5:53 PM PDT up reply actions

Yeah, I was thinking more the standard zone blitz, which Seattle may use

But I guess in essence it’s somewhat similar to the standard 3-4, where one linebacker rushes, you just trade a DE for a LB.

To be fair, the over five rusher rule would definitively be the easiest way to tally, especially if you have to take in account 3-4 defenses. I’m not sure how often Seattle will use the standard zone blitz, if at all, so it’s probably not worth it unless it’s used often. I was just thinking it might yield some skewed results if they do and you discount them simply because there were only four rushers. To conserve your time, it would probably be best to do it with just 5+, and then if a more detail look at the blitzing looks to be worth it, it’s something that could be done all on its own. Not that I’m trying to give you anything to do, you do a hell of a job as it is.

by B.B.Finnegan on Aug 25, 2009 7:28 PM PDT up reply actions

Is this Sando inspired?

You should compare notes.

by Groundhog on Aug 25, 2009 2:41 PM PDT reply actions

Yep.

I’ve always liked the charts he used and never had the time to do them like I wanted.

by John Morgan on Aug 25, 2009 3:31 PM PDT up reply actions

This will be per game as well as cumulative?

by Groundhog on Aug 25, 2009 4:02 PM PDT up reply actions

That is the goal.

And instead of using “3 WRs” I’ll use official formations.

by John Morgan on Aug 25, 2009 4:44 PM PDT up reply actions

Great idea.

For some strange reason, though, I thought I remembered there being a two TE set on offense on their second drive. I must be mistaken.

Sam Bradford, future Seattle Seahawk.

by Carl Shinyama on Aug 25, 2009 8:06 PM PDT reply actions

The first play of the second drive had two tight ends

It also had two wide receivers.

by John Morgan on Aug 25, 2009 8:57 PM PDT up reply actions

That'll explain it.

Sam Bradford, future Seattle Seahawk.

by Carl Shinyama on Aug 25, 2009 9:52 PM PDT up reply actions

how to chart

I’ve charted about 6,000 plays charted since 2006 and can tell you the key to doing this is figuring out quickly what NOT to chart. Time is of the essence when you’re trying to chart accurately in real time. Whatever you chart, you need to be able to do it as you watch the game live. That way, you are only making corrections on the re-watch. Fixing mistakes is the hardest part of the process. You will get the down or distance wrong sometimes. It’s not because you’re careless. It just happens when you try to track 1200 plays in a season (and that is just the offense). You’ll see a spread look at write down 1TE/3WR even though you mistook the second tight end for a WR based on where he was lined up. Best of luck with it and I’d be happy to compare notes over time. I didn’t chart Seattle last season because there was no point after a while. But I’ll try to do it this season.

by Mike Sando on Aug 25, 2009 11:36 PM PDT reply actions

omghaiSando!

The demise of the Broncos in '09 is our future. Pray hard.

by Nick Andron on Aug 26, 2009 8:08 AM PDT up reply actions

another thought

I could chart one thing if you wanted to chart other things. I’ve become pretty good at charting personnel groups (TE/RB/WR). We could match up the information and have something more powerful. Would be nice to chart number of pass rushers, QB drop depth (3/5/7), play-action, etc.

by Mike Sando on Aug 25, 2009 11:39 PM PDT reply actions