Always play the matchups.
Weighing matchups helps you make difficult decisions between similarly valued players, and every week, it's important to analyze the best (and worst).
So what exactly constitutes a favorable or unfavorable matchup?
The answer to this is never trickier than it is annually in Week 1. Matchups data are the sketchiest they'll be all season in the opening week, and it's a mistake to present "matchups numbers" that pertain exclusively to last season's or preseason statistics. Citing solely 2017 statistics assumes no change to defensive personnel, whether players on the roster or injury situations. Citing 2018 preseason statistics puts weight on a four-game sample in which first-team defensive players played limited snaps (often one-third of those games or less).
An additional wrinkle to consider with 2017 statistics: Fantasy point totals against individual skill positions are useful tools, but they need to be taken in context, considering the strength of schedule those teams faced.
For example, the Detroit Lions in 2017 afforded the 13th-fewest fantasy points to opposing quarterbacks (244.20, or 15.3 per game). They also, however, played only seven of 16 games against quarterbacks who finished among the top 20 at the position in fantasy points and only three against top-10 passers. Totaling these, the Lions wound up facing the fourth-easiest schedule against quarterbacks of any NFL team last season and easily the easiest such schedule of any NFC squad. Adjusting for this schedule -- comparing opponents' seasonal averages to the team's fantasy point total allowed in each game -- the Lions actually finished 20th against quarterbacks, or seven spots worse than their raw fantasy point total allowed, illustrating that they were a below-average unit.
That's where the "Matchups Map" comes in. Each week, I'll provide a schedule-independent method to determine strength of positional matchups using the most recent, relevant data. Check back for updated numbers each week, including matchups highlights at each position -- both favorable and unfavorable -- based on those statistics. For these purposes, we will use PPR (Point Per Reception) scoring, though I have analyzed this data for both PPR and non-PPR and have found that the rankings would scarcely change (if at all). These do, therefore, apply to both scoring formats.
For Week 1, the maps include two measures: The first, "Rk," is my personal ranking of how favorable/unfavorable I consider that positional matchup; the second, "Adj. FPA," reflects how far above or below a player's average that defense held opponents at that position. For Week 1, 2017 full-season data is used for the latter, so take those with a grain -- or several grains -- of salt. Beginning in Week 4, we'll use 2018 data (three weeks in the books at that point), and starting in Week 6, we'll use the most recent five weeks.
Finally, a caveat: Remember that matchups are only one ingredient in my rankings formula. Not every favorable matchup should be exploited; not every unfavorable matchup should be avoided. To get the most complete recipe for whom to start and sit, consult my weekly rankings.