One of the ideas that we’ve written about on the site is what we call “Winning the Flex” and is related to Shawn Siegele’s guidance that the flex position wins fantasy championships. Winning the Flex is also an idea that is tangentially related to Zero RB, to the extent that Zero RB says that you should load up on WRs because they’re the highest scoring position in PPR leagues, and when the flex is accounted for you can start a lot of them.
In the past I’ve used expected fantasy points based on Average Draft Position to show that the oft cited criticism of Zero RB1 is pure semantics if WRs are structurally undervalued. But that’s the past and what’s happened in 2016 is a historic increase in value at the WR position. In fact, WRs have gotten so expensive relative to their historic levels that it’s fair to wonder whether things might have over-corrected. If there’s an over-correction then Anthony Amico’s strategy of going RB heavy this year could be a league winner.
To look at this issue I made a simple tool which lets you create a purely ADP based model to project fantasy points. The only explanatory model variable being used for this tool is the player’s positional ADP rank. So all we’re doing is feeding in some past ADP data, and past fantasy points scored, then we’re using a LOESS model to smooth things out, and we’re applying that model to the year that you choose in the app. For example, if I select PPR scoring, select 2016 as the year to predict, and then set the slider tools years to 2010 to 2015, the app will do the following:
- Use 2010 to 2015 data to create a model where positional rank is used to explain end of season PPR fantasy scoring – this is done on a per position basis
- Apply that model to the 2016 ADP data, which is pulled from My Fantasy League
- Sort the results by expected points according to the model we’ve just created
- Calculate EQUITY, which is the difference between the player’s overall ADP, and their overall rank in the new expected points number we’ve calculated
I created this tool in about an hour, so my intent is not to be exhaustive in this exercise. I just wanted to be able to look quickly at whether WR ADP had over-corrected. For that reason I’m only using one source of ADP (I didn’t go looking for standard scoring specific ADP). But what you can do is download the expected points derived from this model and use the data as you draft various teams. I’ve been in some drafts lately where it really did seem like WRs had gotten so expensive that I might be better off drafting RBs. But after looking at this data I don’t think that’s the case anymore.
My reason for making that conclusion is that if I set the sliders to train the model on the most recent four years (2012-2015), the WR position still has the positive Equity numbers. RBs are still negative. One of the things that I encounter in monitoring the discourse on Zero RB is that there exists a group of fantasy players who never got on board with Zero RB, who haven’t changed their mind at all, and who now are just promoting the strategy that they never abandoned as being the contrarian one. But if you draft as if you’re just going to take the highest predicted scoring flex eligible position for about six rounds, you’re still going to end up going Zero RB. This isn’t to say that things couldn’t change to make RB-heavy a league winner in 2015, it’s just that if we inform our expectations for 2016 based on what’s happened in the past four-ish years, we won’t be drafting RBs early. That’s despite the fact that the market appears to be closer to catching on that WRs have been more valuable than RBs.
Note this all is conditional on PPR scoring. You can look at the app to see how things change in ordering if you use standard scoring. You can also change the app sliders so that it’s focused on 2000-2006, and then you would see that at one time the focus on RBs as league winners really probably was warranted.
Right now this is just a simple tool to let you see what things would look like if we used an ADP-only model to project fantasy points. If I have more time to work on it then I might be able to pull in more ADP sources so that you could download more customized cheat sheets using this data.
- The critique is that you shouldn’t tie yourself to any strategy and you should always take best player available. (back)