You know what they say: variety is the spice of life.

When it comes to fantasy sports, change doesn't always happen quickly. (How many of your season-long teams still do snake drafts instead of auctions? How many of your leagues are just head-to-head formats with the same old scoring rules you had 10 years ago?)

FanDuel is changing that for us by introducing the superflex to DFS.

What exactly does that mean? I'm glad you asked.

What's a Superflex?

Superflex, for FanDuel's purposes, means that you can roster a second quarterback at one of your flex spots.

What's that? "One" of your flex spots? Yup. The superflex setup eradicates one wide receiver slot, as well as kickers and defenses, from the player pool and gives you two different flex slots. In one of them, you have the ability to roster a quarterback, running back, wide receiver, or tight end. In the other, you can go with a running back, wide receiver, or tight end.

This gives us the chance to roster up to two quarterbacks, three tight ends, or four running backs or wide receivers.

So, what should you do with that superflex?

Filling the Superflex

The most obvious way to attack the superflex is to roster a second quarterback, for reasons we'll detail in a second. But just know that pricing is a bit different when joining superflex contests -- but only for quarterbacks.

In the first week of the new format, quarterbacks are priced anywhere between $100 to $1,300 more in the superflex contests than in regular contests. In the grand scheme of pricing, that's a small change.

The biggest boost is a 20% increase in pricing (for Nick Foles, who is sliding into a starting job). Bryce Petty, who is also taking over as a starter, is up 15% in pricing, and so is Tom Brady, who is just priced up as a result of the new format. We can probably use that as the baseline: a max of 15% in salary bump.

So, let's experiment. We'll dig into past results, over three seasons from 2014 through 2016, and give all quarterbacks a 20% boost in salary from what their FanDuel price actually was in a given week. That's probably aggressive, but it still doesn't even matter. Check out these results over these three seasons in terms of fantasy points per $1,000 when looking at roughly 1,500 total games for each position (sorted by projected FanDuel points to help eradicate outliers and injuries).

Even with that likely over-aggressive multiplier of salary for quarterbacks, they've reached each value threshold better than any other position. The gap gets narrower when we look at four-times value (4.0 FanDuel points per $1,000 in salary) because an $8,000 passer would need 32 FanDuel points to hit that mark. A $5,000 running back needs only 20, by comparison.

But in cash games, you should be opting to roll out a second quarterback in basically every circumstance because of the likelihood that they produce at least 3.0 FanDuel points per $1,000. Quarterbacks are the most consistent and easiest-to-predict position in fantasy football.

And, of course, keep in mind that we've added 20% to the weekly salary for all quarterbacks for this chart. Salaries could be boosted by only 10% or less in most instances, based on the first week of data.

Filling the Other Flex

Well, we can look back to the chart in the prior section for guidance here.

Running backs and receivers offered similar odds of returning 1.5 FanDuel points per $1,000, but on the higher end, running backs had produced a higher rate of games with 3.0 and 4.0 FanDuel points per $1,000. Essentially, among players with reasonably high projections who were expected to post a solid fantasy outing, the position outproduced wide receivers in terms of providing a ceiling.

That may seem counterintuitive, as running backs don't provide big play ability quite like wide receivers do, but on FanDuel, touchdowns generate more fantasy value than on full PPR sites.

Perhaps the most important takeaway is that, while rostering two cheap tight ends might let you spend up on stars elsewhere, you're certainly lowering your lineup's safety while not necessarily increasing the ceiling.

If you look solely at the stud tight ends in ostensibly good matchups (215 in this split with a projection of at least 10.0 FanDuel points), they hit 2.0 FanDuel point per $1,000 mark 34.4% of the time and 4.0 points per $1,000 3.7% of the time. There's a case to be made for rostering two stud tight ends, but a double punt is not ideal.

Stacking

Here's where things get extra fun. According to research by 4for4.com's Chris Raybon, a quarterback's performance correlates pretty strongly (relative to other in-game correlations) with the performance of the opposing quarterback. It's not quite as strong as, say, a quarterback and his top wide receiver, but you can now run out two QB-WR1 stacks in the same lineup.

That means giving a long look at a full game stack, including both quarterbacks from a single game.

Across our sample -- from 2014 through 2016 with quarterbacks projected for at least 10.0 FanDuel points -- we have more than 700 games with both quarterbacks meeting that minimum projection. In that sample, quarterback fantasy point totals from passers in the same game have a correlation of 0.526. Based on the value thresholds (rather than raw fantasy point totals), same-game quarterbacks are linked even stronger at 0.587. There's a lot of merit to pairing two starting quarterbacks from the same game.

Should You Ever Not Flex a Quarterback?

Based on what we've discussed so far, your answer is probably, "No, I should always flex a quarterback." I sure don't think you're wrong, but there are some instances where you may want to avoid that.

The first is just general game theory. If everyone else is building lineups the same way -- starting with two quarterbacks -- you can differentiate by going against the grain. After all, with quarterbacks priced up, you can likely find some leverage in opting for a stud running back over a mid-tier quarterback. Peep these results if we stick with the 20% boost to quarterback pricing.

Quarterbacks -- even with their prices boosted -- are safest for giving you a bare minimum return on investment, but stud running backs and receivers -- your Le'Veon Bell and Antonio Brown types or your Leonard Fournette and A.J. Greens to a lesser extent -- might now be the same price as a mid-level quarterback.

You don't always have to roster two passers, but finding two quarterbacks with cake matchups should be your first line of thinking when approaching the superflex.

Takeaways

Quarterbacks are strong fantasy football commodities, and rostering two puts a nice twist on the usual format of DFS. Pairing opposing passers together can unlock some nice upside for your squads, especially if you stack them with the right pass-catchers.

You don't always have to plug in two quarterbacks no matter what. After all, we might find a week where two minimum-salary running backs are slated to be bellcows, which can change everything. DFS is always contingent on the slates, matchups, and pricing. That's what makes it so great.

But these general takeaways should put you on the right path as you head down the super exciting superflex format.