Every year, I take a three-year sample of the end-of-season actual points scored at each position. This way, I can assess the expected value of, for instance, the TE1 compared with the WR5; and the RB10 compared with, say, the WR10. I used the Pro-Football-Reference database and sorted by half-point PPR scoring. Let’s leave quarterbacks out of this because I am strongly opposed to a strategy that seeks to maximize points by drafting quarterbacks early — but more on that later.
The overriding objective when drafting running backs, wide receivers and tight ends in fantasy football is to get the most points, regardless of position. If you’re making the 25th pick and 18 running backs have been selected, you probably don’t want to take a running back. And if the three tight ends who are in that top tier at the position are taken all in a row before you select, you probably want to cross that position off your cheat sheet and focus somewhere else.
These are the actual averages for the players at each position, ranking by PPR points — all the WR1s, RB15s, TE5s, etc. But this has nothing to do with where the players were actually drafted. So the “RB10” represents how many points the 10th-highest-scoring RB averaged, regardless of where he was drafted. The objective is to know the expected points when you draft the 10th running back in your league. You may have that player ranked higher, of course, but we want to know exactly where the points bar is that you have to beat with that pick, and whether it would be wiser to focus instead on another position.
TE1 matches the expected points of WR5, meaning that a strategy of just waiting on TEs when no one has drafted one and continuing to pound the WR queue after well more than five of those players have been taken — that’s leaving expected points on the table. Travis Kelce going around WR5 is perfectly reasonable. I prefer drafting Mark Andrews over Kelce, as Andrews was the No. 1 TE in 2021 and Kelce is over 30. But you probably need an early-round pick to take Andrews after Kelce in the second round when it snakes back to you. Picking fourth, fifth or sixth probably allows you to snake Andrews in Round 2 about 85 percent of the time.
TE5 equals WR30 in points but the fifth TE off the board has been going (according to ADP) where the WR20 typically gets drafted. The play here is to pass on that TE and take the WR. The reason is that you only have to get what you pay for at WR to beat the TE handily, as opposed to drafting the TE5 and hoping he scores much higher than where he’s drafted among TEs. We hope all of our players score higher than where we pick them but we want that to be a bonus — we don’t want to actually pay for that.
Here’s a different way of looking at the data, with the positions and three-year-average point finishes all in one chart:
RB1407WR1391RB5270WR5285RB10231WR10253TE1289WR15237RB15207WR20225RB20193TE5188WR25209WR30189RB25169WR35177RB30153WR40162WR45153RB35139TE10153WR50143RB40123TE15133RB45107TE20115
So with perfect foresight, you would obviously prioritize the No. 1 scoring running back. That’s going to be the most valuable non-QB, on average, beating the No. 1 wide receiver by 16 points the past three years, 407 to 391.
But as soon as you think the No. 1 point-scoring running back is likely off the board, you should prioritize wideouts, because the No. 1 receiver is going to beat the RB5, for example, easily (by an expected 120 points). And that continues in the early rounds. The WR5 goes about where the RB10 (or even later) is drafted, and that’s about a 50-point expected edge for the WR pick. WR10 beats RB15 by a similar amount. Then, most drafts back off RBs and focus on WRs. (So it’s nice to already have your WR studs by then.)
Finally let’s look at two different strategy constructions, measured by expected points — just getting exactly what you paid for and not needing to be some great player picker. We’ll do 1EliteRB versus RB Heavy.
1EliteRB (meaning you take an elite RB, then focus on WR and TE for your next several picks) — No edge with the first pick since you want that anchor RB just like the RB-heavy team. But then you’re taking WR1-5 with pick 2, WR5-10 or TE1 with 3, WR10-20 in the 4th, WR20-30 in the 5th. That’s an expected total of 1,360 expected points (330 points +330 + 275 + 225 + 200)
RBHeavy (means you likely go RB, RB, RB, WR, RB) gets you 330 + 237 + 180 + 225 + 140 for a total of 1,110 points.
That’s a massacre. But, the RB drafters will say, “OK, but now we get to feast on the higher scoring WRs and you have to draft slug RBs.” That’s somewhat correct. Add three RBs and a WR in the next four rounds for the 1EliteRB team and three WRs and one RB/TE for the RB Heavy team. That’s an expected 135, 135, 135, 150 for the former team (RB, RB, RB, WR) and 150, 150, 150, 135 for the latter (WR, WR, WR, RB/TE).
So the RBHeavy team has picked up about 30 points. But they’re still down an expected 220 points through nine rounds. That’s a lot. That’s 37 touchdowns you have to make up. It’s playing decidedly uphill.
This year, structure your selections in a way that optimizes the expected points rather than minimizing them. Keep the chart handy when your pick comes up and you have a dilemma on who to pick, and you could reap the benefits of proper draft construction.