[This is a continuation of my exploration of FireFactor, which is our way of ranking players compared to replacement players at the same position. I already discussed it in detail here while applying the ranking to the quarterback position.]

Over the past few seasons, quarterbacks, receivers, and even tight ends have been creeping up fantasy draft boards, closer and closer to the first position and even overtaking the latter half of the first round.

The league is morphing into a passing one, so it makes sense to a degree. But these average draft position (ADP) jumps are mostly just a culmination of new (and often misleading) ideas, subversive ideas that are overtaking the traditional drafting approach of selecting running backs early and often. These alternative strategies to early running backs fail to account for just how to fill the necessary running back slots every Sunday.

Last year, early-round running backs flamed out hard, but late-round running backs almost never pan out. I wrote a lot about this prior to the season last year on a website that was so popular it failed to renew its domain name, so I'm sure you all had the chance to read it. But in seriousness, JJ Zachariason wrote about running back bust rates, and I couldn't recommend it enough. I also reference it later.

While I have you, JJ wrote a five-part series on running backs: The Current Misconception, Supply and Demand, Replaceability, Predictability, and Market Value.

So, with all of the dissenting opinion, what are you to do this year in the first few rounds of your fantasy draft?

Let's break it down numberFire style.

FireFactor and Running Backs

Running backs are becoming increasingly platooned in the NFL, and this is part of the reason people feel less and less confident investing early-round picks in them. This plays into the decision for some people to recommend playing it safe and locking up the less volatile non-running back slots, getting Graham and elite quarterbacks and receivers early in the draft.

This is misleading, and it's the opposite of safe. There really is no way to play it safe in a draft - even if you draft running backs. Stick with me. I promise.

Now, it's okay if you disagree with me, but I'll make my case using projections from our fantasy football cheat sheet, which includes the FireFactor rankings and the ability to adjust them according to your personal league settings. It's largely consistent with my independent research done over the past few years of actual statistical results, as well.

This chart indicates the FireFactor rankings for running backs in a 10-team league starting 1 quarterback, 2 running backs, 3 receivers, and 1 tight end, which means that a total of just 20 running backs could be started in a week. It's a bit of an antiquated roster construction, but it'll prove a future point.

It's truncated so that you don't gloss over 28 rows of numbers, but the message is still discernible. By the way, just like for the quarterback analysis, I'm omitting the names of running backs because they differ between PPR and non-PPR settings, and also because I'm interested in conveying strategy and not recommending certain players.

While there were almost no difference between QB4 and QB12 in 10-team leagues, the difference between RB1 and RB22 is immensely significant. Even the difference between RB7 and RB22, in non-PPR leagues, is a wider disparagement in terms of FireFactor than the discrepancy between QB1 and QB12 regardless of passing touchdown point value.

To put it simply, the margin between elite running backs and fringe fantasy starters is basically larger than the gap between the best and worst starting quarterback in a 10-team league. And very few leagues have a 20-running back cut off.

Let that sink in.

The Weekly Game

The discrepancy between quarterbacks and running backs becomes more obvious when splitting total points into weekly averages. I know tallies aren't compiled evenly game-to-game, but visualizing the weekly expectations should make you rethink relying on weak running backs in order to secure strong players at other positions.

I'll switch things up and examine 12-team PPR leagues that include a running back and receiver flex, meaning that 36 running backs could be started in a given week. These league settings will be important later on, as well.

There are some pretty big drop-offs in this chart. Having a running back outside the top 24 provides very little help compared to truly elite backs, something that's not really the case between the afterthought quarterbacks and the studs.

Those elite backs, the ones who can post the absurd levels of points, will generally provide nearly a 9.5-point advantage over 12th-ranked backs. Or, to put it another way, the gap we project between RB1 and RB12 is larger than the gap between the one we project between QB1 and QB16.

Let that sink in again.

And whether you agree with our player rankings or not, you should find merit in the point distribution among positions. That's what this is all about anyway: breaking everything down and visualizing how hard it is to replace running back productivity with weaker fantasy backs.

Applying FireFactor to Your Fantasy Draft

Unless you have some really offbeat drafters in your league or play in a non-traditional league, Jamaal Charles, LeSean McCoy, and Adrian Peterson will probably be off the board by the fourth pick. Most people slotted fourth will be confident with Matt Forte in that Marc Trestman offense.

Then what?

Pick five could really swing the momentum of a league this year in most drafts, as the surest bets to be worthwhile at running back will be gone. Would you chance it on Eddie Lacy? Marshawn Lynch? Maybe Arian Foster in a comeback year?

The point is that there are crippling question marks surrounding nearly ever other running back at this point. So why not just draft Calvin Johnson or Jimmy Graham? Hopefully, if you've read my quarterback analysis, you're ruling out quarterbacks.

Well, this is where you need to take a risk, according to the numbers. It's a risk in the sense that you're willing to spend the fifth overall pick on a running back who might finish as the 12th-best running back instead of Johnson or Graham, likely locks to be a top-3 receiver and the best tight end.

I'll cite JJ Zachariason for some numbers to help prove my point. (Here's the link again.) Over the past five seasons in 12-team PPR leagues, running backs who were top-6 in average draft position finished as top-12 backs 50.33% of the time, and backs who were 7th through 12th in ADP finished as top-12 backs 46.67% of the time. That's not great, but it is compared to 30.00% for backs with an ADP of 13-18 and 20.00% of backs 18-24.

As Zachariason's research shows, top-12 and top-24 backs tend to come from the consensus top-6. As my chart above indicates, you don't want to be messing around with backs outside the top 24 (or, colloquially, RB3s and Flex players) for weekly production.

To summate, top-12 backs are typically drafted early, and it's hard to find good running back production if you wait. You can find it all over the place anecdotally from the guy who drafted Knowshon Moreno or Zac Stacy, but statistically, you need to invest early in the position to get a good lead back for your fantasy squad.

Implementing FireFactor Into Your Fantasy Season

I'm going to do what I did for the quarterback piece and provide a bit of a value comparison chart between positions, based on FireFactor. If you like tight ends, you may want to stop reading here. There's not much left because the chart speaks for itself. This is for a non-PPR league with four-point passing touchdowns and a receiver and running back flex position.

Just to explain, you'd be better off having a low-level RB1 (meaning roughly the 8th-best to 12th-best back) than the best quarterback or receiver. And basically a low-level RB2 is as valuable as the best tight end (meaning Graham).

So, what did we learn? Running backs are so valuable that even low level RB2s can be about as valuable over replacement running backs as Jimmy Graham is over replacement tight ends, depending on the projection for Graham.

Just remember that when you think about taking Graham over a number of question-mark backs in rounds one or two.