The Model Intro
Let’s get this out of the way, shall we?
This is not subjective. I did not put any personal interest, thought, or subjectivity into the outcomes that you are about to see below.
Now that we have gotten past the disclaimer, let’s dive into to how predictive modeling works, the foundation to the outcomes, and the top 50 players predicted by this model for the 2020 season.
Player predictability is something that is quite difficult in the world of football. It is so volatile and the ultimate team sport that predicting individuality can be quite difficult. However, we can get close, and this closeness can help you with a foundational pillar that can begin the process of building out your own rankings.
Ultimately, there are two steps to creating a predictive model on a PPG basis.
- Time Series analysis to create a player lifetime curve based on age and statistical evidence of breakouts
- Regression analysis on seasons prior to predict future success.Both of these factors are combined to create my Wide Receiver Fantasy PPG Prediction Model. How do the results look? Below is an example of last season’s PPG, vs. the predicted PPG from the model.
Although there are some differences, it is extremely consistent in terms of the outcomes. So much so that it has an R^2 of 0.56. In terms of an analytical PPG model, you will be hard pressed to find one that is as close to this.
Now that you understand the legitimacy of the model, let’s jump into the projections for 2020, shall we?
Most of this is self explanatory, however, I’ll include a bit of commentary about players that may be surprising, and why they landed in the spot that they did.
2020 Model Results: Top 1-10
The top 10 already has some spicy headliners. Allen Robinson finding his way into WR4 may surprise people, but there is so much to like coming into this next season. He has elite red zone production, which actually has stickiness year over year. On top of his high volume, he posts an elite 2.8 yards per route run, which is one of the most predictive measures in terms of future wide receiver success.
Courtland Sutton is one that will surprise others as well. However, maybe it shouldn’t. Sure, the Broncos drafted additional weapons. However, it is tough for rookie receivers to come in and make a major impact. Additionally, like Robinson, Sutton was an elite red zone threat, posting an incredible 15 receptions around the red zone last season. No one that was brought in to Denver posts that kind of threat for red zone production. Additionally, a 13.5 yards per reception and 2.62 yards per route run were extremely productive for the position. Although WR8 may be ambitious, it’s not out of the realm of possibilities.
2020 Model Results: Top 11-20
The interesting results continue with Tyler Boyd entering the top 20 in the model slotted at WR17 with a predicted 13.5 fantasy points per game in 2020.
Maybe this shouldn’t come as a shock, however. Boyd had an incredible 147 targets last season while posting a very average 9.3 yards per reception and 2.01 yards per route run. It can be predicted along a time series that QB play will increase, comparatively to last season. Although A.J. Green is coming back into the fold, the model thinks that this could be a positive for Boyd, as even if his targets go down from last season, all efficiency metrics are pointed in the right direction.
2020 Model Results: Top 21-30
I think the more shocking aspect of the top 30 is more about who isn’t on it than who is on it at this point. Most of these players seem pretty stable and a really good bet for hitting the top 30.
Julian Edelman may be a bit of a shock, however. His stability over time has been proven and he really has no true target competition behind James White (assuming N’Keal Harry does not make the jump, which all of my models show that he will not.)
2020 Model Results: Top 31-40
Oh, look. Everyone’s favorite sleepers are sitting here in the 31-40 range. This is where a lot of the contradiction may come, analytically speaking vs. actual outcome. However, there are certain risk factors that do come with the players in this tier and that is factored into the model.
With Juju, we have never seen him be productive on his own as the Number 1. His talent group around him is all projected to take a step up, and how sure are we that Big Ben will look anything close to himself this next season? It’s understandable that he is a bit lower in these type of rankings, but this does feel extreme.
Additionally, A.J. Green is one that people are really excited about for this upcoming season. How many targets will there really be to go around for the Bengals in 2020? Additionally, how healthy is A.J. Green? These are risk factors that do play into his predicted PPG, as well as a yearly decline in predictive statistics over his last couple of healthy seasons.
I personally think Adam Thielen will outproduce this ranking. He has no target competition in an offense I believe will be very efficient.
2020 Model Results: Top 41-50
Not a lot to see here, not a lot to get excited about. The only thing I can really say about this group, and the WR group as a whole, is that it is DEEP.
This feels like a season where you can totally punt on WR heading into your redraft leagues. You should feel comfortable taking your first wide receiver in the fourth or fifth round this upcoming year, taking elite talents at the RB and TE positions earlier in the draft. You can feel really comfortable with a lot of these guys sitting in your WR2 spot in your redraft leagues, and you don’t have to spend high draft capital to acquire them.
Conclusion
Do not go and take these rankings and make them your own. That is not how this thing works. However, this gives you an extremely strong foundational footprint as to what the analytical side of evaluation is saying about these players, and where a realistic outcome for these guys may be.
Some of the rankings may seem far fetched, but keep in mind with a 55% success rate on ranking these, it is right more often than it is wrong. Most of the incorrect aspects of this come from confounding variables outside of the actual football production.
It’s important for us, as fantasy football players, to accept and absorb all sorts of knowledge. This can be one of the many tools you use to create, plan, and execute your strategy going into this next season.
