With the draft season upon us, it’s time to dial in our process and create a winning draft strategy. Part of that winning strategy involves accurate player projections, whether creating them yourself or using a trusted source. Then you need to implement these projections into your decision-making process in a strategic and informed way.
The problem with raw projections is that they don’t tell the whole story. No matter how accurate they may be, they don’t do a great job of evaluating true player value across multiple positions. In most leagues and scoring formats, quarterbacks score more points than any other position. The 12th-best quarterback in any given season may score more than the best wide receiver or running back.
Of course, this doesn’t mean you should only draft quarterbacks. Quite the opposite. Positional scarcity, the number of starters per position required in your lineup, how players’ projections compare against others at their position, and how they compare to other positions drafted in similar spots in the draft, all should influence your drafting decisions.
When searching for an edge in making these evaluations, we can look at common metrics used in fantasy analysis, like value over replacement (VOR) and VOR scoring drop-off, to help create an advantage while building your roster. While these concepts aren’t new or groundbreaking, understanding what they are and how they work can make you a better, more informed drafter.
Below, we will examine how these concepts are derived, why they are valuable, and how you can incorporate them into a successful draft strategy.
Defining Value Over Replacement
Value over replacement (VOR) is a commonly used metric in fantasy football that measures a player’s projection relative to the baseline-level projection of a player at their position. This comparison gives you a better idea of how much more valuable a player is relative to the baseline player at their position.
Calculating The Baseline Player
The trick with calculating VOR is determining who the baseline player is at each position. This baseline player becomes the projection against which all other players at the position are measured. There are numerous ways to do this, and the chosen method can greatly impact the results of your VOR calculation.
The most common approach is to use the “worst starter” as the baseline, using the projected points of the last starter at a position. So, in a 12-team league that starts one quarterback, the worst starter would be the number 12 quarterback based on projections. If your league starts two running backs, the baseline starter would be RB24, and so on.
Other methods include using average starter production, a more complicated “man-games” approach, or using a particular spot in the draft and using that as your positional baseline. For example, using the first 10 rounds as the target, the total number of players taken at each position within the top 120 picks becomes your baseline player.
While all of these have merit for different reasons, I wanted something driven by the data I had compiled over the years. With most of the approaches above, I didn’t feel they adequately addressed the nuance of how flex spots were filled. In leagues with two flex spots, how do you know how many are allocated to each skill position? The worst starter method doesn’t fully address this, and the specific draft spot cutoff method assumes rational drafting with fantasy managers correctly allocating their roster spots.
Given this, I set out to create a baseline player selection method that quantified historical flex-worthy appearances by position using weekly scoring data. I wanted the method to be flexible, drawing from this historical dataset while incorporating a weighting component to give more relevance to recent seasons and better capture scoring trends.
To reflect these trends, I applied a weighted approach to the past six seasons of flex-worthy data, assigning greater weight to more recent years. For example, if running backs qualified for flex consideration 100 times last year and that season carries a weight of 5, its contribution to the total is 500. This method helps capture evolving scoring patterns while still grounding the model in a meaningful historical sample.
The easy part is figuring out how many players for each position are required every week. For this discussion, we’ll use a 12-team league starting one quarterback, two running backs, two wide receivers, and one tight end. There are also two flex spots that can be filled by any of the skill positions. It’s pretty easy math, but just to lay everything out, the REQUIRED weekly starters are:
12 Quarterbacks
24 Running Backs
24 Wide Receivers
12 Tight Ends
72 total required weekly positional starters, 60 for the skill positions
In my data model, which includes weekly statistics dating back to 2010, I identify the specific players who should have filled the 60 required skill position spots each week. These players are then excluded from flex consideration in that week’s analysis.
As we dig into the flex spots, each team has two spots to fill each week. That is 24 additional skill positions that need to be worked into lineups. After removing the required starters for each skill position (60 players in total), we must determine who the NEXT BEST 24 players are, regardless of skill position. These are the 24 players each week who are flex-worthy and should be in lineups.
Going back to 2010, using point-per-reception (PPR) scoring, wide receivers earned the flex-worthy designation 74.79% of the time, running backs 17.99% of the time, and tight ends the remaining 7.23% of the time. Breaking this down, out of the 24 flex spots that need to be filled weekly, the allocation would look like this (standard mathematical rounding used):
Wide Receiver = 18 spots (24 * 74.79% = 17.95)
Running Back = 4 spots (24 * 17.99% = 4.32)
Tight End = 2 spots (24 * 7.23% = 1.73)
So in any given week, three out of four flex spots should be filled by wide receivers, with the remaining 25% divvied up amongst running backs and tight ends. In reality, we know this isn’t how it always works. The diversity of your bench depth, injuries, and decisions of floor versus upside projections play into your weekly considerations. Regardless, this shows where the next best 24 weekly scorers came from over the past 15 seasons.
As I mentioned earlier, my desired baseline calculation is weighted to favor the most recent seasons’ flex allocation to capture trends – for example, if running backs have factored in more over the past couple of years, they would receive slightly more preference in the baseline calculation.
After adding the flex allocations to the number of required starters, my baseline calculation for the 2025 season is as follows:
Running Back = RB28 (raw baseline calculation 28.3)
Wide Receiver = WR42 (raw baseline calculation 41.7)
Tight End = TE14 (raw baseline calculation 14)
As you can see, the weighting did not impact the final positional allocations. Nor should it, unless a significant trend has truly emerged and changed scoring patterns.
It is important to consider how your league’s scoring settings can impact the baseline calculation. For tight end premium (TEP) leagues, tight end flex-worthy starts jump up to 16.71%, bringing their baseline allocation to 17, robbing one spot from running backs and two from wide receivers.
Calculating VOR
Having established our baseline players, we can now calculate VOR. A simple example using quarterbacks: assume the QB1 is projected to score 418 points and the QB12 is projected to score 337 points.
The difference between these two is 81 points. This is your VOR score for QB1. As you go down the list of quarterbacks, the QB12’s projected points get subtracted from each of the previous quarterbacks (QB1-QB11) to create their VOR score. The QB12 would therefore have a VOR score of 0.
Now that we have established our baseline player rankings for each position and have determined how to calculate our VOR, let’s explore how to apply this during our drafts.
Value-Based Drafting
Value-based drafting (VBD), like VOR, isn't a new concept. The idea of VBD goes back to the mid-1990s when co-owner of FootballGuys Joe Bryant introduced it to the fantasy football world. While it has evolved over time and many have given their own spin to the concept, the core principles and objectives are the same.
VBD applies the VOR scores across all players in your draft and assesses how those values compare to other players drafted in a similar range. Once you have your VOR score calculated for all players, you can incorporate them into the list of players, which can include total projected points, average draft position (ADP), and any number of other data points you wish to use in your evaluation.
For our example, my list includes just a few items to keep it simple without adding additional data to possibly complicate things. You can sort it in multiple ways depending on how you want to evaluate the data.
You can sort by ADP to see which player in each round offers the highest VOR. You can sort by VOR to see which positions/players are near the top. You can sort by total projected points to better understand the relationship between total points scored by positions like quarterback, and the comparative value of other positions based on scarcity.
The table above is sorted by VOR. You will see the top four spots are occupied by wide receivers, with one more making it into the top 12. There are five running backs and two tight ends rounding out the top 12.
Looking at PPR ADP from FantasyData, nine of the top 12 VOR player scores are drafted inside the first round. Brock Bowers, Devon Achane, and Trey McBride are drafted in the second and third rounds.
About halfway down the list, you see Jayden Daniels, the top overall projected scorer this year based on Mike Clay’s PPR projections, yet his VOR score is 19th on the list. Daniels’s VOR score as the overall QB1 is only 77.05, less than half of the VOR score of the projected WR1 in Ja’Marr Chase.
Using VOR within rounds to compare the relative value of players and positions can give you insight into who might be the best pick. If going into a draft with a position-agnostic strategy and simply looking to draft the “best player available”, using VOR can guide you in the right direction.
But what if you are choosing between several players with similar projections, which one do you choose? Which position should you focus on if everyone looks the same? This is where we can incorporate VOR scoring drop-off by position to add another element to your decision.
Positional Scoring Drop-Off
When you get to a certain point in your draft, you may start to think that everyone in this area feels the same. Especially once you get past the elite producers at their position and are swimming in the murky WR3/RB3 player pools, it gets even more difficult to decide who to draft. Even with VOR, very close projections within a position will result in clusters of similar VOR scores.
We can look at how and where scoring drops off by position to further inform our decision. Sometimes, just looking at a list of numbers can make that analysis tricky, so creating a visual can help us spot these drop-offs more quickly.
The line chart above represents the VOR score by position for each positional ranking 1-42 (as 42 was our largest baseline player rank for wide receivers). This chart clearly shows where the dramatic positional scoring drop-offs occur based on this year’s projections.
WR1 to WR2 represents a substantial drop-off with a 34-point drop in VOR score between Chase and Justin Jefferson. The drop-off after WR2/3 is pretty stable and steadily declines throughout the rest of the positional rankings.
With running backs, you see the drop-off start to occur after RB4, with a 10-point VOR drop to RB5 and then another 10-point drop to RB6. After that, it’s a gradual slope that roughly follows the wide receiver slope until you hit RB18. RB18 to RB21 represents a significant 32.1-point VOR drop. ADP indicates these four running backs are being taken within seven spots of each other.
Tight ends experience their largest VOR drop after TE2. Once McBride is off the board, there is a 29-point drop to George Kittle at TE3, and by TE4 with Sam LaPorta, the value has dropped an additional 41 points. This drop from TE3 to TE4 represents the single largest VOR score drop-off.
Lastly, with quarterback, the largest drop-off occurs from QB4 Lamar Jackson to QB5 Joe Burrow with a 26-point VOR drop. Then QB8 Baker Mayfield to QB9 Kyler Murray is another 10-point drop. Otherwise, the quarterbacks drop at a pretty consistent rate and, as we touched on earlier, have the lowest starting VOR score for the top-ranked player at any position.
So, how do you use this information? Instinctually, you want to strike before the drop-offs occur. To do so, you need to have an idea where they might happen by incorporating ADP, as positional drop-offs are only relevant relative to where these players are being drafted in your leagues.
Utilizing two of the more dramatic VOR drop-offs in our above examples, let’s start with a tight end comparison and finish with a running back comparison. We know that the drop from McBride to Kittle is roughly 29 points. Using current PPR ADP data from FantasyData:
McBride has a third-round ADP of 27 and a VOR of 105.30
Kittle has a fourth-round ADP of 45 and a VOR of 76.20
They are coincidentally taken around the same area of the draft board, meaning this is a realistic decision that will have to be made by drafters.
Players within several picks of McBride and their VOR scores: Josh Allen (73.25), Kyren Williams (67.80), Tyreek Hill (64.40), Chase Brown (47)
Players within several picks of Kittle and their VOR scores: Jalen Hurts (71.15), Kenneth Walker (55), Joe Mixon (44.20), Chuba Hubbard (43.50), Omarion Hampton (31.70), D.K. Metcalf (44.50)
Given this, let's examine the opportunity cost using VOR scores to see whether drafting McBride or Kittle makes more sense:
Scenario 1: Draft McBride in Round Three If you draft McBride, you lose the opportunity to draft Allen, Williams, Hill, or Brown. However, in the next round, you could draft one of the players selected around Kittle’s ADP.
Scenario 2: Wait for Kittle in Round Four By passing on McBride, you can take any of Allen, Williams, Hill, or Brown who are all projected to score more total points. However, you miss out on McBride's superior VOR score and will have to settle for a much lower-value tight end option later.
Comparing the Trade-Offs The drop-off from Allen to Hurts is only ~2 points. The drop-off from Williams to Walker is ~ 13 points. Hill to Metcalf is ~20 points.
The Verdict Given the relative drop-offs from each of the other players around McBride to the players drafted around Kittle, both VBD and positional scarcity say that you should draft McBride in the third round.
In these opportunity cost exercises, we can evaluate the outcomes with what has been dubbed “2v2” – pitting two teams of two players against each other to see which team represents the best pairing. Rather than doing each possible combination, I will pair the highest projected companion players for each tight end option to see how things shake out:
McBride (257.90) + Hurts (416.90) = 674.80
Allen (419) + Kittle (228.80) = 647.80
This very simple binary comparison shows the McBride/Hurts combo to represent a 27-point advantage.
Next, we’ll go through a similar exercise with the large drop-off at running back, examining the RB18 to RB21 drop, which we determined earlier was a 32.1-point difference.
Chuba Hubbard is the RB18 and currently has a fourth-round ADP of 47.
RJ Harvey is the RB21 with a fifth-round ADP of 53.
Like McBride and Kittle, they are drafted in similar areas of the draft board, so it represents a realistic scenario.
Players drafted around Hubbard and their VOR scores: Jalen Hurts (71.15), Kittle (76.20), Mixon (44.20), Metcalf (44.50)
Players drafted around Harvey and their VOR scores: Xavier Worthy (33.50), James Conner (51.90), Zay Flowers (36.20)
Not drafted near Harvey, but later in the sixth round is Sam LaPorta with an ADP of 62 and a VOR score of 34.90. The next quarterback taken after Hurts in the early fourth round is Patrick Mahomes taken in the mid-sixth round with an ADP of 64 and a VOR of 36.35.
Scenario 1: Draft Hubbard in Round Four Drafting Hubbard costs you the opportunity to secure a projected top-four quarterback or a top-three tight end.
Scenario 2: Wait to Draft Harvey in Round Five Passing on Hubbard allows access to elite players like Hurts and Kittle, though you accept a significant 32-point VOR drop-off at running back by waiting for Harvey.
Comparing Trade-Offs The gap from Hurts to Mahomes is substantial (34.80 points), while the drop from Kittle to LaPorta is even steeper at 41.3 points. The drop-off from Metcalf to Worthy is 11 points (and 8.3 to Flowers).
The Verdict Even with the 32-point VOR drop from RB18 to RB21 (Hubbard to Harvey), VBD principles suggest that if you need a quarterback or tight end and Hurts or Kittle are available, they should be the choice over Hubbard.
An added caveat to this scenario which makes the 2v2 more difficult is that Conner is being drafted one spot after Harvey, yet is projected for over 40 points more than Harvey. In fact, Conner is projected to score more points than Hubbard as well. So if you trust the projections, the verdict is even clearer – take the elite quarterback or tight end in round four and draft Conner in round five.
This is the essence of how opportunity cost works. Looking at the drop-off for running back, drafters may feel an urgency to grab the last running back before the drastic value drop-off occurs. However, by looking closer at all options available, drafters can avoid this potential urgency trap and capitalize on the knowledge of more significant value drop-offs at other positions and possible values later in the draft.
The goal is to make up the difference in points from drafting player/position B later versus drafting player/position A earlier by drafting an alternative position with larger comparative point differentials. This helps avoid the trap of drafting based on perceived positional urgency, but instead focus on actual scoring differentials that can make a difference in your lineup weekly.
While the two examples above are realistic based on current ADPs, they ignore other factors that cannot be known within their narrow focus, such as players you have already drafted or league-mates' preferences, among other factors. One could also extrapolate these examples several rounds into the future, plotting out multiple picks beyond the ones discussed.
So now that you know about VOR, scoring drop-offs, and VBD, that’s all you need. Right?
Is VOR a Flawless Strategy?
I would be remiss if I didn’t discuss the pitfalls of using VOR, scoring drop-offs, and VBD strategies. Like any strategy, there is nuance and nothing is perfect.
If a drafter is strictly beholden to using VOR scores and isn’t paying attention, they may ignore aspects like positional team needs, the needs of other teams in their league driving their decisions, or other values based on shifts in ADP. Lastly, ignoring values presented by the market undervaluing players based on their projected points could prove costly. These oversights could result in critical errors during their draft.
What Can VOR Do For You?
The draft is my favorite part of fantasy football. It’s one of the things that makes fantasy football fun. Weighing the opportunity cost of drafting Player A instead of Player B, guessing what curveball your league-mates will throw at you next, calculating the chances you will get sniped on your next pick, or be fortunate enough to have an unexpected player value fall to you in the next round.
Heading into your fantasy drafts with a strategy is important. Equally important is flexibility, allowing new information during the draft to help with those ever-critical decision points.
Understanding VOR, VBD, and scoring drop-offs within positions and at different points in the draft can help you avoid making bad decisions influenced by perceptions of positional urgency. Instead, you can stay cool, calm, and collected during your draft, making sharper, data-backed choices that can tilt your league in your favor.
Before your next draft, calculate VOR scores for your projections, identify the key drop-off points by position, and map them against current ADP data. When draft day arrives, you'll have the tools to recognize value others miss and the confidence to make contrarian picks that win leagues.
Ultimately, the best drafts are those where strategy meets adaptability – and VOR can be your compass when things get unpredictable.