Why you shouldn't pick your fantasy QB too early -- or too late

Knowing exactly when to draft your quarterback -- even a top star like Aaron Rodgers -- is a key to fantasy success. EPA/LARRY W. SMITH

I know what you're thinking: That 2017 season seemed like a relative disappointment for the quarterback position.

It was, as for the second straight year, the position's fantasy point total declined -- by five percent after having dropped by 3.5 percent from 2015 to 2016. Matt Ryan's precipitous drop in production contributed, as did a disappointing year by Drew Brees.

Nevertheless, the collective group of NFL quarterbacks scored the sixth-most fantasy points in any single season in history, continuing to establish this as the "Decade of the Quarterback." To give you an idea of how high the bar has been raised at the position -- and this has been a development that largely began in 2012 -- quarterbacks as a whole scored 18 percent more fantasy points last season than they did in an average season during the century's first decade. There were 22 quarterbacks who reached the 200-point threshold in 2017, third-most in any season.

Last season's No. 11 scorer, Dak Prescott, scored 260.66 fantasy points, which would've been good for fourth-most at the position 10 years earlier.

Still, it's not the top-shelf scorers who are the story at the quarterback position, but rather the position's progression as a whole. Fantasy production hasn't only increased at the upper tiers, but also at its replacement level, with the forgettable folks at the position improving just as the stars have.

That brings us right back to the conclusion drawn in this very space in each of the past two seasons: There is absolutely no reason to ever pay sticker price for a quarterback.

To be clear, that line does not read, "Always wait for your quarterback." In an age of record-setting fantasy point totals at the position, such an approach is both lazy and dangerous, for two reasons: One, it can result in handing your competition a sizable advantage in the event he/she is quicker to draft an upper-tier quarterback who inexplicably lingers multiple rounds beyond sticker price, and two, in leagues that either warrant higher prices on quarterbacks (such as two-quarterback leagues), have historically drafted the position more aggressively or have done their homework and know the precise, proper values for said quarterbacks, you could wind up forced to pick from replacement-level choices.

It remains imperative that you draft yourself one of the top-scoring quarterbacks each and every week, but you need to do it for the cheapest price possible. In a way, it's fantasy football's version of extreme couponing.

What, then, is "sticker price"?

That varies by league, but we need to use value-based drafting (VBD) in order to determine quarterback's value relative to replacement level and compare it to the difference for players at other positions. Quarterbacks' raw fantasy point totals aren't relevant, because the position's scoring is so inherently stacked toward it occupying the top spots on the season-ending leaderboard. It's all about what a player can provide relative to what's on your league's free-agent list.

To save you the trouble, I pulled five years' worth of VBD data and mapped quarterbacks' expected draft-day prices. The chart below, under "FPTS Required," shows the number of fantasy points a quarterback needs to score in a given year to warrant his selection in that round (averages of all players in that round), and under "quarterbacks by projected FPTS," shows all quarterbacks who should be priced in that round using ESPN's projected fantasy points for 2018.

Strategically speaking, the idea at quarterback is to wait for as long as possible, even beyond where they're listed in this chart, to select yours. In my experience, at least one round -- 10 picks -- is recommended, and anything beyond 15-20 picks is excellent value. As you might expect, players in the later rounds tend to be the ones available at deeper discounts, so keep in mind in the event that if Aaron Rodgers remains out there until the 50th overall pick, that'd be the wiser deal than, say, Marcus Mariota 109th overall.

These player valuations, remember, are based upon ESPN's projections, so feel free to shuffle names around if you formulate your own projections. I've done that with my own rankings and projections -- remember that I'm the guy who would never draft Tom Brady in the first 50 overall picks (in a one-quarterback league).

That in 2017, Alex Smith (No. 4 scorer among quarterbacks; drafted 203rd overall and in only 8.2 percent of ESPN leagues), Carson Wentz (No. 5; drafted 135th, 14th round, 42.1 percent), Matthew Stafford (No. 7; drafted 107th, 11th round, 100.0 percent), Philip Rivers (No. 8; drafted 140th, 15th round, 86.0 percent), Ben Roethlisberger (No. 10; drafted 106th, 11th round, 99.5 percent) and Jared Goff (No. 12; drafted 220th and in only 3.9 percent) were all plenty productive players who cost practically nothing on draft day only illustrates the benefit of patience. This came on the heels of 2015 and 2016 seasons in which an even greater number of out-of-nowhere quarterbacks were available either in the late rounds or as in-season free agents.

What about non-PPR leagues?

Though quarterback fantasy scoring scarcely changes between PPR (point per reception) and non-PPR leagues -- quarterbacks as a whole caught just two passes in 2017 -- the position's value over replacement player does change a decent amount.

In PPR scoring, the upper-tier running backs, wide receivers and, to a lesser degree, tight ends gain more of an advantage relative to replacement. That makes all three positions more important to address with early picks than quarterback, thus pushing the quarterback top tier down slightly. In a non-PPR league, that's less the case, and there's an effective "accordion" effect at the position, with the higher-tier quarterbacks gaining a slight amount of value, but the lower-tier quarterbacks losing a similar amount of value.

One issue with this: With so few quarterbacks projected to score an elite number of fantasy points -- not one is expected to reach 320 -- none of the quarterbacks in the top tiers really gains on the chart, though their odds of returning third- or fourth-round value is greater than in PPR.

Again using VBD, here's a look at proper quarterback pricing in a non-PPR league:

Since it's more important to pad your weekly point total with a high-scoring quarterback, it's imperative that you avoid dipping into the replacement pool in this format. The sweet spot for picking a quarterback here might well be the Rounds 9-11 range, should any of those players slide.

Two-quarterback leagues radically shift the prices

The wrinkle in all this is the two-quarterback league, a format that makes a heck of a lot more sense in this era of record passing numbers. In two-quarterback leagues -- or in "superflex" leagues that include quarterback as an eligible flex player -- draft strategy at the position is even more critical.

Using VBD numbers, a top-tier quarterback is generally worth between 65 and 70 additional fantasy points over a replacement-level performer compared to a one-quarterback league -- that's using five-year (2013-17) data -- which vaults the position several rounds up in the pricing system. In fact, last season's top-scoring quarterback, Russell Wilson, was by VBD the fourth most-valuable player in a two-quarterback league, behind only Todd Gurley II, Le'Veon Bell and Alvin Kamara.

Using the same VBD method as above, here's how ESPN's projections would price quarterbacks in a two-quarterback, PPR league:

Now, a two-quarterback league with non-PPR scoring:

The key difference in a two-quarterback or superflex league, however, is that quarterbacks rarely come at as extreme discounts as in one-quarterback formats, mainly due to there being a finite number to fill a league's weekly lineup spots. For example, in weeks with six byes, there are only 26 starting quarterbacks available to play, leaving precious few extras in a 10-team, two-quarterback league (20 starters at the position) and even fewer in a league with more teams.

It's for that reason that any player who lingers even one round longer than his listed price is an outstanding value, especially in Rounds 1-5, and the case can be made that no player should last even five picks beyond his projected value.

Some final strategic notes

In-season quarterback management requires a vastly different -- and arguably entirely opposite -- approach, as impatience with players is often wiser.

Consider that in a 10-team, ESPN standard non-PPR league last season, the average quarterback (using our new fractional scoring) across all leagues would've scored 17.2 fantasy points, and a "replacement level" player in those formats would've averaged 15.5 fantasy points. Those are extremely high bars, amounting to 275 and 248 fantasy points respectively over a 16-game seasons, and they illustrate why mixing and matching your quarterbacks weekly, especially when dealing with a struggling draft pick, is a wiser strategy than locking one player in all season.

Backups matter. A quick look at the charts above for one-quarterback leagues show that 26-28 quarterbacks are priced as "draft-worthy" in that format. With as many as 20-plus quarterbacks annually proving themselves productive weekly scorers these days, it's ever wiser to give yourself two options from which to play matchups weekly, even if that second quarterback is one you rotate throughout the season. Never get complacent, starting only the one quarterback you drafted.