It's not predictability, it's design and talent

The Essential Online Seattle Football Fan Forum Community. There simply is NO substitute. LANGUAGE RATING: PG-13
  • https://www.fieldgulls.com/2019/6/7/186 ... n-3rd-down

    So while I read over the article and found some of it interesting, the thing that got me was the point that the Hawks predictability wasn't quite that abnormal compared to other teams situationally. Especially on third downs between 5-10 yards. Most teams are passing 85%+ of the time there. So at the very base, the idea that we are predictable on 3rd down is kind of a canard.

    BUT...BUT...If we do a little sleuthing via Pro Football reference we can see that the outcomes of what we did are different to average in composition. The conversion rate to 1st down? Almost completely average relative to the field. Likewise on distance to down. Slightly above average in rank (Good Thing). Total of 3rd down plays, also slightly above average in rank (Good thing).

    But here is where we are divergent.

    Sacks on 3rd down. Tied for 2nd in the league @26 with Oakland and only behind GB @28. They did have the highest sack rate on 3rd downs given the lower amount of 3rd down passing plays.
    TDs. Led the league @ 16. 2nd was New England @12.
    Turnovers. 5th in the league @ 6.5% behind Buffalo, Arizona, LA Rams, and Pittsburgh.

    So how can we make sense of some of this?

    1. Perhaps, there is a design element that elevates sacks and TDs and really puts a boom bust stamp on how 3rd downs are approached.

    2. Perhaps, there is a talent gap on the field where the execution of 3rd downs results in these divergent outcomes. RW on down the line.

    3. Perhaps, that's just the way it is with the Hawks on 3rd down and nothing to think about. We enjoy the spoils and spoilage of the approach.

    Just kind of interesting to me how the Hawks aren't different in some ways, but quite different in others. I am particularly intrigued by the similarity in turnovers to the Rams and Steelers who are our closest 'like' peer in that regard, but the divergence between sacks and TDs for both relative to us and one another.

    Thoughts?
    mrt144
    NET Veteran
     
    Posts: 4065
    Joined: Thu Dec 30, 2010 2:21 pm


  • Great post, interesting stuff. I look forward to reviewing it further. :2thumbs:
    ivotuk
    * NET Staff Alumni *
     
    Posts: 19156
    Joined: Sat Mar 03, 2007 8:29 pm
    Location: North Pole, Alaska


  • mrt144 wrote:https://www.fieldgulls.com/2019/6/7/18656268/how-predictable-seattle-seahawks-offense-1st-down-2018-2-minute-4th-quarter-russell-wilson-3rd-down

    So while I read over the article and found some of it interesting, the thing that got me was the point that the Hawks predictability wasn't quite that abnormal compared to other teams situationally. Especially on third downs between 5-10 yards. Most teams are passing 85%+ of the time there. So at the very base, the idea that we are predictable on 3rd down is kind of a canard.

    BUT...BUT...If we do a little sleuthing via Pro Football reference we can see that the outcomes of what we did are different to average in composition. The conversion rate to 1st down? Almost completely average relative to the field. Likewise on distance to down. Slightly above average in rank (Good Thing). Total of 3rd down plays, also slightly above average in rank (Good thing).

    But here is where we are divergent.

    Sacks on 3rd down. Tied for 2nd in the league @26 with Oakland and only behind GB @28. They did have the highest sack rate on 3rd downs given the lower amount of 3rd down passing plays.
    TDs. Led the league @ 16. 2nd was New England @12.
    Turnovers. 5th in the league @ 6.5% behind Buffalo, Arizona, LA Rams, and Pittsburgh.

    So how can we make sense of some of this?

    1. Perhaps, there is a design element that elevates sacks and TDs and really puts a boom bust stamp on how 3rd downs are approached.

    2. Perhaps, there is a talent gap on the field where the execution of 3rd downs results in these divergent outcomes. RW on down the line.

    3. Perhaps, that's just the way it is with the Hawks on 3rd down and nothing to think about. We enjoy the spoils and spoilage of the approach.

    Just kind of interesting to me how the Hawks aren't different in some ways, but quite different in others. I am particularly intrigued by the similarity in turnovers to the Rams and Steelers who are our closest 'like' peer in that regard, but the divergence between sacks and TDs for both relative to us and one another.

    Thoughts?


    Okay except the place were we are different which would impact 3rd down, is we go run, run, pass on 75% of our drives. That is a huge difference. We are predictable because everyone knows we will do run, run pass, most of the time and that hurts us on 3rd down.
    John63
    NET Veteran
     
    Posts: 1565
    Joined: Wed Dec 19, 2018 2:56 pm


  • How about pass-pass-punt....that will fool them ... LOL
    xray
    NET Veteran
     
    Posts: 1292
    Joined: Sun Aug 19, 2018 6:29 am
    Location: AZ


  • Run run pass has given the Hawks much more manageable 3rd down yardage vs the previous 2 years of pass heavy offense that regularly put us in 3rd and very long. The OL was never as bad as advertised. RW is an extremely conservative QB. He'll eat it rather than take risks the vast majority of the time and I'm sure Pete wants it that way. Ain't no gun-slinging up in here.
    vin.couve12
    .NET Poster of the Month
     
    Posts: 4993
    Joined: Fri May 27, 2011 11:19 am
    Location: Vancouver, WA


  • John63 wrote:
    mrt144 wrote:https://www.fieldgulls.com/2019/6/7/18656268/how-predictable-seattle-seahawks-offense-1st-down-2018-2-minute-4th-quarter-russell-wilson-3rd-down

    So while I read over the article and found some of it interesting, the thing that got me was the point that the Hawks predictability wasn't quite that abnormal compared to other teams situationally. Especially on third downs between 5-10 yards. Most teams are passing 85%+ of the time there. So at the very base, the idea that we are predictable on 3rd down is kind of a canard.

    BUT...BUT...If we do a little sleuthing via Pro Football reference we can see that the outcomes of what we did are different to average in composition. The conversion rate to 1st down? Almost completely average relative to the field. Likewise on distance to down. Slightly above average in rank (Good Thing). Total of 3rd down plays, also slightly above average in rank (Good thing).

    But here is where we are divergent.

    Sacks on 3rd down. Tied for 2nd in the league @26 with Oakland and only behind GB @28. They did have the highest sack rate on 3rd downs given the lower amount of 3rd down passing plays.
    TDs. Led the league @ 16. 2nd was New England @12.
    Turnovers. 5th in the league @ 6.5% behind Buffalo, Arizona, LA Rams, and Pittsburgh.

    So how can we make sense of some of this?

    1. Perhaps, there is a design element that elevates sacks and TDs and really puts a boom bust stamp on how 3rd downs are approached.

    2. Perhaps, there is a talent gap on the field where the execution of 3rd downs results in these divergent outcomes. RW on down the line.

    3. Perhaps, that's just the way it is with the Hawks on 3rd down and nothing to think about. We enjoy the spoils and spoilage of the approach.

    Just kind of interesting to me how the Hawks aren't different in some ways, but quite different in others. I am particularly intrigued by the similarity in turnovers to the Rams and Steelers who are our closest 'like' peer in that regard, but the divergence between sacks and TDs for both relative to us and one another.

    Thoughts?


    Okay except the place were we are different which would impact 3rd down, is we go run, run, pass on 75% of our drives. That is a huge difference. We are predictable because everyone knows we will do run, run pass, most of the time and that hurts us on 3rd down.


    Predictability doesn't exactly explain why the output on 3rd down was much more boom/bust than other teams. I mean, that is a tendency of the offense as it existed in 2018 but how does the defense exactly play boom/bust? I don't see it, really.

    Now the setup for 3rd downs, sure run-run-pass is the mode we operated in much of the time but I thought this was an interesting wrinkle to run my fingers over.
    mrt144
    NET Veteran
     
    Posts: 4065
    Joined: Thu Dec 30, 2010 2:21 pm


  • In my view our third down sack rate is partly due to our head coach's philosophy (20%) and mostly due to our unique QB (80%). That split is apportioned mostly towards Russ because Pete does occasionally talk about needing to get the ball out more quickly. If you remember after the Bears loss, Pete had a whole series of interviews where he talked about Russ trying to do a little too much and suggesting that they would be working on throwing the ball away more often in future games.

    On the whole, however, I think Pete is fine with Russ taking the occasional sack so long as he holds onto the football. Pete really hates turnovers and putting his defense on a short field. A sack is a vastly better outcome than an interception in his book and Pete doesn't mind punting with good field position. Pete recognizes that telling Russ to throw it away after 2.5 seconds each play would limit our upside.

    Regarding Russ, his sack rate is a negative when looked at by itself. However, I think most of us see it as a negative symptom of his positive ability to extend plays. In 2018 he was #3 out of #39 QBs in time to throw (3.01 seconds) which led to many of his sacks. It's also probably a big reason why he was #1 in PFF's "big time throw" metric which supposedly charts throws into tight windows.
    AgentDib
    NET Veteran
     
    Posts: 3977
    Joined: Fri Oct 01, 2010 10:08 pm
    Location: Seattle


  • The boom or bust on 3rd down makes sense to me. Russ hangs on to the ball way too long and either gets sacked, forces it, or the defense breaks down.
    Own The West
    NET Rookie
     
    Posts: 276
    Joined: Wed Jan 13, 2016 8:20 pm


  • Own The West wrote:The boom or bust on 3rd down makes sense to me. Russ hangs on to the ball way too long and either gets sacked, forces it, or the defense breaks down.

    Forcing it is not really the norm for him. It happens once in a while, but he'll eat it or throw it away 90 to 95% of the time if the defense doesn't break down. Used to be more throw aways, but now it's more tuck and get down. He's not as quick and agile as he was.

    I'm not faulting him for not forcing throws, btw. The TD to INT ratio is a direct derivative of that.
    vin.couve12
    .NET Poster of the Month
     
    Posts: 4993
    Joined: Fri May 27, 2011 11:19 am
    Location: Vancouver, WA


  • xray wrote:How about pass-pass-punt....that will fool them ... LOL



    Punting on 3rd down. Brilliant!!! This is a game changer.
    niveky
    NET Veteran
     
    Posts: 685
    Joined: Fri May 01, 2009 10:58 pm


  • AgentDib wrote:On the whole, however, I think Pete is fine with Russ taking the occasional sack so long as he holds onto the football. Pete really hates turnovers and putting his defense on a short field. A sack is a vastly better outcome than an interception in his book and Pete doesn't mind punting with good field position. Pete recognizes that telling Russ to throw it away after 2.5 seconds each play would limit our upside.

    Regarding Russ, his sack rate is a negative when looked at by itself. However, I think most of us see it as a negative symptom of his positive ability to extend plays. In 2018 he was #3 out of #39 QBs in time to throw (3.01 seconds) which led to many of his sacks. It's also probably a big reason why he was #1 in PFF's "big time throw" metric which supposedly charts throws into tight windows.


    I would say this.

    Realistically, what's the difference between taking a sack on 3rd down and punting, versus throwing the ball essentially away (out of bounds or too early and safely in a spot nobody gets to it). The result is in essence the same. And if I were a coach, that's exactly how I'd prefer to do it.

    The ultimate barometer (ability to convert 3rd downs) is largely unaffected or rather it's mostly average to above average. But by taking sacks instead of directing panic throws -- it reduces the chances of interceptions (batted passes, errant throw aways or poorly timed throws that DBs can make better plays on the ball). It's a sound tradeoff and I have always felt that the bloated sack values for our offense have been largely a benign result of our philosophy and situational design.

    If I were to change something, it would be to focus more on short outlet passing (flats/crossing the face of LB zones). Those patterns are generally safe chain movers in the context of "3rd down and manageable". Not utilizing those high percentage throws for free yardage kind of eliminates the value of getting to manageable distance in the first place. We either don't or can't implement those gift patterns. My guess is that we don't have the talent to make those plays. Procise was supposed to be that at the RB position. McKissic was actually pretty good at that. But in general we don't seem to be at our best in the short middle. Nor have we been successful at screen passes really since Holmgren left.
    Attyla the Hawk
    NET Veteran
     
    Posts: 2311
    Joined: Mon Mar 11, 2013 3:38 pm


  • John63 wrote:
    mrt144 wrote:https://www.fieldgulls.com/2019/6/7/18656268/how-predictable-seattle-seahawks-offense-1st-down-2018-2-minute-4th-quarter-russell-wilson-3rd-down

    So while I read over the article and found some of it interesting, the thing that got me was the point that the Hawks predictability wasn't quite that abnormal compared to other teams situationally. Especially on third downs between 5-10 yards. Most teams are passing 85%+ of the time there. So at the very base, the idea that we are predictable on 3rd down is kind of a canard.

    BUT...BUT...If we do a little sleuthing via Pro Football reference we can see that the outcomes of what we did are different to average in composition. The conversion rate to 1st down? Almost completely average relative to the field. Likewise on distance to down. Slightly above average in rank (Good Thing). Total of 3rd down plays, also slightly above average in rank (Good thing).

    But here is where we are divergent.

    Sacks on 3rd down. Tied for 2nd in the league @26 with Oakland and only behind GB @28. They did have the highest sack rate on 3rd downs given the lower amount of 3rd down passing plays.
    TDs. Led the league @ 16. 2nd was New England @12.
    Turnovers. 5th in the league @ 6.5% behind Buffalo, Arizona, LA Rams, and Pittsburgh.

    So how can we make sense of some of this?

    1. Perhaps, there is a design element that elevates sacks and TDs and really puts a boom bust stamp on how 3rd downs are approached.

    2. Perhaps, there is a talent gap on the field where the execution of 3rd downs results in these divergent outcomes. RW on down the line.

    3. Perhaps, that's just the way it is with the Hawks on 3rd down and nothing to think about. We enjoy the spoils and spoilage of the approach.

    Just kind of interesting to me how the Hawks aren't different in some ways, but quite different in others. I am particularly intrigued by the similarity in turnovers to the Rams and Steelers who are our closest 'like' peer in that regard, but the divergence between sacks and TDs for both relative to us and one another.

    Thoughts?


    Okay except the place were we are different which would impact 3rd down, is we go run, run, pass on 75% of our drives. That is a huge difference. We are predictable because everyone knows we will do run, run pass, most of the time and that hurts us on 3rd down.


    ^This

    I don't care if we run the ball but when the other team knows WHEN you are going to run you are, yes, VERY predictable.
    DomeHawk
    NET Veteran
     
    Posts: 1984
    Joined: Tue Jan 09, 2018 11:20 am
    Location: Ravenna


  • There are 9 combinations of binary play types through 3 downs. So every team is going to be predictable in some aspect. I think it would be easier to criticize the actual play-calls instead of the type of play called.
    knownone
    NET Veteran
     
    Posts: 2168
    Joined: Tue Sep 07, 2010 12:10 am


  • knownone wrote:There are 9 combinations of binary play types through 3 downs. So every team is going to be predictable in some aspect. I think it would be easier to criticize the actual play-calls instead of the type of play called.


    Really? Go back and watch the playoff game against the Cowboys.

    "Every" team? Right, every team that almost always runs the ball on 1st and 2nd down. It's pretty simple, just stack the box. The Cowboys showed every team in the NFL how to beat the Seahawks just as the Patriots showed every team how to beat the Rams.
    DomeHawk
    NET Veteran
     
    Posts: 1984
    Joined: Tue Jan 09, 2018 11:20 am
    Location: Ravenna


  • DomeHawk wrote:
    knownone wrote:There are 9 combinations of binary play types through 3 downs. So every team is going to be predictable in some aspect. I think it would be easier to criticize the actual play-calls instead of the type of play called.


    Really? Go back and watch the playoff game against the Cowboys.

    "Every" team? Right, every team that almost always runs the ball on 1st and 2nd down. It's pretty simple, just stack the box. The Cowboys showed every team in the NFL how to beat the Seahawks just as the Patriots showed every team how to beat the Rams.

    Yes, find me a team that is not predictable in their binary play calling. I'll save you the time; you won't.

    The problem here is that people are using 'predictability' with regards to running only, but in order to do that you must throw out the logical inverse. Consider this, the Seahawk's run the ball 59% of the time on 1st and 2nd down, the Chiefs pass the ball 59% of the time on 1st and 2nd down. Which team is more predictable? What conclusion can be drawn?

    Using the authors metrics, the least predictable teams in binary play calling on 1st and 2nd down are the 49ers, Jaguars, Dolphins and Jet's. So what does relative predictability tell us about offensive success abstract from the actual play being called? As far as I can tell, absolutely nothing.
    knownone
    NET Veteran
     
    Posts: 2168
    Joined: Tue Sep 07, 2010 12:10 am


  • DomeHawk wrote:
    knownone wrote:There are 9 combinations of binary play types through 3 downs. So every team is going to be predictable in some aspect. I think it would be easier to criticize the actual play-calls instead of the type of play called.


    Really? Go back and watch the playoff game against the Cowboys.

    "Every" team? Right, every team that almost always runs the ball on 1st and 2nd down. It's pretty simple, just stack the box. The Cowboys showed every team in the NFL how to beat the Seahawks just as the Patriots showed every team how to beat the Rams.


    They did, oh you mean with your best WR injured and two Guard that should have been on IR.

    But ok I wonder how well the Cowboys would have done with Amari Cooper out and their two starting guards playing on one leg.
    chris98251
    .NET Hijacker
     
    Posts: 28629
    Joined: Tue Feb 27, 2007 11:52 pm
    Location: Renton Wa.


  • chris98251 wrote:They did, oh you mean with your best WR injured and two Guard that should have been on IR.

    This is a great example of the connection between running and passing that lead to data analysis problems. With our interior OL seriously banged up we had difficulty running the ball and that led many to be upset after the fact that we didn't pass it more often. However, as off-season football forum posters well know a functional OL is also important for pass protection. We didn't see the possible outcome in which we came out throwing the ball like in Week 1, Russ got sacked eight times, and the fan base was instead furious that we abandoned the run.

    Whether that outcome would have occurred or not is entirely speculation and is not very well informed by looking at statistics that include games where we had a healthy OL. The interconnected aspects and much lower sample size is what makes the NFL much more interesting to follow and debate than the other major sports IMO because there are few obvious answers.
    AgentDib
    NET Veteran
     
    Posts: 3977
    Joined: Fri Oct 01, 2010 10:08 pm
    Location: Seattle


  • I don't care significantly if it is an incomplete, short of 1st down or a sack on third down. Exception within our own 20 yard line

    I would be curious if you looked at the stats that way how they come out. Does it really matter if you take a sack or throw it away on 3rd down? RW extends the play and guess what boom or bust is the result.

    Heck I am fine with the hail mary interceptions deep in opponent territory as long as we are in position to stop the person that intercepts the pass.

    The problem with stats is that they often treat everything the same. Big difference between an int on 1st down over middle of the field at the 40 yard line vs a bomb / hail mary / desperate throw that gets downed around the 10 yard of the opponent side
    mikeak
    NET Veteran
     
    Posts: 7730
    Joined: Fri Aug 24, 2012 7:24 pm
    Location: Anchorage, AK


  • mikeak wrote:I don't care significantly if it is an incomplete, short of 1st down or a sack on third down. Exception within our own 20 yard line

    I would be curious if you looked at the stats that way how they come out. Does it really matter if you take a sack or throw it away on 3rd down? RW extends the play and guess what boom or bust is the result.

    Heck I am fine with the hail mary interceptions deep in opponent territory as long as we are in position to stop the person that intercepts the pass.

    The problem with stats is that they often treat everything the same. Big difference between an int on 1st down over middle of the field at the 40 yard line vs a bomb / hail mary / desperate throw that gets downed around the 10 yard of the opponent side


    I'll take some time and try and suss out some situational stuff. FWIW, there were around 150 third down plays during the season so coming away with firm stuff might be hard but I'll take a look at it.

    I just had kind of an 'aha' moment of curiosity when I saw that the conversion rates were ho hum but the outcomes were more disparate than other teams.
    mrt144
    NET Veteran
     
    Posts: 4065
    Joined: Thu Dec 30, 2010 2:21 pm


  • Since the Dallas game reared it's ugly head in this topic .It wasn't so much that the Hawks offense was predictable ; because they won 10 games with the same predictability . It was the Hawks defense that lost that game. IMO
    xray
    NET Veteran
     
    Posts: 1292
    Joined: Sun Aug 19, 2018 6:29 am
    Location: AZ


  • xray wrote:Since the Dallas game reared it's ugly head in this topic .It wasn't so much that the Hawks offense was predictable ; because they won 10 games with the same predictability . It was the Hawks defense that lost that game. IMO


    No offense Xray but that is pure baloney, whatever problems the defense had were often the result of the offense not being able to sustain drives and putting the defense on the field for too long.

    Dallas stuffed the run and virtually every sportswriter AND Pete agreed we waited too long to open up the offense.

    Dallas had 23 1st downs compared to our 11.

    They had the ball almost 10 minutes more than we did.

    It was truly one of the worst coaching performances in the PC/JS era.

    https://www.espn.com/nfl/game?gameId=401038953
    DomeHawk
    NET Veteran
     
    Posts: 1984
    Joined: Tue Jan 09, 2018 11:20 am
    Location: Ravenna


  • Pete will always take the blame on his coaching staff rather than throwing players under the bus.

    DomeHawk wrote:They had the ball almost 10 minutes more than we did.

    How do you know that abandoning the run in the first half would have increased our time of possession rather than decreasing it further? It's also not clear how much time of possession even affected our defense on the two 4th quarter TDs they gave up. We made numerous mistakes on the first drive and the final one was sustained by two huge third down PI calls.

    DomeHawk wrote:It was truly one of the worst coaching performances in the PC/JS era.

    We narrowly lost a close road playoff game. Much of the outrage is because the game looked winnable, but the catch is that it looked winnable with our conservative strategy. Perhaps our odds would have gone up with more aggressive offensive approach and perhaps they would have gone down.

    You're confident about your opinion but you saw one roll on a dice where you don't know what is on each side or even how many sides there are.
    AgentDib
    NET Veteran
     
    Posts: 3977
    Joined: Fri Oct 01, 2010 10:08 pm
    Location: Seattle


  • DomeHawk wrote:
    xray wrote:Since the Dallas game reared it's ugly head in this topic .It wasn't so much that the Hawks offense was predictable ; because they won 10 games with the same predictability . It was the Hawks defense that lost that game. IMO


    No offense Xray but that is pure baloney, whatever problems the defense had were often the result of the offense not being able to sustain drives and putting the defense on the field for too long.

    Dallas stuffed the run and virtually every sportswriter AND Pete agreed we waited too long to open up the offense.

    Dallas had 23 1st downs compared to our 11.

    They had the ball almost 10 minutes more than we did.

    It was truly one of the worst coaching performances in the PC/JS era.

    https://www.espn.com/nfl/game?gameId=401038953

    The Hawks had the overall 17th ranked defense going into that game last season. And it showed badly ....if you want to blame the offense and the coaching that's fine.
    t
    xray
    NET Veteran
     
    Posts: 1292
    Joined: Sun Aug 19, 2018 6:29 am
    Location: AZ




It is currently Fri Nov 22, 2019 12:47 am

Please REGISTER to become a member

Return to [ THE OFFICIAL NET NATION FAN FORUM ]




Information
  • Who is online