Everyone has been there before. After the draft, all of the players gather around the draft board, commiserating over picks and offering ridiculous trades, when someone says “I will never draft DeAndre Hopkins, he killed me last year.” Usually the other people gathered around nod sagely and murmur sympathetically, and maybe the person who drafted Hopkins makes some half-hearted defense of the pick by saying “he was a stud two years ago” or “his quarterback can’t possibly be as bad as Brock Osweiler” and then the conversation will move on. The entire time, everyone in the room ignores the fact that “I won’t draft him, he killed me last year” is a pretty dumb way to run your fantasy football team.

Look, if you don’t want to relive the emotional trauma of watching DeAndre Hopkins catch only 52 percent of his targets and completely disappear from games, that is totally fair. Also, there are plenty of really good receivers going right around Hopkins in drafts, so unless everyone in your league owned Hopkins last year, the odds are you aren’t passing on Hopkins in favor of a complete bum. Fantasy football is supposed to be fun, so if you want to pass on a player because you have emotional scars, he plays for a rival team or you think he is a bad person, knock yourself out. Just know you are using a suboptimal strategy. Because, getting back to my point, carrying the past with you is a bad strategy.

DeAndre Hopkins could, of course, be just as bad in 2017 as he was in 2016, but he could just as easily have a huge season like he did in 2015. After all, Houston’s revolving door of bad quarterbacks didn’t hold Hopkins back in 2015, and at 25 years old, Hopkins is firmly in his prime. While a smart projection for Hopkins probably falls firmly between the two extremes of 2015 and 2016, it would be an error to ignore either possibility. Fantasy football players tend to look at the most recent performance and assume that is the new norm when that is often not the case. If Hopkins had been awful in 2015 and great last season his ADP would likely be significantly higher, even though his outlook for 2017 would largely remain the same.

Hopkins is probably the best example of why overrating last year’s performance can be a mistake, but there are plenty of others.  DeVante Adams probably isn’t catching 12 touchdowns next season, especially considering he had four TDs in his first two seasons combined. Matt Ryan had career-highs in yards, touchdowns and yards per attempt last season to go with a career-low in interceptions, but the eight previous seasons when he was just a solid QB1 still count. While 2017 may be the new norm for Ryan, it is not a fait accompli.

While overcoming recency bias is important for evaluating individual players, it is even more important in trying to determine overall fantasy strategy. 2014 was a banner year for rookie wide receivers, with Odell Beckham, Mike Evans, Kelvin Benjamin, Sammy Watkins and Jordan Matthews all going for at least 872 receiving yards and six touchdowns. That resulted in a number of rookie wide receivers getting overdrafted in 2015, because as it turns out, rookie wide receivers don’t tend to have huge seasons, and 2014 was an outlier. Just like Hopkins and Allen Robinson being terrible in 2016 was probably an outlier as well.

Fantasy players also tend to believe because they did things a certain way in the past and got a certain result, the same will be true in 2017. That is rarely true. Every season is different, and so is every fantasy league. Just because going zero RB blew up in your face doesn’t mean there was anything wrong with the strategy, or even the way it was implemented. It could have been anything from your evaluation of the players to just dumb luck. The guy who drafts Aaron Rodgers and two of his receivers every year because it worked like a charm in 2011 and 2012 probably missed the playoffs in 2013 and 2015.

While fantasy players obviously use the past to inform their opinions on players’ skills and situations, it is important to understand how much the NFL changes from year to year. Proper strategy should be dictated by your evaluation of the players for this season, not on what worked—or didn’t—in past seasons. After all, the past can no longer hurt you, unless you let it.