Forming a Fantasy Football Draft Strategy

By: Toby Junker

As summer begins to get underway, fantasy football fans start their annual migration towards mock draft websites. Mandatory workouts kick off the start of #hypeszn as daily performers in OTA’s begin to create buzz around their names. Though many developments will occur between now and fantasy football draft season, I’m always interested in beginning to formulate a draft plan.

This led me to a FantasyPros article by Donald Gibson looking at the effectiveness of a running back (RB)-heavy approach in the early rounds of fantasy drafts. Gibson looked at how many of the first 22 RBs drafted last year actually managed to finish in the top 22. He found that most of the RBs that did manage to return on their draft price were selected very early in the draft. Thus, there were very few ways in which you could’ve taken 3 RBs in your first 4 picks and “hit” on all 3. To get a better sense of what I’m talking about, be sure to check out his article (link here: Ultimately Gibson seemed to think that it was best to draft a “stud” RB early and then wait until later rounds to finish addressing the position, because the “hit” rate of the RBs taken in rounds 3-4 was so low.

This got me thinking about what the optimal thing to do in these early rounds was. Sure, RBs might not be likely to hit but were other positions better? If, for instance, wide receivers (WR) in that range seemed to do really well, it would make a lot of sense to stock up on RBs in rounds 1 and 2 followed by WRs in rounds 3 and 4. Alternatively, maybe WRs in this range had a similar chance of being successful. If no position was likely to return good value in this range, maybe fantasy owners should look to trade out of these rounds and gain more draft capital in areas where they could more likely get value.

In order to tackle this question, I decided to compile some data based on Average Draft Position (ADP) & Final Rank for players of all positions over the past 3 seasons (2016-2018). I chose 3 seasons because I wanted a larger sample size than just last season. Going back to the Gibson article, maybe last year was particularly bad for RBs that went in rounds 3 and 4; more data could help us see the bigger picture. That being said, I capped it at 3 seasons because the league is always changing and evolving. I wanted recent data so it was relevant. Potentially RBs could have been viable to draft in the early 2000s/2010s, but the passing game's evolution has meant there aren’t as many viable options (not saying that’s definitely true, but it’s a possible change I wanted to account for). I used standard PPR scoring and obtained my final rank data from (season long totals). All of my ADP is from (ADP was taken the week prior to NFL kick-off from approx. 750 mock drafts).

I’ll walk you through all the takeaways I had. Important to note that across the top of the tables are where the players were drafted. “D-1:8” denotes that we’re talking about QBs that were drafted in the top 8 in their position. Likewise, along the side of the table is where the players finished. “F-1:8” denotes that the group finished in the top 8 at the position. Each cell in the tables is the percentage of players in the “drafted range” that ended up in the “final rank range.”

Table 1: Percentages of Where QBs were Drafted and Where They Finished

What jumped out at me was the second group of QBs drafted had a higher success rate than the top group. Over the past few seasons, the fantasy community has adopted the late-round QB approach. The general argument is that the value you can get by taking slightly worse QBs later in the draft is better than that when taking the top guys earlier in the draft. This table indicates not only is it better value to do that, but potentially that the second group of QBs is just better. I think what’s happening here is a statistical phenomenon known as “regression to the mean.” Essentially every year we have QBs that perform above their average level of play (Matt Ryan 2016, Matthew Stafford 2017, etc.). These QBs shoot up draft boards the following year due to their recent success, but fail to live up to the hype when they return to their normal level of play (hence why the top 8 QBs draft are most likely to finish in the mid range). Likewise, the good QBs who underperformed the previous year are “slept” on the following season (Cam Newton after 2016, Matt Ryan after 2017, etc.). They get drafted in the 9-16 range, but when they return to their normal level of play the following season, they end up finishing in the 1-8 range and giving great value for fantasy owners that selected them.

My other takeaway from looking at the QBs was that you don’t want to go “overboard” with the late-round QB strategy. Looking at the QBs drafted 17-24 in the past 3 seasons, it seems that they rarely end up playing well. A little over 40% of the time, players in this range finish outside the top 24; this means in 12 team standard PPR leagues the player would basically be a non-useable asset. Furthermore, only 5% of the time did we see QBs in this range break-out (finish top 8), and only about 1 in 4 of these QBs did better than we would have expected.

So what does all of this mean? Well, when drafting QBs I suggest staying away from the first and third groups. If you really want to take a QB early, make sure it’s someone who has performed consistently over the past few seasons. Don’t be too quick to believe last years breakout QB will do it again this year. If you’re someone who likes to wait on QBs, make sure to pick a couple of mid range guys instead of waiting until the very end of the draft. In terms of rounds, I suggest trying to grab 2 QBs in rounds 9-13. Look for guys that can provide high floors with their rushing stats, are likely to have large volume, and/or are attached to high powered offenses. QBs Jared Goff, Jameis Winston, and Kyler Murray are all guys I’m targeting in those late middle rounds.

Table 2: Percentages of Where RBs were Drafted and Where They Finished

Table 3: Percentages of Where WRs were Drafted and Where They Finished

I want to discuss RBs and WRs at the same time because so often these two positions make up the overwhelming majority of the selections in the first 5(ish) rounds. To begin I wanted to look at what to draft in the first round. I generally think there is a narrative that RBs are more likely to get injured, or that WRs are safer to select early in the draft. But given my findings I disagree: the top 12 RBs drafted finished as top 12 RBs more than the top 12 WRs finished as top 12 WRs. Furthermore, the chances that either finished outside the top 36 at the position (the chance they “busted”) were virtually identical. I noticed that most of the time, when the top drafted RBs or WRs were finishing outside of the top 36, it was due to injury. Though definitely not a perfect measure, this group that finishes outside the top 36 could be looked at as how often these high ranked RBs/WRs end up with significant injuries. Getting back to overall “bust” rate, I would argue that a top 12 RB/WR finishing in the 25-36 range for their position should also be considered a “bust.” Incorporating those rates, we find that top 12 RBs are about 6% less likely to “bust” than top 12 WRs. Based off of the past 3 seasons, it seems that top RBs offer a better chance at reward while presenting less risk.

Looking now to the middle groups of these two positions, we see similar results. Both positions return their expected value about 28-30 percent of the time. WRs seem to have a slightly higher “hit” rate and a slightly lower “bust” rate. I lean towards taking mid WRs over mid RBs, but this should be a flexible time in your draft where you select based on roster need.

The third group of WRs and RBs is quite concerning. For both positions, there was over a 50% chance that the players in that range finished outside of the top 36 at their position. While for some of these guys that isn’t very far from their ADP, someone outside the top 36 won't provide much value, as they can’t (*shouldn’t*) be trusted on a week-to-week basis. Furthermore, there was only about a 22% chance that RBs or WRs in this range finished in the top 24 at their position. This struck me as a group that offered little reward with large amounts of risk. These guys usually go in the second half of round 5 to about the end of round 7 in 12 team leagues. Looking over the ADP from previous seasons, this group usually consists of overlooked veterans and up-and-coming players that are projected to carve out a role in their offense. I didn’t run the numbers to figure out exact percentages, but looking over the players in this range that did end up performing well, it was more often that the overlooked veteran managed to finish top 24 in their position, than a young player that was projected to take the next step.

So, in terms of drafting RBs and WRs early on in drafts, I suggest the following. If you take all RBs and WRs in the first 8 or so rounds, I recommend targeting 2 RBs and 2 WRs in the first four rounds. The order in which you do this isn’t super important, so make sure you’re getting good value at each pick. Then, use rounds 5-8 to shore up whichever position you feel is weaker. You probably want to exit round 8 with 4 RBs and 4 WRs, but if you feel like your RB’s are weak after the first 4 rounds, maybe grab back to back RB’s in rounds 5 and 6 before returning to WR. If you like to take a tight end (TE) early in drafts (in the first 4 rounds), I suggest you target a RB in the first round, a WR in the second round, a TE in the third round, and a WR in the fourth round. Theoretically your top RB is pretty likely to hit so you don’t need to worry as much about that position in the top 4 rounds. I’d go WR in round 2 as you could still probably get someone in that top tier of WRs. This year, TEs seem to be going really early but in round 3 you can still probably get Ertz or Kittle, giving you a positional advantage each week. Now, this goes for both types of drafters, figure out if your league allows draft pick trades and trade away your 5-7 picks. If you pick in the later half of the 5th round try packaging your 5 and 6 for a 4 and 9. On paper, you’re only moving up one round but dropping back 3, meaning your leaguemate might be likely to accept it. The extra 4 could let you take another mid tier WR; potentially you could then go RB round 1, WR round 2, TE round 3, and a RB & WR in round 4. The extra 9 could allow you to take another stab at QB, securing 2 of those mid-tier QBs; hopefully one of them breaks out.

Table 4: Percentages of Where TEs were Drafted and Where They Finished

The TE position is bleak. Looking above, we can see that even the top group of TEs, the first 6 off the board, end up being totally useless (finishing outside the top 18) about a quarter of the time. Additionally, there’s less than a 40% chance that a top 6 drafted TE finishes in the top 6. Even less encouraging, is that the middle group of TEs, those drafted 7-12, rarely break out. Over the past 3 seasons, on average, only one of the 6 TEs drafted in that range has broken out (finished top 6). It’s over twice as likely that a TE in that range finishes outside the top 18, than inside the top 6. My strategy for TEs would to either be to go after the elite tier in round 3, or if you miss on that, just take a few stabs at that last tier and hope for the best. Sure it’s somewhat likely that if you take 2 TEs neither ends up being much of anything, but the draft capitol you have to invest in these guys is so small that it isn’t very big of a risk. Maybe you could even consider taking 3 guys in this range very late, and just dropping the 2 least productive ones after a few weeks.

Recommended Draft Strategy

I’m going to break up this section into two separate parts: those who are not able to trade draft picks, and those who are.

No Draft Pick Trades

This is going to be pretty simple: go RB & WR heavy. I suggest that every two rounds you target selecting 1 RB & 1 WR and to do this for the first 8 rounds. This will give you 4 RBs and 4 WRs and likely you’ll get at least two at each position that are pretty good. In rounds 9 - 13 I would try to get 2 QBs , 1 TE, and 1 RB/WR. And in the last two rounds I would take 2 fliers on RBs/WRs/TEs. I’m sure you’ve heard this a million times, but don’t worry about drafting a kicker or a defense. Those are positions that you should be streaming every week based on matchup. Wait until a few days before the regular season kicks off and drop a couple guys on your roster to make room for a kicker and defense. Maybe one of your players got hurt, maybe one of your fliers isn’t looking so hot, or maybe a TE you drafted doesn’t seem to have chemistry with his QB. We’ll all make mistakes at some point, so just drop the two players that seem to have the least value at that point.

Draft Pick Trades

For those of you who are able to trade draft here is my strategy for you: trade out of rounds 5-7. I talked about this a bit in the RBs/WRs section but try to pair your 5 & 6 round picks for a 4 & 9. This shouldn’t be too hard to pull off, as (on paper) it looks like your league mate is getting the better end of the deal. After you’ve done this I suggest you target 1 RB and 1 WR in the first two rounds. Now, because you have an extra 4th round pick, you have the luxury of going after a TE in round 3. If none of the elite TEs are there, you can draft a RB or WR here but ideally you’d get a top TE. With your two fourth round picks I would again target 1 RB & 1 WR. If all goes according to plan, you’ll have 2 RBs, 2 WRs and a TE after round 4. I also suggest trying to trade your 7 & 15 round (or last round) picks for a 9 &  10/11/12 (to be clear, you only get one of those three). You now have 8 picks in rounds 8-12. The draft plan from here on out is pretty simple. Make sure you get 2 QBs in rounds 8-12, if you weren’t able to get a TE earlier draft a TE here. Otherwise draft RBs and WRs you have good feelings about. The plan for the last 3 rounds (remember, you only have 2 picks) is to grab fliers you like. If you didn’t get a TE early, one of these two picks should be a TE. Otherwise it’s up to you what to draft here. Remember again that kickers and defenses shouldn’t be drafted. Wait until the start of the season and drop your 2 least valuable players to make room. If you can’t pull off either of these trades, don’t worry. Just refer to the section with no draft pick trades to see what to do in those rounds and other affected rounds!

Final Notes

To conclude I wanted to touch on 3 things to keep in mind during your upcoming drafts.

First, you mock draft for a reason. Mock drafts are practice; they allow you to experiment with draft plans, roster construction, etc. with no risk. Try using different strategies in your mock drafts. Figure out what plan gives you the roster that you like the most. If you do the same thing in every mock draft you’ll never know if there is a better alternative. And, if something unexpected happens in your draft, you won’t have had any practice in how to adjust accordingly.

Second, the cutoffs for the sections I used in this article are arbitrary. I did put thought into how I wanted to break up each position, but ultimately the cutoffs are made up. I wasn’t particularly high on the 25-36 RBs/WRs group, but if you feel confident about your 25th WR, don’t be afraid to take that player. Think of my groups as tiers. Instead of strictly thinking that the top 12 RBs will return their value 75% of the time, think that the elite tier of RBs will “hit” about 75% of the time. For me, there are only 7 RBs in that elite tier this year. I’ll feel very comfortable taking any of them, but after that I’ll probably treat my next few tiers like they’re the 13-24 ranked RBs.

Third, have a draft plan but also be flexible. This is a delicate relationship, and it can be difficult to master. I recommend having a draft strategy before going into every real draft. Have an idea of what positions you want to tackle early/mid/late. Have some guys that you’re really high on; have some guys that you’re really low on. Have some sleepers that you’ll be targeting in the last few rounds. That being said, don’t be afraid to be flexible. Rarely ever is it the best plan to completely abandon your draft plan, but if you were planning on taking a WR in round 2 and someone like Melvin Gordon or David Johnson slips into the second round, don’t pass on that value just because you were planning to take a WR. Remember to keep in mind that you are a little off course and to look for value at the position you ended up not drafting later on.

If you are interested in looking at the data I collected for this article, or the year by year breakdowns of the “hit” rate for each position, you should check out the links below. And if you enjoyed this article, be sure to drop a follow on my Twitter (@itsfootballtime)!

Links to Data

Raw ADP vs Finish Rank

Compiled “Hit Rates”

4 thoughts on “Forming a Fantasy Football Draft Strategy

  1. Great write up! I did a similar process a few years ago but looked closer at overall ADP gaps to draw my positional lines. I settled on an upper tier of top 3 for the onesie positions, which usually get drafted around the same time as the top 12 for RBs and WRs. In most leagues only 16 QBs or TEs are rostered at any given time.

    I then took those "hit rates" and multiplied them by the average difference between the top scorer from the lowest starter scorer (RB1-RB25) or a value over replacement for guidelines on which positions to get where. That value over replacement addition to the equation really favors the top end RB strategy.

    Great stuff!

    1. Sorry about getting back to you so late, our comment section was a little buggy! Sounds like a super interesting article, would be happy to check it out if you wanted to drop a link! Also feels good that we ended up with similar stuff; great minds!

  2. This is a fascinating article with a lot of good information. However, I wonder how position scarcity comes into play. You recommended 4 RB and 4 WR in the first eight rounds, but in almost all leagues I have ever been in, there have been more starting positions for WRs than for RBs. Drafting an equal number of WRs and RBs in those rounds may create a team with a strong bench, but not necessarily the one with the highest point production overall. So, how do you account for position scarcity?

    1. I think knowing your league settings is really important; if you're in a league where (for example) 4 WRs get started and only 1 RB does, it would definitely make sense to slightly adjust for those roster settings. Even if your league has one more WR starting than RB, I would still recommend drafting them pretty evenly.

      One thing that you have to balance while drafting is value vs positional need. I tend to favor taking value over positional need because I'm confident that if my players hit, I'll be able to trade them for players at the positions I need that I wouldn't have been able to draft. But again, that comes down to knowing your league. If I can get really good value at each pick, I'll be much more willing to wait on "position X" in a league where I know everyone because I'll feel confident that I can actually engage in trade talks. The flip side of that, is that if it's just a random league I've join or even a custom league where I don't know everyone, I might be more willing to make sure my starting lineup is solid (draft for positional need) because I'm unsure if I'll be able to address my weaknesses via trade as the season goes on.

      The other reason that I like drafting a similar amount of RBs and WRs is that as soon as you get out of that elite tier of RBs and WRs, the bust rates get pretty alarming. Even in the second tier for both position, there's above a 35% chance that they finish outside the top 36 at their position. If I knew that all of my RBs would hit, I would feel much more comfortable drafting for positional need, but because (realistically) at least one of the RBs I draft earlier probably won't work out, if I only walk out of the first 8 rounds with 3 or even 2 RBs, I could quickly become "up a creek without a paddle".

      One final note, I really do appreciate everyone who reads my articles; if you have more thoughts about this one (or even just general fantasy football questions) I'd be happy to hear them!


Leave a Reply