State of the Roster: Building over time

Last year I provided illustrations, by color, of the Eagles offense and defense, using it to identify weaknesses and strengths.  Today I want to take that to the next level, for two main reasons.  First, PFF is now providing the same visuals.  Maybe they did this before and I didn’t know about it, but in any case, there’s no reason for me to duplicate what they’ve already done.  Here is their projected line-up.  Clicking it will take you to the source write-up.

Screen Shot 2014-02-24 at 3.46.18 PM

The second, and more important, reason that I want to do things differently is because the above chart misses out on a vital aspect of team construction.


You can’t sustain success if you’re entire focus is the next season.  The team needs to be built so that it can contend OVER TIME.  Given how much luck there is in the game, it makes more sense to contend over a long period of time rather than “go for it” in any one season.  Andy Reid’s Eagles did this perfectly, they just never had things fall the right way.

This is particularly important for the Eagles because of where the team is currently in its team-building process.  I mentioned in my last post that there are still a lot of holes, and some people pushed back.  While there’s plenty of room to disagree over the projected quality of each player, I realized we first have to agree on just what timeframe we’re looking at.

For example, if we are just considering next season, than Jason Peters is far from a “hole”.  However, if we’re looking at creating a 3-4 year “window” of contention, then things become a bit more difficult.  Peters is 31 years old.  Do you think he’ll still be an above-average OT in his mid 30s?  It’s possible, but the point is that it’s not enough to just look at this coming season.

The question then becomes:  What’s the best way to alter the graphic above to incorporate longer time-projections?

Today I’ll take a stab at that.  First, though, I want to note that the goal is obviously to build a team that contends for a lot longer than 3-4 years.  However, we have to recognize the limitations, or margin of error, in any attempt to project future performance.  The farther out we go, the less accurate or predictions will be.  Therefore, once we go out to 5+ years, there seems to be very little value in attempting to projecting player performance.  That might not be right, but it’s the constraint I’m operating within for right now.

So….the projections:

As you’ll see below, I simply listed every starting player on offense and defense, then assigned them a color based on how I believe they will perform in the relevant year.  For now, we’re going very low-resolution, so I’ve separated players into just 3 groups.  Red is below-average/bad, yellow is average/mediocre, green is above-average/objectively good.  There’s definitely room to refine this (McCoy could be blue, as PFF did above), and I’ll do so after FA and the draft.

The Offense

Screen Shot 2014-02-24 at 4.17.09 PM

As you can see, I also included the salary cap numbers, mainly to show when players’ contracts were up.  Overall, the offense looks pretty good.  There are just 2 areas that could use some immediate improvement, RG and the TE/WR situation behind D-Jax.  Of course, if Maclin resigns and comes back healthy, he could easy go “green” for the foreseeable future.  That would help a lot.  Still, though, the offense looks like it’s in really good shape.

The concern comes in year 3.  As illustrated, Peters, Mathis, and Herremans are all in their 30’s, and while they can certainly perform at a high level into their mid-30’s, we have to ask ourselves:  what are the chances of that happening?  Moreover, what are the chances that all 3 of them do so?  Slim, at least in my opinion.  That’s why I really wouldn’t be upset to see the team add an OL in the draft (beyond the 7th round OL the team should be drafting every year, but that’s a different issue).

I certainly anticipate Zack Ertz moving into the starting TE role, and being a good player for a long time.  Notice, though, that still leaves a whole in the roster.  Anyway you cut it, the offense has space for another receiving weapon.

The Defense

Screen Shot 2014-02-24 at 4.17.18 PM

Now we get to the fun part, the defense.  It won’t shock anyone to learn that the defense has a lot more holes than the offense.  There are some tough calls to make here, as far as projections go.  Can Earl Wolff go “green”?  Yes, but I think the odds still favor him being just OK as a starting safety.  Similarly, some people will probably argue that Bennie Logan deserves a better rating.  Again, its possible, but I’m not convinced.  Additionally, it’s very harmful to the construction process if you assume guys will hit their “ceiling”.  Many don’t, or rather their ceiling gets adjusted downwards as their career progresses.

Most notable here is the lack of good players in the secondary.  Boykin might be able to slide over to the #1 or #2 CB spot, but that still leaves a hole to fill.  Also, remember not to focus on just the first year.  Look at year 3 above.  Naturally, the team has a few offseason to address the issues, so their not urgent, but it has to be accounted for.  Demeco Ryans isn’t getting any better, and how much longer do you think Trent Cole can play effectively for?

You surely noticed the thick black box in both graphics above.  That’s what I’m looking at as the “strike zone” for this team.  If I’m Howie Roseman, I want to turn as many of those boxes green as I can, even if it’s at the expense of next season.  Also, I want to eliminate every red box.  This is what I mean when I mention accounting for time.  You’re not just trying to build the best team you can.  You’re trying to get a lot of different pieces to fit come together at the right time.  It won’t work if your young players improve to “good” just in time for your veterans to decline.

Again, some more work to be done here, and I’m hoping to do a little research into the general career arcs of different positions (so we know things like what age an OT should be expected to decline).  Keep this in mind though when you hear about potential FAs.  How well to they fill those boxes?


16 thoughts on “State of the Roster: Building over time

  1. Great post. Don’t misinterpret my comments as negative or disagreeing overall, but I don’t think it’s practical to project an NFL team beyond 3 years. Things are just too fluid.

    For example – when has a ‘stud’ RB ever been a ‘green’ level player for more than 5 seasons? Mccoy already has 2 under his belt. I know I wouldn’t put money on him being a stud RB four seasons from now, not with the number of carries he gets/will be getting. In the league right I can think of Marshawn Lynch as a guy who has seemingly remained green for 5 years. Can’t think of anyone else but I’m sure there’s at least one or two. Then again, I’m biased: I think RBs are more a product of the system. Mccoy is special no doubt, but considering Bryce Brown came into Andy’s system and broke and dominated 3 weeks in a row, and quite frankly, looked better than Mccoy did during that time….I don’t know.

    The only two positions I can think of where players consistently play well into their mid-30s are QB and OL.

    Loved seeing this:
    “Andy Reid’s Eagles did this perfectly, they just never had things fall the right way.” Well put! You are the first person I’ve ever read to write this. I spent the 00’s spouting this to my fellow Eagles fans, as they kept up with the mythical “Lurie and Banner are cheap!” mantras….and then the Eagles screwed us and seemingly listened to the dope fan base and went “all in” and blew the whole thing up as a result.

    • Fair critique. RBs are especially tough to project, and you could certainly convince me that he deserves fall-off earlier. You and I agree on the time-frame bit; I know 5 years is stretching it. However, only showing 3 years really wouldn’t have allowed me to include the contract details I wanted to (or to make the point about the OL). Maybe I can adjust the shades so that the reflect a lack of confidence rather than helping to differentiate the contract status.

      To the age point, I believe there’s been some research done on the average career arcs of players, but I haven’t had a chance to find/review it. If not, i’ll see if I can put something together.

      Andy Reid is a great coach and deserves a lot more credit than he receives from a lot of Eagles fans. Then again, casual fans tend to reject the notion that luck plays a huge role in determining the outcome of games.

  2. I love the set-up and the long-term thinking, but as a color blind person I have a lot of trouble interpreting your charts. Even with a color picker tool, it seems that you’re a little inconsistent about the green or the red or the yellow you use, so it’s tough. Can you add a legend? Also, in general, for color blind people, blue-yellow-red is a good combo to use because the colors are so different.

    • Just so you know you’re not going crazy, there are indeed different shades of red, green and yellow – the distinction is that the boxes with a number in are a brighter shade than those without but the ones with numbers and the ones without numbers seem to be the same as each other (not sure if that helps you figure out which is which among the rest of them).

      • There are different shades. I was trying distinguish between under-contract years and after-expiration years. So each color has two shades. Sorry about the color-blind confusion, I’ll try to use blue instead of green.

  3. I love this way of thinking and mapping out the state of the roster. One comment–as a color blind person, it’s very difficult to interpret red/green charts (though yours also has yellow). Blue-yellow-red would be a lot easier to interpret for me and many others who are color blind. Alternatively, you could be more consistent in the colors (there are several reds and greens, I think), add a legend, and then I could use a color picker tool.

    (My apologies if you see two similar comments from me; not sure whether the first went through.)

  4. Hi, as a research economist, I always like your quantitative approach (your glob made me think Mmm… I might want to do some analysis myself…). If you think about risk factor, I think you could think about probability (maybe conditional on position) of getting injured. I don’t want to sound too negative, but what is the probability that Shady survives the “box” without missing a big chunk of time. I think it’s almost zero, considering the likelihood of injury for (especially heavily used RB). And unless Foles becomes like Brady, our season might be over once that happens. That’s I’m worried about OL, too. the first backup seems pretty decent, but at this point, the likelihood of having two guys missing a large chunk of time is nonegligible.

    Anyway, keep good work!

    • Injuries are definitely a wild-card, but I don’t see a good way of representing that. Perhaps outlining the boxes of injury-prone players/positions with red would help. That of course raises the question of what constitutes “injury-prone” though.

      I think the best way to handle it is simply to expand the chart to include the rest of the depth chart. That way, you can see where a single injury will have a dramatic effect (like Foles being replaced by Barkley perhaps).

      Agree that McCoy is a particularly large source of risk given his skill and importance to the offense. Another reason why we might see a “skill” position FA signing and/or high draft pick. The offense is the team’s strength, but need to build up the resilience as well.

  5. Great analysis. I would love to see you also focus on depth at the positions as well. It would be harder to determine color progression for players that don’t get significant game time, but identifying areas of relative depth strength (especially as it relates to potential and youth) would go a long way to determining an offseason strategy. Is there a conceivable heir apparent to an aging position on the roster (think LB and OL projects)? What are the ceilings to those prospects, and how can you ensure that you have multiple options for future slots if certain young players don’t pan out? I think you could build off of your model to visualize a bit of that information.

    Thanks for always focusing on original, forward-thinking analysis instead of recycling the same content found elsewhere on the web. It’s nice to know that when you start reading one of your articles you’re going to learn something new, or at least re-visualize information that you don’t question enough. Chip Kelly would be proud.

    • No doubt. If I was Roseman, I’d have this chart for the entire team. For this post I just wanted to make things clear. Including everyone would have made that difficult. I’ll add to it though, and try to include more players and group them by position. I’ll likely wait until after FA and the draft though.

      Thanks for the compliment. Doing what I can. I realized that I tend to think/visualize things differently than most of what I was reading online, so figured I might as well provide a different perspective whenever I can.

      I do think Chip would approve…

      • You should figure out a way to send this directly to Howie Roseman. Seriously – if he doesn’t have a chart like this then I bet he’d immediately see the value in it. Next thing we know you’d be The Assistant to the Assistant General Manager for the Eagles!!!! (don’t laugh, that’s how Roseman started!) : )

  6. Great analysis as always. The chart reinforces my desire to get a big time wr earlier rather than later in the draft. The chart also nicely highlights the issue with Ryans. I wouldn’t be opposed to the team going ilb early in the draft.

    I agree with then utility of somehow building.risk into the chart and the issue with shady (and really desean). The next few years are a critical window if we want to capitalize on their current high level of play.

    If the chart is to help make more informed decisions about the draft and fa, there are at least two other factors that could be useful to consider. First, the relative importance of a position given scheme. There should be some differences across teams, but there will be some similarities (e.g. Qb is important everywhere). Second, scarcity of decent players at a position. If good safeties are scarce and a big hole on our roster and are important to the scheme, we should prioritize accordingly.

  7. Hey Brent,

    i am an eagle fan from Germany (sorry for my bad english, i have been out of school for years now, but i am trying my very best 😛 ) and first of all: Love your analyses and blogs! They truly deserve more attention and viewer. Walk on 🙂 !

    You have mentioned Maclin, but not Cooper. How would you classifiy him?

    • Cooper is an interesting case. He has the physical tools to be a good WR, but last year it really seemed to me that he just benefited from the increase snap count and targets. That’s not to say he’s bad, just that there are a lot of WRs in the league who could do what he did last year given the same opportunity. I’d be happy brining Cooper back, just not at an unreasonable salary. If I put him in the chart, he’d be “yellow”.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s