Over the last few seasons, the NFL has transitioned into a pass-first league in part due to rule changes allowing less contact between defensive backs and wide receivers. The change in offensive philosophy has resulted in some believing that the fantasy value of quarterbacks, wide receivers and tight ends are higher than ever. This notion couldn't be further from the truth. In fact, the shift in play-calling is actually increasing the fantasy value of running backs.

As teams pass the ball more, there are more serviceable WR/TE to be found on waivers and more depth than ever before. As teams run the ball less, there are collectively less RB touches to go around and less quality fantasy players at the position. For this reason, it is rare for a second-string RBs to have legitimate fantasy value, while second-string WRs are consistently productive fantasy players (i.e. Randall Cobb).

Because there are fewer desirable options and less depth at RB than at other positions, elite RBs are the most valuable assets in standard fantasy football. There is a larger fantasy point differential between an elite RB and a replacement level running back than there is between an elite QB, elite WR, elite TE and a replacement level player (RLP) at that position. 

This point differential can be referred to as PAR (points above replacement). The more PAR a player provides, the more valuable that player is to your team. A replacement level player (RLP) can be defined a number of different ways. The best way to describe a RLP is a player that you already have on your bench or could acquire with little difficulty through waivers or trade. 

For this example, I'll define an RLP for a ten team, standard-roster league (1 QB, 2 RB, 2 WR, 1 TE, 1 FLEX, 1 K, 1 D/ST). Assuming that four teams start a RB at Flex, four start a WR at FLEX and two start TE (the number of that position in the starting lineup multiplied by two) and account for some teams being stacked at certain positions, replacement level players can be defined as the following:

• QB - QB13 (Ten in starting lineups, two quality backups)

• RB - RB28 (20 in starting lineups, four at FLEX, three quality backups)

• WR - WR28 (20 in starting lineups, four at FLEX, three quality backups)

• TE - TE15 (Ten in starting lineups, two at FLEX, two quality backups)

To quantify how many PAR an elite player at each position provided last season, we recorded the point differential between the third overall scorer and a RLP at QB, RB, WR and TE for all fantasy relevant weeks (1-16) of the 2014 season. 

Due to injury, matchup or other reasons, normally solid players are not startable some weeks whereas normally mediocre players are occasionally great options (i.e. Knile Davis when Jamaal Charles went down last year.) By analyzing replacement level production on a week to week basis we can get a better idea of what the depth of the player pool was like at that moment in time. This is why its important to look at replacement level production from each week of the season rather than the player who finished as the RB28 for the season.

Charles LeClaire-USA TODAY Sports

For purposes of comparing what is considered elite to a RLP, the third highest scoring player at each position is a better indicator than the first overall scorer. Looking at the top scoring player in a given week at each position can be misleading because there will always be outliers at the top of any numerical distribution (i.e. Jonas Gray scoring 43.9 points in Week 11). Analyzing the third overall player gives a more accurate depiction of reasonable expectations for an elite player (i.e. Jamaal Charles scoring 27.8 points in Week 11 as the RB3). 

The following data pertains strictly to leagues with a standard lineup (1 QB/2 RB/2 WR/1 TE/1 FLEX) and a standard scoring system. If your league starts two QB then elite QB will provide more PAR than the numbers below. Same goes for WR if your league starts three WR.

All of the following numbers are the average of the weekly point totals produced by the player who happened to be the QB3, RB28, etc. that week. All stats are courtesy of fantasypros.com. 

2014 Weekly PAR:

QB3 Avg (per week): 26.39 pts // QB13 Avg (per week): 17.03 pts // Avg PAR (per week): 9.36 pts

RB3 Avg: 23.46 pts // RB28 Avg: 7.36 pts // Avg PAR: 16.1 pts

WR3 Avg: 21.96 pts // WR28 Avg: 8.91 pts// Avg PAR: 13.05 pts

TE3 Avg: 14.53 pts // TE15 Avg: 5.83 pts // Avg PAR: 8.69 pts

These numbers show that high-level production from RB is going to give you more of an advantage over your opponent on average than similar production from every other position. Elite WR production is also important and WR should be prioritized over QB and TE.

While it is important to analyze replacement level production on a weekly basis, there is far less turnover at the top of each position group. As a result, comparing the average weekly production of the player who finished last season third overall at each position to a RLP at the same position will give us further insight as to which position provides the most value. 

Below is a compilation of the PAR provided by the third overall player for the 2014 season. The positional value trend remains the same.

2014 Season-Long PAR:

QB3: Russell Wilson - Weekly Avg: 20.5 Pts // QB13 Avg: 17.03 Pts // Wilson's Avg PAR per Week: 3.48 Pts

RB3: Marshawn Lynch - Weekly Avg: 16.6 Pts // RB28 Avg: 7.36 Pts // Lynch's Avg PAR per Week: 9.24 Pts  

WR3: Demaryius Thomas - Weekly Avg: 14.4 Pts // WR28 Avg: 8.91 Pts // Thomas' Avg PAR per Week: 5.49 Pts

TE3: Jimmy Graham< /strong> - Weekly Avg: 9.1 Pts // TE15 Avg: 5.83 Pts // Graham's Avg PAR per Week: 3.27 Pts

No matter how you slice it, elite RB are the most valuable fantasy commodity. WR provide more value than TE and QB, but Marshawn Lynch was still markedly more valuable than Demaryius Thomas last year.

The main takeaway here is that when you only have to start one player at a position each week (QB/TE), there are plenty of serviceable options available on waivers or through trades. When you have to start two players at a position each week (RB/WR) and also count on that position to fill your flex spot, quality fill-ins are much harder to come by. 

How to Incorporate PAR and RLP Into Your Fantasy Strategy

One interesting implication of analyzing players through the scope of PAR is that it ends with the conclusion that players who drop in the draft due to being perceived as injury-prone are often underrated. Arian Foster and Rob Gronkowski are two examples. Both have had their fair share of injuries over the years but have produced at an elite level when healthy.

Richard Mackson-USA TODAY Sports

When analyzing an injury prone player, it is crucial to remember that if Foster or Gronk get hurt, you’re not going to take a zero at that spot for the week. You'll plug in a RLP or a quality backup and hope for the best.

When Gronk/Foster are in the lineup, they have the potential to explode any given week and provide the almighty PAR that will lead you to victory. In 2014, Arian Foster scored the fifth most fantasy points (235.5) among RB in standard leagues. However, he was second in fantasy points per game (18.1) in the 13 games in which he did play. 

After adding replacement level production (7.36 points) for the three games in which Foster missed last season to his 2014 fantasy point total (3 games x 7.36 points = 22.08 pts + 235.5 pts = 257.08 pts), he was actually the fourth most valuable RB behind only Marshawn Lynch, DeMarco Murray and Le'Veon Bell. Factoring in replacement level production shows that Foster was actually more valuable than Matt Forte (244.6 pts in 16 games) in 2014 despite accumulating less total points. 

Foster's groin injury is not a minor one and will rightfully drop him down draft boards. However, I'm absolutely open to taking him in the middle rounds and I currently have him at 52 on my big board. Foster may not play until the middle of the season, but I'll gladly store him on IR and hope he regains enough health to get on the field and run behind the Texans' solid offensive line. 

To roughly estimate where you would take a player based on where he is ranked on your board, my rule is to add 20%. For Foster, this means look at taking him in the early 60s (52 x 0.2 = 10.4 + 52 = 62.4). This rule accounts for the fact that you're almost never going to take your 49th ranked player with your 49th pick because other owners will always take players earlier than you feel they should.

I'm more than willing to risk a pick in the 60s on a running back who could be a crucial piece for a playoff run. It could come back to bite me, but that bite won't hurt nearly as bad as watching a healthy Arian Foster crush me in the fantasy playoffs.

Now, back to Gronk. TE in general don't score at an overly high rate, but Gronk is the exception to that rule. Despite only playing one TE in most leagues, he can still provide significant PAR when he is in the lineup.

He averaged 12.3 fantasy points per game (fppg) in 2014, a whopping 2.7 fppg more than the TE2 last year (Antonio Gates - 9.6 fppg). The difference between the #1 and #2 QB, RB, and WR was 0.2, 0.4, and 1.2 fppg respectively. For that reason, Gronk is still an extremely valuable fantasy commodity this year and I currently have him 7th overall on my board.

Drafting RB and Maintaining RB Depth

Elite running backs and running back depth in general are still the key to success in fantasy football. In standard leagues with a 16 player roster, you should have a minimum of five RB on your team at all times and preferably 6-7. 

There is no reason to have more than one K or D/ST, so that's two spots. You really only need one QB but sometimes you might need a backup for a bye week, so QB can count for two. The same rule for QB applies to TE, so we can also allot TE two spots. 

That's six spots that you should allot to non-RB/WR. That leaves ten spots for WR and RB. If you have enough RB depth to start one at FLEX, there is no reason to carry more than four WR on your team. There are replacement level WR on waivers deep into the season. The RB waiver crop dries up much more quickly. 

As long as you draft multiple RB and pay attention to the waiver wire, you should always have three RB to fill out your RB and FLEX spots. If you have any RB left over, you can always trade with an RB-needy owner.

Don't necessarily pass on a stud WR or even stud QB or TE in favor of a middling RB. Owning an elite RB is the best way to win a standard fantasy football league and that you should organize your big board accordingly. 

If you are in a league where the objective is not to get last place, then the strategy changes a bit. Going WR at the top may be a more effective strategy for playing risk-averse due to injury concerns with RB. If your goal is to win the league and there is no difference between finishing second or last, drafting RB early and often will give you the best chance to come home with the trophy. 

I'll release my full fantasy big board next weekend and analyze players I am particularly high or low on relative to general public opinion. If you have questions about where I have certain players ranked or about anything football related, feel to reach out on twitter @mrosekNFL.

Good luck next season, and make sure you're not the guy desperate for RB come Week 4.