If you want a laugh, go back and look at FantasyAlarm’s 2016 NFL Bold Predictions. Looking back, I’m actually really happy with how my picks turned out, even though I had Kevin White with 80+ receptions, I picked the Bears as my sleeper team and I had the Bengals and Packers meeting in the Super Bowl.
Some of the other highlights from last year’s predictions: the Rams and Jets were picked for favorite fantasy defense, and also for sleeper team. Todd Gurley was a popular pick for leading rusher, and more than one person chose DeAndre Hopkins as the leading receiver.

The point of this exercise isn’t to criticize anyone’s preseason predictions (except for my own), but rather to point out how difficult it is to predict anything in the NFL from year to year. Cam Newton should have easily been a top-10 fantasy quarterback in 2016 after he was the top fantasy QB in 2015. Conventional wisdom held the Panthers, Jets and Rams would all continue to field top defenses after they kept most of their players, and most believed Josh Norman going to Washington would make that defense better. None of those things came to fruition last season, and if you entered your fantasy drafts treating them as fait accompli, you likely got burned.

All of this brings us to the notion of strength of schedule. Like all of the fantasy myths I discussed earlier in the draft guide, relying on strength of schedule to construct your draft board can be a huge mistake. Strength of schedule can be used in a lot of different ways in the preseason, and I think it is useful to take a look at the flaws in each individual argument.

My favorite example of the strength of schedule myth blowing up in the faces of fantasy players is Joe Flacco in 2016. Flacco was a popular target among fantasy players looking for a cheap quarterback to play Weeks 1-4 while Tom Brady served his suspension. Flacco faced the Bills, Browns, Jaguars and Raiders those four weeks, and most fantasy players agreed this was a favorable schedule. They were right about the Browns and, to a lesser extent, the Raiders. Baltimore’s first four opponents finished 20th, first, 24th and 11th in fantasy points allowed to quarterbacks last season. In total that probably was a favorable schedule for Flacco, but likely not as favorable as people thought when they drafted him. In those four weeks, Flacco had 12.02, 16.18, 11.76 and 22.02 fantasy points, which ranked him 23rd, 19th, 21st and sixth at his position.

The best part about this example is it was the first quarter of the season. If there is any part of the season we should be able to predict, it should be the first four games. Later in the season when injuries take more of a toll, it is understandable that some teams will be better or worse than we think. We don’t know which team will lose its leading pass rusher or suffer multiple injuries in the secondary, but we know most teams will have most of their players early in the season. Of course, that knowledge didn’t help us see Carolina’s slow start coming. We all thought Denver’s running backs had a difficult matchup against Carolina in Week 1, but they combined for 34.10 fantasy points. Even with his strong start, I doubt anyone thought Matt Ryan would throw for 503 yards and four touchdowns against Carolina in Week 4. And yet, by the end of the season, Carolina’s defense was formidable once again.

If we don’t have a handle on which matchups are good in the first quarter of the season, we certainly don’t know who has a good matchup during the bye weeks, or the fantasy playoffs. Every year I see people in my fantasy leagues reach for a bad quarterback because he has an allegedly good matchup when their starting quarterback is on a bye. I can already picture Russell Wilson and Dak Prescott owners drafting Jay Cutler because he faces the Falcons in Atlanta Week 6. Of course, just because the Falcons were the second-best matchup for fantasy quarterbacks in 2016 does not mean they will be in 2017. Don’t believe me, just look at the New York Giants, who allowed the fewest fantasy points to opposing quarterbacks in 2016 after allowing the most passing yards in the NFL in 2015.

I have to admit, there is one strength of schedule argument I found compelling, at least on its surface, and that is avoiding wide receivers who face a lot of difficult secondaries. This is one of the best arguments against Pierre Garcon, aside from him having a terrible quarterback in a bad offense and also not being that good. The 49ers face the Panthers, Seahawks, Rams and Cardinals the first four weeks of the season and later play the Redskins and Jaguars. All of those teams have potential shutdown corners they can put on Garcon, and they have no reason not to do so because the rest of the receivers are so bad.

The thing is, cornerbacks are nearly as unpredictable as everything else in the NFL. Do we really believe James Bradberry is a shutdown corner after one pretty good NFL season? Are Richard Sherman, Josh Norman and Patrick Peterson still as good as their reputations? It certainly didn’t look that way on film. And even though it is a bit counterintuitive, great cornerbacks can become mediocre in the blink of an eye in the NFL.

Joe Haden was a shutdown cornerback—arguably the shutdown cornerback—in the NFL as recently as 2015. He was repeatedly beaten like a Cherokee drum last season. If Josh Norman and Richard Sherman were such impossible matchups, would the Panthers have released Norman? Would the Seahawks have attempted to trade Sherman?

You might attribute Haden’s struggles in 2016 to playing on a terrible defense that generated little pass rush, which only further illustrate my point that cornerbacks and secondaries should not necessarily be feared. We have seen Aqib Talib struggle at times when his line can’t get much of a rush, and most secondaries are that way. The same goes for Patrick Peterson. It is way too early to start worrying about who is going to cover your favorite wide receiver in any given week.

The bottom line is, you should construct your draft board based on your evaluation of each players’ talent and opportunity. I believe Pierre Garcon could be a target monster in San Francisco, and I believe he actually has the talent to take advantage of the situation, which I never thought about Torrey Smith. The rest will most likely take care of itself.

I feel like even if we were much better at identifying which schedules are good and which ones aren’t, we would still overrate how important that actually is. The Browns gave up more fantasy points to opposing quarterbacks than any team last season, and Flacco only managed 16.18 fantasy points against them back in Week 2. In the end, it comes down to one thing: Joe Flacco isn’t actually a good fantasy quarterback. Having a good schedule is better than having a bad one, obviously, but how much better? At what point would you rather have a good quarterback with a bad schedule rather than a bad quarterback with a good schedule? The smart thing is to always take the better player, since we are much better at evaluating players than schedules.