Finding 2020's League-Winning Fantasy Wide Receivers Through Season Simulations
We've all heard by now that fantasy football seasons aren't won in the draft but that they can be lost in the draft.
And that's usually true, anecdotally. If your first three picks get hurt or suck, you're gonna have a bad time. If you hit on some late-round picks, you're gonna have a good time.
Historically, one way to start off your draft with a safer investment is to draft receivers instead of running backs. This is due to the fact that early receivers typically pay off with big fantasy seasons more often than early running backs do.
There's a whole drafting philosophy around it called ZeroRB, and it makes sense when you compare historical ranges of outcomes of receivers (which is what I'm looking at here) and running backs (which I did already).
Now, receivers are typically easier to project than running backs are, but there's still variance involved here. Lots of it. There's touchdown variance, injury variance, and positional shuffling variance in the sense that not a whole lot separates a wide tier of receivers once we get past the top six or so. It's usually really tight.
And that's why I'm simulating the season 10,000 times to see which specific receivers pop with high-end upside and to see what we can learn from the process.
Projections, Simulations, and Historical Precedent
I discussed this in my quarterback article and running back piece so far, but it's vital that we discuss it again here. Projections are great to build, but no projections are perfect, and along with projections, we can easily be off base even if we're being cautious.
A very specific receiver who is plaguing me this year is A.J. Brown. Last season, Brown posted a league-best 12.51 yards per target among 79 receivers with at least 50 targets (only 11 other players since 2000 held even a 12.0 yards per target rate on at least 50 targets). The 75th-percentile yards per target rate in that 2019 sample was 9.31 yards, so that put Brown 3.2 yards per target better than an already good season.
Regression is likely coming -- not only for Brown but also his quarterback -- but to what degree? Put him at a still-great 9.31 yards per target compared to 12.51, and that's a change of 320 yards (or 32 fantasy points) over 100 targets. If Brown winds up a tinge below the still-elite tier, and we're looking at roughly a 40-point drop or more.
You can see why it's a little tough to pinpoint the "correct" yards per target rate for someone such as Brown. There's no correct option, but the only incorrect one is to put down 12.51 next to Brown's name for 2020 again and assume that's going to stick. Do this exercise for around 100 receivers, and it gets maddening -- and we're never exactly right about rates anyway.
That's why I've just started preferring using ranges of outcomes from median projections to look at upcoming seasons.
So that we're clear, I use numberFire's Net Expected Points (NEP) data and average target depth to project yards per target and catch rate, as this has proven to be more fruitful when projecting out future rate stats.
Using this, I have my median projections for this year -- which I audit against consensus projections and ranks to check for any significant outliers -- and can now apply historical ranges of outcomes to them to see how the season may play out.
And, yeah, yeah, some players are more volatile than others. We should know that, naturally. The ways receivers reach high-end upside historically is through elite volume or elite efficiency from either a yards per target (or catch) rate or a touchdown rate. That means: we want a ton of targets or some outlier rate stats. Finding those in play for top-five target numbers is a hell of a lot easier. Players with higher-end target shares and touchdown rates (due to their offense/quarterback) are tweaked.
One final note: here are historical hit and bust rates for receivers, bucketed by average draft position tiers (data comes from MyFantasyLeague). If the projections and historical ranges are in the ballpark, the sims will look something like this (they do).
| Finish Tier Odds by ADP (2013-2019) |
Top-6 Result |
Top-12 Result |
Top-24 Result |
Worse Than 36 Result |
|---|---|---|---|---|
| WR1-6 | 54.8% | 66.7% | 78.6% | 9.5% |
| WR7-12 | 16.7% | 38.1% | 64.3% | 21.4% |
| WR13-18 | 9.5% | 21.4% | 57.1% | 28.6% |
| WR19-24 | 11.9% | 26.2% | 52.4% | 33.3% |
| WR25-30 | 4.8% | 19.0% | 28.6% | 64.3% |
| WR31-36 | 2.4% | 9.5% | 19.0% | 64.3% |
Top-6 receivers in terms of average draft cost (treated here as the top guys in projected fantasy points) end up being top-6 performers more than half the time and are top-12 assets two thirds of the time -- plus top-24 players nearly 80% of the time.
Things taper off from there a bit, and it's not a complete dice roll that the next three tiers are good. Tiers 5 and 6 (receivers outside the top 24) are low-upside plays, historically.
Okay, now, no more dawdling. Let's sim this puppy out a bunch and see what happens.
Simulation Results
It wasn't too hard to get the projected ranges to replicate historical ranges, which is a good thing. That's how it should have worked. That lets me know that I'm on the right track.
As usual, you can disagree with my projections. It's fine. Just know that someone projected for roughly 100 fewer fantasy points than the WR1 (usually around the WR30-WR36 range) just don't often turn in top-12 seasons. It's fine if that's the only thing you get from this. Being realistic about what happens historically.
The table here is sorted by the odds a specific player finished as a top-12 receiver in the simulations.
| Player | Top-6 | Top-12 | Top-24 | Worse Than WR36 |
|---|---|---|---|---|
| Michael Thomas | 63.0% | 78.5% | 90.9% | 4.5% |
| Davante Adams | 53.6% | 72.0% | 87.1% | 6.4% |
| Julio Jones | 51.1% | 69.6% | 86.7% | 6.6% |
| Chris Godwin | 48.6% | 68.8% | 86.9% | 6.1% |
| DeAndre Hopkins | 43.3% | 61.9% | 80.7% | 11.2% |
| Tyreek Hill | 40.1% | 60.1% | 78.9% | 11.5% |
| Mike Evans | 31.1% | 54.3% | 78.0% | 11.1% |
| Allen Robinson | 25.7% | 47.7% | 74.2% | 13.1% |
| Kenny Golladay | 19.6% | 40.0% | 67.7% | 18.0% |
| Odell Beckham | 17.8% | 39.9% | 69.3% | 15.4% |
| DJ Moore | 15.3% | 37.2% | 69.4% | 14.5% |
| Amari Cooper | 13.7% | 34.9% | 67.7% | 15.7% |
| Adam Thielen | 12.3% | 31.9% | 63.4% | 18.6% |
| JuJu Smith-Schuster | 14.2% | 31.5% | 60.2% | 23.0% |
| Courtland Sutton | 13.4% | 31.3% | 59.0% | 24.3% |
| Robert Woods | 11.0% | 31.2% | 64.3% | 17.5% |
| AJ Brown | 13.4% | 30.5% | 57.2% | 26.1% |
| Calvin Ridley | 10.7% | 29.8% | 61.8% | 20.6% |
| Tyler Lockett | 12.3% | 29.4% | 57.9% | 23.9% |
| Cooper Kupp | 11.5% | 28.8% | 59.4% | 22.3% |
| TY Hilton | 10.1% | 26.8% | 57.0% | 24.4% |
| Keenan Allen | 8.9% | 26.4% | 59.4% | 20.9% |
| DeVante Parker | 9.4% | 26.1% | 57.2% | 23.6% |
| Stefon Diggs | 7.4% | 21.8% | 50.6% | 29.8% |
| DK Metcalf | 7.2% | 19.9% | 46.1% | 34.9% |
| DJ Chark | 5.4% | 18.8% | 51.2% | 26.1% |
| Terry McLaurin | 4.9% | 18.6% | 51.0% | 26.2% |
| Julian Edelman | 4.4% | 17.2% | 49.9% | 27.4% |
| Deebo Samuel | 3.7% | 16.3% | 48.0% | 29.3% |
| Jarvis Landry | 3.5% | 15.6% | 47.1% | 29.0% |
| Marvin Jones | 2.7% | 11.6% | 37.6% | 39.0% |
| Sterling Shepard | 1.6% | 9.2% | 36.1% | 38.0% |
| Tyler Boyd | 1.7% | 8.9% | 35.1% | 39.3% |
| Michael Gallup | 1.7% | 8.6% | 34.8% | 39.1% |
| AJ Green | 1.6% | 7.9% | 32.9% | 41.2% |
| Marquise Brown | 0.8% | 6.0% | 26.7% | 50.0% |
| Will Fuller | 0.7% | 4.3% | 23.0% | 54.1% |
| Brandin Cooks | 0.5% | 3.3% | 18.9% | 59.7% |
| Diontae Johnson | 0.5% | 3.3% | 19.7% | 56.4% |
| Christian Kirk | 0.3% | 3.1% | 20.0% | 56.5% |
| Jamison Crowder | 0.2% | 2.4% | 18.7% | 55.9% |
| Golden Tate | 0.2% | 1.9% | 16.0% | 61.6% |
I'm not alone in this, but putting Michael Thomas in a tier of his own does plenty to boost his odds of a top-six season. Then from there, it's a pretty tight tier with Davante Adams, Julio Jones, Chris Godwin, DeAndre Hopkins, and Tyreek Hill, who are the WR2 through WR5 in Bestball10 12-team ADP. (Again, though these are my projections, they're pretty on par with consensus.)
To be transparent, their top-six rates are probably a smidgeon too low, but that's okay. Mostly everything checks out, and we're also projecting all these receivers for 16-game seasons as the baseline. Injuries to players in the second and third tiers, for example, would reduce their top-six rates, but we don't see receivers step in for injured studs and simply become studs.
Moving on. There's a clear tier drop before we get to Mike Evans (and Allen Robinson, for me), and then it's a lot of similar projections and ranges. This actually reflects history quite well. While there are drops from WR7 down the list in terms of thresholds, they're fairly flat and steady.
Players such as JuJu Smith-Schuster or Robert Woods aren't that much less likely to be a top-12 receiver than players a bit higher on the list, such as D.J. Moore and Amari Cooper. I think this is always a key takeaway from studying receiver probabilities: the top guys are usually the top for the reason, but then it's a steady -- not precipitous -- decline.
Now, the table also shows that plenty of receivers have chances to have top-12 seasons but that it does level off after WR24 or so. We've seen from the historical results table that this tracks. Those fifth and sixth tiers (WR25-30 and WR31-36) fall outside the top 36 well more than half the time at season's end.
This is another key learning point.
Yeah, wide receiver is deep, and it especially feels deep this season, but take your pick of your favorite Tier 5 or 6 receiver (players such as D.J. Chark, Terry McLaurin, Deebo Samuel, T.Y. Hilton, Jarvis Landry, Michael Gallup, Tyler Boyd, Will Fuller, Marquise Brown -- you get the idea). History shows that this is not a super bankable tier.
And it makes sense. Some of those guys have questionable offensive situations (namely Chark and McLaurin), some are not their team's lead receiver (Gallup and potentially Landry and Boyd), some are in run-heavy offenses (Samuel and Brown). You get it. These players won't be in the mix for league-leading volume, and unless they notch the high end of their outcomes, they can be just lukewarm fantasy assets with nothing much separating one from the next.
Again, they may still be usable WR3s, but history shows that this isn't generally a league-winning tier and that the true differentiating pieces at receiver are off the board by this point. What, then, separates one guy from the next? Preference, mostly. It's merely a teachable moment that we won't hit this tier significantly often because it's a pretty similar group overall. They all have some question marks.
Plus, the natural way that receivers work is inherently different than with running backs, who attain their value from volume. Two to three receivers can be relevant on a team. If a team has two running backs, we don't really have one fantasy relevant back (at least when seeking any modicum of upside). These are the types of receivers we're looking at in this range.
Now, I will say that certain players in this range have different ranges of outcomes due to touchdown or volume upside. I'll likely ramp these up more for personal use, but for these purposes, I wanted to keep this close to historical rates.
In my projections, touchdowns are tied to quarterback play and yardage, which is tied to volume. Players with higher volume and (therefore, generally) higher touchdown projections project with some higher upside than their counterparts.
Breakout receivers often come from ambiguous depth charts in good offenses, so that speaks volumes for a few players in this general tier: Robert Woods/Cooper Kupp, D.K. Metcalf/Tyler Lockett, Brandin Cooks/Will Fuller, A.J. Green/Tyler Boyd. You get the idea. It's usually clear which receiver we should prefer, but there's probably value in settling for the cheaper option.
Piecing It All Together
Based on the historical results and current projections, we should still feel very safe in drafting receivers early -- and often. The first round or two is full of elite fantasy assets this season, players whose volume and role aren't questioned in the least.
From there, it's a steady drop until we get to the shrug range where we're kind of just closing our eyes and hoping we hit the players who hit the higher end of their still somewhat capped upside. The info shows that the WR1 isn't likely to come from the sixth round.
Pairing that with the understanding that RB1s typically come from the first round, you have to pick your poison: take a big risk to find one of the few game-changing running backs or play it safer with a top-tier receiver. This year, there are, probably, five backs you feel great about and probably six receivers, so just about any first-round pick will make itself for you.
From there, with the knowledge that the next round or two is crucial for still getting realistic shots at top-6 and top-12 performances from backs and receivers, we should just be hammering those positions.
Given that both positions historically taper off in the middle of the draft and that the mid-round quarterbacks look to have league-altering upside, I'm prepping for a bit of a cookie-cutter draft season so far: running backs in the first half of the first round and receivers in the back end, backs and receivers from there, and then actually taking a stab with the second tier of dual-threat quarterbacks before they're off the board.
Unless, of course, I find that tight end upside shows that it's clearly worth drafting Travis Kelce or George Kittle early. We'll look at that next.
