I’ve spent a lot of time with the Prediction Machine lately. I’m enjoying the process of generating predictions for various offenses based on recent history, Vegas lines, and coaching staff changes. I’ve taken quick looks at both the New York Jets and New York Giants. Up today, my hometown Lions.
Refer to the table below as much or little as you wish. It contains the details of how I set the input variables for the App. The Projection Machine allows you to fine tune many aspects of a team-level prediction. I wanted to give you the information I used, but didn’t want to spend a thousand words explaining all my choices.1 In general, my approach was as follows:
- Make team level estimates based on the history of Jim Caldwell and Joe Lombardi (e.g. New Orleans) offenses.
- Make player level estimates based on the player’s recent performance and tendencies from Caldwell/Lombardi offenses.
- Incorporate offseason news and research in a conservative manner.
2014 Lions Projections
|Pass tend||0.04||Det/NO Hx|
|QB SK RT||0.035||Stafford Hx|
|QB RUSH PCT||0.07||Det/NO Hx|
|QB INT RT||0.025||Det/NO Hx|
|WR1 TGT||0.25||Johnson/Graham Hx|
|WR1 CR||0.6||Johnson/Graham Hx|
|WR1 YPT||9.25||Johnson/Graham Hx|
|WR1 TDRT||0.065||Johnson/Graham Hx|
|WR2 TGT||0.17||Det/NO/Sea Hx|
|WR2 CR||0.6||Det/NO/Sea Hx|
|WR2 YPT||8.75||Det/NO/Sea Hx|
|WR2 TDRT||0.06||Det/NO/Sea Hx|
|WR3 TGT||0.08||based on rest of parameters|
|TE1 TGT||0.1||Hx ave rd 1 TEs|
|TE1 CR||0.59||Subpar to reflect scout concerns|
|TE1 TDRT||0.035||Subpar to reflect scout concerns|
|TE2 TGT||0.08||Fauria/Pettigrew blend|
|TE2 CR||0.63||Fauria/Pettigrew blend|
|TE2 YPT||6.55||Fauria/Pettigrew blend|
|TE2 TDRT||0.07||Fauria/Pettigrew blend|
|RB1 RUSH PCT||0.44||Det/NO Hx, Offseason news|
|RB1 YPC||4.1||Det/NO Hx, Offseason news|
|RB1 TDRT||0.032||Det/NO Hx, Offseason news|
|RB1 TGT||0.13||Det/NO Hx, Offseason news|
|RB1 CR||0.7||Det/NO Hx, Offseason news|
|RB1 YPT||6||Det/NO Hx, Offseason news|
|RB2 RUSH PCT||0.39||Det/NO Hx, Offseason news|
|RB2 YPC||4.1||Det/NO Hx, Offseason news|
|RB2 TDRT||0.04||Det/NO Hx, Offseason news|
|RB2 TGT||0.11||Det/NO Hx, Offseason news|
|RB2 CR||0.72||Det/NO Hx, Offseason news|
|RB2 YPT||6||Det/NO Hx, Offseason news|
|RB3 RUSH PCT||0.09||Theo Riddick guesstimate|
|RB3 TGT||0.03||based on rest of parameters|
That’s near the top of the league in projected number of plays, but fewer than Detroit ran last season, and almost identical to what Baltimore (Caldwell) and New Orleans (Lombardi) ran. Pass attempts and pass/run ratios are right in between Detroit and New Orleans’ 2013 seasons. Rushing attempts fall in between Detroit/Baltimore and New Orleans. In short, I think this is fairly close.
Matthew Stafford is being drafted as one of the top fantasy quarterbacks this season, and with apparent good reason. A drop in interceptions and a slight bump in accuracy from last year, and voila, a projected top three finish. Later on we’ll look at what tweaking some inputs does to his projection. In the meantime, know that he’s a point spread bully and is undervalued in dynasty.
We don’t need to spend much time on Calvin Johnson. He’s firmly atop our rankings. Golden Tate’s projection also seems reasonable, but higher than we initially ranked him. Last season Nate Burleson and Kris Durham combined for 140 targets while playing second fiddle to Megatron. I projected Tate a bit lower partly to be conservative, partly because the New Orleans offense targeted its second receiver less frequently, and partly because I expect Eric Ebron to get a decent number of targets. The NFL Career Graphs App suggests that the yards/reception and catch rate numbers for Tate and Johnson are reasonable.
The third wide receiver projection was negligible. Given the abundance of targets elsewhere, and crowded situation after Johnson and Tate, I wasn’t sure who to project. Even though he’s apparently healthy and might not suck, I’m not sure the third WR will be Ryan Broyles. And even though he could eventually be Antonio Brown, T.J. Jones isn’t healthy. I basically allocated the pass targets not designated for anyone else to the WR3 position, and left the inputs at league average.
I struggled with the running back projection. I think this could go a few different ways. In the end, I opted for a roughly even split, based mostly on how Detroit used its RBs last year. I also threw in a bit of work for Theo Riddick, based mostly on offseason reports about his performance.
The player with the most to gain here is probably Reggie Bush. His projected workload varies the most from his historical use. I also undersold the yards per reception for Bush and Joique Bell, to better match what New Orleans’ backs have done recently. In other words, if the workload breakdown holds, there’s upside for both Bush and Bell. From the Career Graphs App, take a look at the rushing and target percentages, touchdown rates, and yards per reception. I tried to balance things, but if Bush and Bell perform closer to their recent benchmarks, they can finish higher in the fantasy standings- Bush perhaps even in the top 10.
Tight end time. That stat line for Ebron came about as follows. I used the Pro Football Reference Player Screener tool to find all first-round tight ends since 2000. After eliminating Ben Watson (only one game as a rookie) and Kellen Winslow (only two games as a rookie), the rest of the TEs averaged 10 percent of their team’s pass targets in their rookie season.
So, that’s what I used for Ebron. I set his catch rate at 59 percent and TD rate at 3.5 percent to reflect concerns about those parts of his game. I personally don’t think either is an issue, but I’m trying to be a bit conservative here.
For the TE2 numbers, I expect these to be mostly Joseph Fauria targets for a couple reasons. First, the way Detroit is using its TEs in training camp. But mostly, they’d be crazy not to target Fauria, especially in the red zone, and especially if Ebron isn’t as good in that area of the field. In order to account for the fact that Brandon Pettigrew is still likely to get some targets, however, I set the TD rate and yards per catch a bit below Fauria’s 2013 benchmarks. The Career Graph App shows both Pettigrew and Fauria’s performance last season.2
If you came looking for something earth shattering . . . sorry. I think these projections, if the players generally stay healthy, are reasonable. What I think is valuable however, is that doing this exercise gives me a sense of confidence about how the Lions offense will perform from a fantasy perspective. Tate seems a bit undervalued, Ebron looks promising long term but maybe overvalued in redraft, and I expect Bush and Bell to reprise their top 20 finishes from last year.
I also think it’s good to look at a range of outcomes, so let’s see what happens if we tweak some of the settings. What if we just set the team level settings to league median: an average point margin/play of -1.5, a pass tendency of 0, and a pace expectation of 0. I think that’s unlikely; Detroit has been consistently positive in pass tendency and pace, New Orleans is consistently a plus pass tendency and plus pace expectation team, and Caldwell’s Ravens teams were as well.
But should such a scenario come to pass, it would shave 2.5 points per game from Stafford’s fantasy production, dropping him to QB10. Megatron takes a 2.4 fantasy point per game hit and drops to WR12. Tate drops over a point per game and falls to WR32. The impact to Bush and Bell is minimal, however.
I present this scenario not to caution you against any of these players (although if you feel confident in the workload projected, the RBs look more or less immune to changes in game script and have some upside based on potentially being more efficient than projected), but to suggest that this is the likely floor for Detroit’s skill players. Hopefully this exercise increases your confidence about their potential 2014 production, and helps crystallize where they represent value in redrafts.