The Gut Check No.499: 12 Lessons From 2019 Fantasy Projections

Photo: USA Today Sports

I posted my first incarnation of 2020 fantasy rankings (PPR and non-PPR) early Memorial Day morning. Comments for each ranked player will be forthcoming during the next 10-14 days. These rankings are based on projections that I create and maintain during the preseason.

I do not post projections at Footballguys.com. We have a fine team of writers who do player projections at the highest level, and I encourage you to delve into their work.

Last year, I partnered with Rookie Scouting Portfolio site contributor and high-stakes fantasy competitor Dwayne McFarland to do a projections series podcast for Matt Waldman's RSP Cast. We covered each team in the league and shared our projections for each offensive player, the craft behind projecting players, and why we had similarities or differences with specific players.

While we are discussing the possibility of doing it again, it seems timely to share with you 12 lessons about projecting players with examples from last year's work. I'll list multiple players within each lesson who exemplify the points below as well as my projected performances versus their season outputs.

1. Rely on median Outcomes When Projecting performance—Ceiling outcomes should be rare

Examples: Lamar Jackson and Dak Prescott

Last year, I thought Jackson and Prescott were excellent values based on their average draft positions, and I recommended them as starter-value quarterbacks at a quarterback-by-committee or reserve price. Here are my projections for Jackson and Prescott versus their actual production:

I had excellent accuracy with projecting Jackson's attempts, completions, passing yards, interceptions, and rushing attempts, but I was too low with his completion percentage, yards per attempt, touchdowns, and rushing yards. I rated Jackson as my 15th quarterback before the season began, and he finished as the best fantasy passer of 2019 because of the above-mentioned differences.

That's alright with me. Having strong showings in fantasy analyst accuracy contests for rankings and projections used to be nice when we participated in them in the past, but the most important thing about projections is identifying value within figures that don't account for the player's ceiling of potential.

Every projection has a floor, a midpoint, and a ceiling. Only in exceptional cases do I recommend using a player's ceiling as his projection although it's a great exercise to project all three values for each draftable player.

Counting on everything to go right lacks realism—offensive linemen miss blocks, receivers drop passes, a key talent suffers a significant injury, quarterbacks misread the coverage, or the ball slips from their hand as they are in the middle of their release. Whatever it is, ceiling projections, which usually don't account for typical things that can and will go wrong force you into positions where there's no wiggle room to operate.

This was true with projecting Jackson. After watching him run between and around defenders as a rookie, even the most casual fan could see the realistic possibility of Jackson's ceiling. Anyone could imagine him earning rushing totals that rivaled the best quarterback performances in NFL history.

Jackson earned 79.7 rushing yards per game from Weeks 11 through 17 in 2018. That per-game average projected to 1,275 rushing yards if he maintained that rate for an entire season.

Based on Jackson's skill, the few big hits he takes as a runner, the scheme, and the quality of the offensive line, there were too many factors to discount Jackson's legs as a massive source of fantasy points. But how high should fantasy owners have gone when projecting this ceiling?

In hindsight, the simple answer is that I had room to project 1,275 yards, and that total would have been 39 yards away from an exact match to the real outcome. For me, it comes down to having self-awareness as a fantasy player.

Going with that maximum value is the stuff that wins projection accuracy contests but realistically, it's wholly unnecessary. Jackson's ADP was nowhere near his actual performance outcome.

When you write down numbers for your projection that reflect a much higher ranking than a player's ADP, you have to be honest with yourself about how you'll react to seeing this figure. Will you begin to draft that player based on that number to validate what you're seeing on paper or can you cope with the cognitive dissonance of seeing a much higher figure on the page for a player you have listed much lower?

For me, I'd rather see the median value of the player and know that there's a potential ceiling that's realistic enough to consider him a draft-day value at a lower price. If my projections reflected Jackson's ceiling—a ceiling I thought was possible enough that I recommended him at every opportunity based on his ADP—then I know myself well enough that I'd be prone to drafting Jackson earlier than necessary in part to the leaning too hard on the projections while forgetting the data is his ceiling.

The same could be true for Jackson's passing touchdowns although I had no analysis that led me to believe 36 touchdowns would be within his ceiling of expectation. If there were compelling indicators as there were for Jackson's rushing yardage, then I'd still maintain that Jackson's lack of history as a high producer at the NFL level and his ADP made it best to use a median projection.

I like to use projections and rankings as complementary processes. If I project the median outcome for a player whose upside is compelling enough to consider more, it's easier to see how he fits in a rankings format, because I can always notate in the ranking that his ceiling is high enough to build a compelling draft plan that involves him.

If I project the ceiling then I'm likely ranking him commensurate to that projection or the way I'm noting how long I can wait to draft him becomes more complicated. You don't want to overanalyze if/then selection scenarios on draft day with so many variables in play.

This lesson applies to my projections for Prescott but with a twist. Projecting a yards-per-attempt average above 8 yards would be a career-high for him.

I also didn't heed this lesson with his completion percentage when I estimated a career-high 69 percent or a bests in passing yardage and rushing touchdowns. Why go there?

Because I extrapolated his 2018 production after the Amari Cooper trade and the talent upgrade made an increase compelling. Prescott had never worked with an elite athlete who was also a complete route runner at the position.

It's why I decided it was worth incorporating the ceiling value for passing yards, completion percentage, and rushing touchdowns. In hindsight, I should have maintained the median projection for touchdowns because the outcomes are harder to predict whereas increases to yardage and completion rates had more cause-effect logic.

My projection of Prescott for 2019 still landed him as a value in my rankings. The value wasn't nearly as much as it was for Jackson because I projected closer to Prescott's ceiling. Still, it made it easier for me to see Prescott as a value I could let fall to me.

2. Don't Punish Your Process Due to Injuries

Examples: Patrick Mahomes II and Matthew Stafford

It's a simple lesson that many competent fantasy players neglect. In most cases, predicting injury is a foolhardy exercise. And, it's a lame supporting argument for the regression of a top performer from the year before.

When extrapolating Mahomes' and Stafford's seasons to a full 16 games, we see outcomes much closer to my projections. When extrapolating, I removed the Broncos game where Mahomes got injured from the equation so I could use a per-game average for the categories below based on full games.

As you can see from the results, this isn't some ploy to prove that my projections were correct. While my projected yardage for Mahomes and Stafford were close to the extrapolated "actual," I went for the Mahomes' ceiling with his passing touchdowns.

Although I underestimated Stafford's touchdown totals, he fit the mold of the quarterbacks I touted as mid-to-late picks. Stafford had the realistic potential of possessing a greater upside than his listed projections but it was best to project his median value.

Considering that Stafford's career-best yards-per-attempt prior to 2019 was 7.87 in 2017 and best touchdown total of 32 came in 2015, and they happened under different offenses and staff, I wasn't about to project career-best numbers in those categories.

While projecting Mahomes' touchdowns close to his previous season's record-smashing totals didn't prove correct, there was a logic behind the decision. The continuity of the offense and staff that set the stage for Mahomes to earn 50 touchdowns the year prior led to the possibility that Mahomes was setting a new standard for quarterback production in the way he had as a thrower.

Ultimately, the wisest lesson for this situation is to honor the idea of regression to the mean when a player had a season that was in record-breaking territory. Still, if you're going to push the envelope with the odds, this situation had enough compelling reasons for consideration.

One could even argue that injuries to Tyreek Hill and Sammy Watkins placed limits on Mahomes' ceiling that wouldn't have been there if healthy and playing those seven combined games that they missed. These were players who have a direct impact on completion, touchdown, and yardage totals for Mahomes.

While a rough estimate—and I emphasize, "rough"— if we take Hill's extrapolated average of 84 yards and .69 touchdowns per game in 2019 and determine the difference between that average and Mecole Hardman's production in Weeks 2-5 and Week11, Hill arguably earns another 161 yards and 2 touchdowns during this period.

If we match Byron Pringle's usage during Watkins' three-game absence with Watkins' extrapolated per-game average of 52 yards and .25 touchdowns, Watkins arguably earns a net extra of 29 yards. Add this 189 yards and 2 touchdowns to Mahomes' extrapolated totals and he's at 5,056 passing yards and 33 scores—80 yards above my projected yardage but still 14 touchdowns below my scoring total.

Once again, we learn that predicting touchdowns is more difficult than projecting yardage and if you're going to roll the dice with projecting a player's ceiling, touchdowns are at the bottom fo the list. Still, you can see that injury shouldn't be a reason that gets in the way of a compelling projection.

Unless that player has never come close to playing a full season during his career, there are stylistic behaviors with that player that invite physical play and awkward contact, and you're still projecting a full season from him, projecting injury for every player is a variable to avoid.

3. Offensive line Play Is Vital for the Ceilings of POcket Passers And Inside Runners

Examples: Matt Ryan and Devonta Freeman

Common sense, right? Most of these points are, and they are worth a reminder or additional validation. The Falcons and Rams are excellent examples of teams whose quarterbacks and running games suffered during various points of the season.

The most convenient parties to blame in these situations are the quarterbacks and running backs. Jared Goff is a fraud. Todd Gurley and Devonta Freeman are washed.

I'm careful (although not always successful) about avoiding these conclusions about players. I've been doing this long enough that I've not only seen players rebound after the masses wrote them off but I'm also aware that labeling a player negatively can generate an internal bias about his future prospects without regard to the actual machinations of football.

When you do your best to analyze how football works on the field, the more open you become to the factors that make the game a team sport and not reduce your analysis to a simple credit/blame game that media conglomerates distill to fans because that's what earns the easiest and highest volume of clicks.

I missed wildly on Devonta Freeman's 2019 projections. I predicted Freeman would produce as the 12th-ranked fantasy runner in Dirk Koetter's balanced scheme. Freeman finished 26th among fantasy backs, and that standing doesn't do justice to how disappointing he was as a producer.

Even if I extrapolated Freeman's totals to 16 games based on most of the 3 games that he missed, we're still looking at a 217-carry and 781-yard rushing performance in 2019 that falls well short of my preseason expectations. Worth noting is Freeman's final five games. Extrapolate that production for a season and he earns 246 attempts, 912 yards, and 6 touchdowns—it's still 229 yards and 5 scores short of my projections but markedly better than Freeman played for most of the year.

What was the difference? The right side of Atlanta's offensive line. Rookie right guard Chris Lindstrom didn't see the field until Week 14 and it's the point where Lindstrom and fellow rookie right tackle Kaleb McGary jelled as a run unit during those final four games.

Before that, Jamon Brown struggled mightily throughout the year at the right guard spot. Injuries repeatedly forced Atlanta to shuffle its unit. Once the unit got healthy between Weeks 11 and 13, the ground game earned more traction down the stretch.

I've been doing the weekly recaps for Atlanta for over a decade, Freeman may not be worth what other teams want to pay him based on his age and recent injury history, but he still flashed competency as a runner despite a down year. He's "washed" because the NFL market has a surplus of younger and healthier backs but not because he can't run anymore.

Convincing people this is the case is difficult because football has a context-driven game of complex layers presented as a simplistic event—even when the media conglomerates present data. Freeman had little room to run early in the year. His offensive line routinely missed blocks.

This was also true in the passing game. Ryan's 7.25 yards per attempt in 2019 was his lowest mark in seven years. Ryan's highest yards per attempt was in 2016—a glorious 9.25—when Atlanta solidified its offensive line with a pair of late-season acquisitions and Freeman enjoyed his career year with 1,079 yards and 11 touchdowns—ranking sixth overall among fantasy runners.

I underestimated the health of the line late in the preseason and overestimated Ryan and Freeman's production. Both players rely heavily on time to create between the tackles and the lack of quality guard and tackle play on the right side limited Ryan's passing efficiency.

Although Ryan completed 66 percent of his passes (I projected 66.1), Ryan had to attempt shorter throws and didn't get the maximum yardage possible from plays because of pressure. Atlanta allowed 48 sacks last year—tied for third in the league with the Arizona Cardinals and the mobile Kyler Murray.

A closer look at the sack leaders reveals that NFL passers who can create with their legs still produced at a high rate despite taking a high number of sacks. Deshaun Watson and Russell Wilson are first and second in sacks allowed last year. Murray as third and Aaron Rodgers and Josh Allen are seventh, and Carson Wentz is twelfth. That's half of the fantasy QB1s from last year.

Jamies Winston had excellent fantasy production in leagues that don't penalize interceptions, but he's known for pressure-induced mistakes and he's essentially Matt Ryan who refuses to get rid of the ball fast and/or wisely. Both players need the time that Dak Prescott and Patrick Mahomes II earned last year as the remaining fantasy QB1s from 2019's list.

4. Banking on Touchdowns Is a Risky Projection Point

Examples: Aaron Jones, Jared Goff, Davante Adams, and Alvin Kamara

A common theme throughout many of the projections shared in this piece is that touchdowns are difficult to project. I was often close to the mark with attempts, targets, receptions, completion percentage, and yardage but vastly over- or underestimated touchdowns.

Here's a partial list of more players where touchdowns were the greatest difference between an accurate and inaccurate projection:

• Jared Goff: 22 touchdowns (36 projected)
• Aaron Jones: 19 total touchdowns (12 projected)
• Alvin Kamara: 6 total touchdowns (20 projected)
• Davante Adams: 5 touchdowns (14 projected)

There are two categories of players here: Those I underestimated like Jones, who had a career-year; and those, like Goff, Kamara, and Adams who were coming off career highs in the category. If there's an area where banking on a regression bears serious consideration, it's a player coming off a career-year with touchdowns.

In Goff's case, I thought McVay would adjust to what defenses did to his scheme and Goff would build on his production based on the weapons around him but the combination of Gurley's decline, loss of key linemen, and McVay sticking with his scheme derailed any hope of Goff building on his 2018 season. Of those three factors, the changes to the line were the most important for projections.

Even if Kamara didn't get hurt in 2019, he wouldn't have come close to his 18-score campaign in 2018. It's a reminder that a total of 10-12 touchdowns for a top running back is a reasonable ceiling in a balanced offense and expecting anything more is greedy and unrealistic. When a player of Kamara's acumen reaches paydirt this often during a year, expect a regression.

The same is true for a receiver of Adams' skills. His 13 touchdowns were a career-high and thinking he'd build on it was a foolish thought on my part. 10-12 scores for an elite receiver is a reasonable projection in most cases.

5. If you Have a Process, Don't Let Others (No Matter how KNowledgeable) Talk You Out of It

Example: Derrick Henry

Develop a process for analyzing your subject matter. If you are meticulous with your process, don't change that process until there are compelling reasons to do so that you've tested. These are the two greatest lessons I've learned as an evaluator of rookie talent.

Sometimes I forget the second lesson when it comes to projecting fantasy production. This was the case last year when I had 15 rushing touchdowns for Derrick Henry before I compared notes with my friend and colleague Dwain McFarland.

As mentioned earlier, McFarland is a high stakes player with a pair of top-three finishes in the Footballguys' FFPC contest in recent years. When Dwain and I did our aforementioned podcast series on player projection last year, he gave compelling reasons why Henry would not score anywhere near 15 rushing touchdowns or deliver 1,300 yards as a rusher.

I made the snap decision to change my stance on Henry, dropping his rushing production to 1,100 yards and 10 scores. My rationale for Henry producing at a higher level prior to my conversation with Dwain was his fit in a run-heavy offense as the only talent on the running back depth chart.

I would have earned the correct result if I stuck with my process. However, sticking to your process even if the results are incorrect is more important for your long-term development. I understand that many of you are here just for the answers to the test (and most of you have probably clicked elsewhere who fit this description) and this doesn't apply to you as much as someone who does their own projections and analysis.

For those of you doing the grunt work, you learn faster when you make mistakes rooted in your path rather than switching away from the path and toss the process aside. You lose your bearings this way. What I should have done is what I intend to continue doing as I move forward: If a running back is the load-carrier for an offense with a run-heavy identity, I'm not shying away from the upper limits until I see enough reasons to do so.

Dwain knows his stuff, but I know my path better than he knows my path. If you're traveling toward knowledge the way I am, you have to learn to listen and take note of compelling thoughts that you can monitor while maintaining your own process rather than reacting and changing results without regard to your work.

6. Be willing to Take Stands Where you Make the Player Prove It

Examples: Leonard Fournette

After a dismal 2018 campaign, Fournette became a player I didn't want anything to do with from a fantasy perspective. Jene Bramel has documented Fournette's ankle issues and described Fournette as a player who can be good for a span of seasons until injuries cause him to fall off a steep cliff.

After last year's debacle, I wondered if that time was coming earlier than expected. I love Fournette's talent and running style, but didn't trust his approach to the game or his health. I projected 878 yards on 266 attempts.

The only thing I was right about was the attempts (Fournette earned 265). Fournette earned 1,152 rushing yards and nearly twice the receiving yards that I projected (522 to 288).

In this case, the projection-versus-the-reality didn't matter: I didn't believe in Fournette's health and preparation—even after reading stories that indicated he took a serious approach to the offseason.

Now that I've seen Fournette prove it, I'm a believer. If you project player performance, you shouldn't worry about having a handful of players that you're skeptical of them delivering and want them to show you otherwise before you revisit them with a more optimistic perspective. Fournette was that player for me last year.

7. Exceptional Players Make the Surrounding Talent Look Better Than it Is

Examples: Nick Chubb (and Marlon Mack as the corollary)

The NFL is filled with exceptional athletes. Many are in the 99th percentile of athletes playing football in the world. Still, the NFL has only a handful of exceptional players.

Saquon Barkley is an exceptional athlete with the potential to become an exceptional football player. I think he'll get there.

Nick Chubb is an exceptional athlete and an exceptional football player. He's teaching tape for the NFL as former NFL scout Dan Hatman and I reveal with an in-depth analysis of Chubb's work against stacked boxes where he's literally doing work that helps his linemen recover from missed blocks to make others during the same play.

Despite lacking a pair of capable tackles, working with an offensive staff that was out of its depth with developing a scheme that maximized the talents of its best players, and saddled with a quarterback who has clear issues managing and navigating a pocket, Nick Chubb still came within 46 yards of the NFL's rushing title.

Once again, I overestimated touchdown totals with another player but the touches, yards, and efficiencies were all on-point—and to be aggressively on-point with a player facing the issues described above, that's notable. The lesson learned here is that if you have a compelling rationale that the player does exceptional work, he can overcome certain flaws with surrounding talent.

The interior of Cleveland's offensive line was decent enough that the unit wasn't a complete mess. Chubb's skills to create space with an elite burst, excellent contact balance, rare cutting ability, and to do it all with efficiency and within the context of strong decision-making are exceptional.

Ezekiel Elliott is the other back in this hemisphere of play and he's had a superior offensive line. Yes, I know, there are some dissenters out there who claim the math says Elliott is overrated. I will hopefully have more about this subject later this summer but the short answer? I've seen a claim from a pretty astute source that those in the anti-Elliot camp were perpetuating a take based on an error the original work they were citing.

Chubb creates positive plays from negative situations and enhances the value of positive plays with his skill in ways that others do not. A good corollary for this point is Marlon Mack.

The Colts' 2019 starter earned respectable production for a team that primarily ran gap blocking schemes behind one of the best run-blocking units in the league. Gap schemes don't require the same degree of decision-making expertise and footwork for a running back as zone schemes.

Gap schemes rely more on the runner hitting the crease with conviction and intensity. Give a back with limited conceptual skills and vision a strong offensive line that can gap block, and he can produce, especially if he has speed and acceleration.

Contact balance and power aren't as important either, because the runner can generate momentum-based power and balance with the longer runways that gap blocks often provide. Mack is a runner who gets what his line blocks for him and with the occasional potential of pulling away from second- and third-level defenders thanks to his speed if the crease is wide and clean enough.

This made Mack easy to project:

Put Chubb behind this line and he threatens the 2,000-yard mark as a runner. Jonathan Taylor has some Chubb-like qualities to his game where he can create more than what's blocked for him. Once Mack is completely out of the picture—whether it's this year or next—Taylor's skills will allow you to up the yards per attempt average he earns as long as the offensive line stays intact.

8. Projected Point Totals Is More Important than Player Ranking

Example: Todd Gurley

Last year, I wrote that Gurley would be no worse than a fantasy RB2—a high-producing second back for a lineup. I had Gurley ranked eighth (335 fantasy points) among backs in PPR by late August/early September and fifth (280 fantasy points) in non-PPR formats.

Gurley was 12th in standard formats with 190.4 points and 14th in PPR formats with 221.4 points. Obviously, this wasn't an exact match and wouldn't even be worth touting without the context that most people were scared off by Gurley based on the offseason reports about his knees.

Even so, my projected point values came nowhere close to Gurley's actual performance. Although my projected ranking could be interpreted as a win—and if this were part of a rankings accuracy analysis, the analyst with a preseason ranking closest to the actual ranking would earn the most value for that player—if the point values are far off then you didn't determine the true value of the player.

Gurley might have been an RB1 for me but what if I took Gurley ahead of Derrick Henry because I projected more points for Gurley? Henry outscored Gurley by 79 fantasy points in PPR formats—5 points per game if both had played a 16-game season.

Rankings don't show you this projected point differential. A well-constructed set of tiers have the capacity to do so and can offer a terrific illustration of value among players within the top 50 players ranked on a draft board.

At the same time . . .

9. DON'T "PICK BY NUMBERS" DURING THE LATER ROUNDS

Example: Darren Waller

After the seventh round of a draft, you should start giving additional weight to a player's upside. The best but most time-consuming way to consider a player's upside is to do three projections for each player—his floor, median, and upside value.

Most fantasy drafters won't do this type of work. I don't but I know of people who do. At the very least, you should notate players you believe have a statistical ceiling that places them 2-3 tiers higher than their median projection.

These are the Lamar Jacksons, Dak Prescotts, and DeVante Parkers, and Darren Wallers of the NFL. Waller is a fine example because, until last year, he was an unproven player that Jon Gruden acquired from the Ravens and used as a situational contributor in 2018.

There was no doubt about Waller's upside: He's a big, tall, fast receiver with above-the-rim skills who converted from wide receiver to tight end. If used as a big receiver in the role of a move tight end, he could deliver yardage production on par with Jimmy Graham in his prime.

However, projecting Waller to achieve that upside without a prior history of him doing so is risky and unnecessary. Waller's average draft position was low enough that there was no need to operate according to his projected ceiling of potential.

At the same time, Waller's median value in my projections placed him within the same range as Dontrell Hilliard, Jordan Wilkins, and Ed Dickson. Wilkins may have starter upside as a talented reserve behind an excellent run-blocking offensive line but only Waller was expected to earn contributor playing time that would equate to starter volume.

If you pick by the median numbers because like most of us, you don't do upside and floor projections, then you're hiding the obvious potential value of the player that elevates his standing within your draft tiers. This is why you don't draft strictly by the numbers. The same is true for selecting free agents.

10. DON'T LET Past PROJECTIONS DRIVE WAIVER WIRE DECISIONS

Examples: Raheem Mostert and Tyler Higbee

• My projections for Mostert: 4 attempts, 26 yards, 1 target, 1 reception, and 5 yards.
• My projections for Higbee: 33 targets, 24 receptions, 193 yards, 1 touchdown.

Neither were draftable options on my board. Mostert earned 137 attempts for 772 yards and 8 scores as well as tacking on 14 catches, 180 receiving yards, and 3 scores in the passing game. Higbee caught 69 passes for 734 yards and 3 scores.

Both were strong fantasy starters by year's end despite not beginning the season as such.

Be open to change. The best way to do so if you do projections is either put away your preseason projections or continue updating projections as the season unfolds. Look for where the available points should be based on the existing scheme or the changes to the offense.

The 49ers have a prolific run game, so it was easy to plug Mostert into a lineup and expect at least committee production. The Rams morphed from an offense that leaned heavily on 11 personnel (one back and one tight end) before Week 10 into a unit that used a lot more 12 personnel (two tight ends and one back).

The savvy fantasy player adapted to the Rams' changes and acquired Higbee, who earned 51 of those 69 receptions after Week 10. Higbee out-pointed George Kittle, Mark Andrews, Zach Ertz, and Darren Waller down the stretch as the third-most productive tight end in fantasy football.

We may lament that teams don't change fast enough but when there's compelling information that change is happening, don't be guilty of intractability.

11. Understand the Talent-Scheme Fit And Potential for rEdundancy

Example: A.J. Brown

My favorite projection of the year was A.J. Brown. Since May of 2019, I was touting Brown as the safest rookie pick in dynasty and re-draft formats. I stuck to that view all year, ranking Brown well above my peers, projecting fantasy starter production:

Brown's yards-per-catch efficiency was something I would never project for any receiver, so it was a pleasant surprise. I had a good bead on Brown's receptions, yardage, and touchdowns.

The reason: I understood that Brown was capable of playing the slot and outside, and he'd be used as a hybrid option in those two roles much like former Titan Rishad Matthews, who performed as a fantasy starter in this capacity in Tennessee with Marcus Mariota and also in Miami with Ryan Tannehill.

I used Matthews' production in this capacity as the baseline for my projections with Brown, who I determined from film study was a better talent who could start and thrive immediately. Despite an injury that limited him during training camp and the preseason, I maintained these totals because the news about Browns' availability for the season remained positive overall.

Moreover, I knew that if Mariota and Tannehill both favored Matthews in the past that Brown's physical makeup and skill made him a similar option that both quarterbacks would gravitate towards as a primary target. This is what happened when Tannehill took over midseason and Brown built on strong games with Mariota into a fantasy starter with multiple elite games down the stretch.

Brown matched the scheme based on his skill set, style of play, and role. He had two starter-caliber talents at quarterback on the depth chart, and both passers favored big slot receivers who could also work the perimeter. It was enough to consider Brown's projected totals as the median value for his potential as a rookie.

12. Study Rookie production

There's no doubt that a well-researched rookie pick can make a difference in the outcome of re-draft leagues (see A.J. Brown), but it's more common for fantasy players to contract a case of "rookie fever," and reach too high for these talents. Of course, I'll tout the Rookie Scouting Portfolio publication as an excellent source for showing its readers the top production for rookies at each skill position as a way of framing realistic expectations for a player's first-year production.

I also recommend examining Footballguys.com's Season Stats and Game Logs for each team. Look at the rookies, second-year, and third-year starters and note the role of the player (flanker, split end, scatback, power back, move tight end, in-line tight end, etc.) and his efficiencies. To clarify, efficiencies include the following:

• Quarterback: Completion percentage and yards per attempt.
• Running Back: Yards per carry and yards per catch.
• Wide Receivers and Tight Ends: Catch percentage (Receptions / Targets) and yards per catch.

Gaining this perspective will help you set realistic expectations for rookies.

If you follow these lessons or at least use them to identify projections that stand out in a positive or negative way, you'll be well on your way to unearthing values that can make a difference for your team.

Did you enjoy this article? Find more of Matt Waldman's work here.