Message from the Hawks Cast Team || Hawks Cast Forum Rules

Moneyball, elite QBs, analytics...

  • My good buddy @sammyc521 got me using Twitter this past season. Before joining it, I hadnt really been active in the online Seahawk fan community for a couple years.. wasnt exposed to the tremendous good there is out there, or the annoying bad. Seeing a lot out there lately over whether Russell Wilson is being wasted by the Seahawks coaching staff and Pete in particular. And people using analytics to justify it. And its annoying. So Im going to put my thoughts in here now:

    I want to say this first, before people come at me with strawmen and accusations of me being a dinosaur: I think analytics has a place. I think every team should have a team of people doing them. But I will also say that football is a very difficult sport to use them with for a variety of reasons.

    Here is why most of what is used in these discussions is wrong:

    1. You dont have all the information. Basically if they are honest, any analyst who isnt privy to the teams knowledge is guessing. Some guesses are better than others and some are easier than others ...but you dont know what the call was, what the route was, what the coverage was.. who was allowed and not allowed to freelance.

    2. How MY team is built matters. Comparing stats from all teams from all games from the past X years is fine.. but this arent coin tosses or dice throws..this isnt a matter of "if all other things are equal situations". If I can get a road grader for 2.5 million a year exactly BECAUSE everyone is trying to pass... and if other teams are looking for DTs who can pass rush at the cost of some run stopping ability.. then that changes the equations on what I "should" do with my team. It doesnt matter that in 2015 such and such a system or play worked in certain situations..because again, this is not random or standard.

    3. Its not what it claims to be.. I keep hearing people tell me this is SCIENCE. No. It isnt. There is no control. There is no testing. There is no ability to reproduce the results. And very tellingly..the armchair analytic types spend their time measuring and talking about what has already happened. Specific games ...there is some guessing..I havent seen anything showing they are more accurate. And I dont see pre season predictions based on analytics telling us how the season will end up. And the opposite.. if for example, running ruins your chances to score (an oversimplification I admit of what some of them say)..but if points per rush is zero so youre just wasting your time, then the teams that run the most should score the least points. But we know that isnt true. Because football is too complex to break down that way.

    4. Projection. There is SO MUCH projection going on. So many people know get to say stuff like "the coach was wrong, and here , I have proof"..and its bogus but it sure as heck makes them feel good.

    So thats whats wrong with the process..then ..the use. Analytics is best used to find inefficiencies in the system -- players undervalued. This is exactly the opposite of how its actually used. Ive touched on this before but will write it again:

    If everyone buys in on the same data set and the analytics that go with it.. then what happens? Everyone is going to try to get the same quarterback. The same type of WR. Then they will look for the same type of CB, EDGE, et al.

    So what happens to those players? If youre a classic WR1 type.. Julio Jones, Antonio Brown, Hopkins, etc.. you get PAID. Everyone needs that WR1!! So everyone has to pay for it.AND people are going to pay HUGE money to bring in the "elite" CBs. What if you have a quirky system, designed around a QB that can extend plays.. and so you find a WR that with THAT system can produce efficiently and at a WR1 type level but doesnt "look" like a get that player for 4, 5, 6 million dollars a year less? Now youre identifying inefficiencies in the system. Now youre playing a team that likely ponied up huge money to cancel out conventional WR1s with their shut down CB.. and neutralizing that with how you use your receivers, and using a system that can even get by with cheaper receivers than that.

    If you build that conventional way.. its very difficult to get your competitive advantage. You have to be the one to get the most of those ideal players and youre competing against everyone else.

    But if your analytics people tell you "Hes too short but he produces"..then you get first dibs on a talented QB with a unique skill set you can build around, all because he didnt look right. If make a defensive system that is demanding on CBs but demanding in a different way.. set technique but limited types of routes they need to cover.. then you can make shut down corners and plus level corners out of players other teams cant use.

    NOW youre moneyballing.

    You can run to win. If YOUR team is built that way. And we actually have seen a bit MORE of it, not less lately as teams have figured out: if that teams D line is super expensive but built to rush the passer.. then .. I should be able to run on it.

    Finally.. you STILL NEED an elite QB if youre a run first team. You might , arguably, counter intuitively, need an elite QB more. Because if you have 10 drives a game that need, say 2 difficult pass completions to continue the drive..and your QB is a 40% thrower in those situations..then youre only going to expect to keep 1.6 of those drives going. If you run and so the game is short.. you have 8 drives needing those passes... with the same QB youre now scoring on 1.2 drives. But if you have a QB who makes 65% of those passes.. now youre continuing 3 drives. And you get all the hard to quantify advantages of running the ball -- which are dependent on what players YOU have. What scheme YOU run. What game plan you run and what play calls you call.

    At the end of the day, my point is this: you can win lots of ways and Wilson isnt being wasted. At all.

  • I can't praise/upvote this thread enough.

    I am 100% on your assessment on QBs + Analytics. That's not the argument we usually have which is "how do you pay the QB X at the cost of Y?"

    Essentially Pete and John have cornered the Market on taking advantage of players who are viewed as inefficient to the greater NFL. Sweezy and Fluker are great examples of players that fit what the Seahawks need but doesn't appeal to the masses. This means they can capitalize at greater savings which helps them elsewhere.

    When it comes to the analytical breakdown I usually just give it a cursory glance and not much though for all the reasons you outlined; there's no controlled environment in football. So all the numbers are great indicators of what happens but not what is to come.

    A WR that averges 50+ catches, 700+ yards, 11 ypc and 5 TDs for their first 5 years does not mean that is what they will do in their 6th year. No one expected Tyler Lockett to explode this year except Pete and John. Everyone was freaking out and saying we overpaid and extended him too early... but had the Seahawks waited until FA they would have probably lost him to another team.

    And to end things, pay Wilson what it takes to extend him.

  • This thread grew out of frustrations last night (my time) with a guy on Twitter..who guessing by his twitter handle maybe works for an Analytics place and he was just so smug and so confident.. proclaiming that the Seahawks offense takes away what Wilson should do best and forces him to be inefficient. 2 of the top 5 QB ratings for QB to WR for the past 2 years are Seahawks. Lockett and Baldwin. THAT is efficiency and results from , of course Wilson talent, but also the scheme.

    If the Seahawks are holding back Wilson, what is it they project hed be able to do..? He is superman? The Seahawks should average 50 points a game?

    The dogma some of these people bring to the conversation is weird, and then when you call them on it you get "no one is saying its like baseball". Well, if you admit that, how can you NOT admit youre likely wrong? If the Seahawks were doing the exact wrong thing and were a top 10 offense team..what did you project them at? Youre gonna tell me with Lockett and a broken ADB and Moore as our receivers our passing game last year if wed used it more was going to be better than teams with all pro tackles and WRs?? Can you not see the tail wagging the dog.. teams have built for pass and so passing has become efficient? But that doesnt mean if you build a different way you wont be successful. And as I outlined above , youre arguably MORE likely to be successful if you have a sound philosophy thats different because youre going after a different set of guys. The Seahawks choose their line different, their WRs different, their CBs different than most teams..

    Anyway.. it was driving me a bit mad yesterday. 🙂

  • Great post. The pro-analytics, anti-run crowd manages to get under my skin to the point where I'm finally using my mute button. So many holes in their logic but a complete refusal to consider counter opinions.

  • @Al-U said in Moneyball, elite QBs, analytics...:

    Great post. The pro-analytics, anti-run crowd manages to get under my skin to the point where I'm finally using my mute button. So many holes in their logic but a complete refusal to consider counter opinions.

    Hi Al,

    Welcome to Hawks Cast.

    I agree that a lot of the messaging is looking at the data but then trying to pitch that it can only be viewed as one thing.

  • Very good post even if I don't fully agree. A lot of it is spot. is Duane Brown, who has more information and is closer to the situation than all of us and he sure sounds like he think they needed to lean on Russ more:
    Check out @PatrickClaybon’s Tweet:

    I'm not anti run I just think they needed to lean on Russ much more than they did in the Dallas game. And Pete, Schott, Russ and now Duane to their credit have already admitted it.

Log in to reply