Optimal Quarterback Timing: 2021 Superflex Draft Strategy
Check out our 2022 standard, half PPR, full PPR, and superflex strategy here!
By far the most important superflex draft strategy revolves around answering the question: “When should I draft quarterbacks?” Most fantasy football managers are comfortable with the generic fantasy draft strategy of running backs and receivers early, grab a QB and TE in the mid rounds, and fill in the gaps later. But in superflex, the extra value added to quarterbacks changes which draft strategies are optimal.
In this article, we study superflex draft strategy and attempt to answer the “When should I draft quarterbacks” question in the 2021 fantasy football season. We’ll go through each pick, 1st through 12th overall, and determine the optimal time to take quarterbacks during your fantasy draft. After this, we perform some 100,000 simulations of mock drafts in order to tell you how to structure your first five picks and determine which position to target in each round. To see the specifics for each pick, please follow this link to skip ahead to the simulations and data for each pick and this link to skip ahead to the section in which we determine the best superflex draft strategy for each pick based on tens of thousands of mock draft simulations.
Bottom Line Up Front
Something I’ve learned in communicating technical material over the years is that sometimes it is best to present your findings first and then explain your methods afterwards. We find that the optimal superflex draft strategy begins with drafting your two quarterbacks in rounds 3-5. Moreover, we suggest to take one tight end in the first two rounds if the board falls right and running backs with the other 2 of your first five picks. We also argue that, while this is optimal in an average case sense, the best strategy is to use these suggestions as a guide while ultimately letting the draft board and what other players are doing dictate your strategy.
Superflex Quarterback Draft Strategies
We want to test the best two rounds in which to draft a quarterback through modeling and simulation of mock drafts. However, if we set the upper bound for when you take your second quarterback to be the 10th round, then there are 45 possible combinations of pairs of rounds (between 1 and 10) in which we could possibly take our two quarterbacks. Multiplying by the 12 possible options for which first round pick you have would result in 540 possible settings to simulate Presenting 540 data points to you, the reader, will suffer from the dreaded “conveying no information by presenting too much information” problem.
Moreover, the difference between taking a quarterback in rounds 1 and 7 is probably imperceptibly different from taking a quarterback in rounds 1 and 8. Similarly, a particular strategy at the first overall pick is likely to be nearly identical to using that same strategy at the second overall pick. The differences in those 540 possible combinations will be too small to easily determine trends. We won’t grok the point.
Instead, we group things together to get a better idea of what is going on. Instead of testing every possible combination of which two rounds we could draft a quarterback, we group them up into ‘Early QB’, ‘Mid-round QB’, ‘Late QB’, and ‘Very late QB’. These categories correspond, respectively, to drafting a quarterback in rounds 1-2, 3-5, 6-8, and 9-10, respectively. Furthermore, we also group the pick numbers into pick ranges as an ‘Early Pick’, ‘Middle Pick’, and ‘Late Pick’. If you have an early pick, your first round selection happens between 1st and 4th overall. ‘Middle Pick’ and ‘Late Pick’ are similarly defined to be drafting in the 5-8 and 9-12 ranges, respectively.
Simulation Method and Details
In this way, we simulate every possible combination of these groups to see which quarterback-based superflex draft strategy works best for each pick range. For example, if we test the ‘Early QB-Early QB’ strategy for an Early pick, then we are taking a quarterback in rounds 1 and 2 when our pick is between 1st and 4th in the first round. A bit more complicated, suppose we want to test the ‘Mid-Round QB + Very Late QB’ strategy when we are drafting from the ‘Late Pick’ portion of the round. To do this, we first randomly choose a draft slot in the 9-12 (the Late) range. Then, we force our simulations to take one quarterback somewhere in rounds 3-5 (‘Mid-Round QB’), one quarterback somewhere in rounds 9-10 (‘Very Late QB’) and non-quarterbacks in every other round. In this way we end up with only 30 simulations, with 30 unique pieces of data to present. This is much more palatable than the 540 we would have done otherwise. The data will be presented in a table format where the column headings are the rounds in which we take our first quarterback and the row labels are when we take the second quarterback. The tables will be symmetric and contain redundant information because, for instance, the ‘Early QB – Mid QB’ cell is identical to the ‘Mid QB – Early QB’ cell.
Our simulation data, including ADP and projections, are borrowed from the excellent work over at Beer Sheets. In an identical manner to what we did in our previous two articles about draft strategy in half ppr leagues, we add some randomness to each player’s ADP to determine a draft order. Then, we determine which draft strategies ‘work’ by adding up the total points of your starting lineup (minus kicker + defense). The most points for your starting lineup is determined to be an optimal draft strategy. (You may argue that there is value to depth but that is an entire other article together). Now, we continue on to study the best superflex draft strategy.
Early Pick Superflex QB Strategies
| Round Range | Early | Mid | Late | Very Late |
|---|---|---|---|---|
Early | 125.6 | – | – | – |
Mid | 126.3 | 126.7 | – | – |
Late | 125.8 | 126.3 | 125.4 | – |
Very Late | 124.7 | 125.2 | 124 | 122.9 |
Middle Pick Superflex QB Strategies
| Round Range | Early | Mid | Late | Very Late |
|---|---|---|---|---|
Early | 124.6 | – | – | – |
Mid | 124.9 | 125.4 | – | – |
Late | 124.8 | 125.1 | 124.7 | – |
Very Late | 123 | 123.5 | 122.8 | 120.7 |
Late Pick Superflex QB Strategies
| Round Range | Early | Mid | Late | Very Late |
|---|---|---|---|---|
Early | 123.5 | – | – | – |
Mid | 124.3 | 124.7 | – | – |
Late | 124 | 124.3 | 123.7 | – |
Very Late | 121.5 | 121.7 | 121.1 | 118.2 |
A Summary of Our Findings
In all three cases, the trends and conclusions about the best superflex draft strategy are extremely similar. The key location to be taking quarterbacks in superflex is in the middle rounds, in the 3-5 range. This is evident in a few different ways. First, look at the ‘Early’ column in each of the three tables. When we force the first pick to be in rounds 1 or 2, the ‘peak’ value is when you take your second quarterback in the middle rounds. This trend is true no matter which pick range our pick falls in – early, mid, or late. The second place in which this is evident is that the largest number in each table – the best strategy according to our simulation method – is when both of your quarterbacks are drafted in rounds 3-5.
Perhaps more surprising is how small the difference is when you draft both quarterbacks anywhere in the ‘Early’, ‘Mid’, or ‘Late’ round ranges. For example, if you have an early pick the difference between the optimal ‘Mid-Mid’ strategy and ‘Late-Late’ is only 1.3 points. This difference is so small as to be almost imperceptible. This means that if you want to use a significantly different strategy, you won’t lose much.
The clearest takeaway from the above analysis is a clear refutation of waiting for either of your QBs to be drafted in the ‘Very Late’ range, in rounds 9 or 10. Based on current ADP, a few quarterbacks that might be expected to be drafted in this range are Derek Carr, Carson Wentz, and Sam Darnold. Even though these guys are starters, the difference between them and the earlier blue-chip QBs is significant.
The Mid-Round QB Draft Strategy for Superflex
OK. So the optimal superflex quarterback draft strategy is to take two of them in rounds 3-5. This should give you plenty of time to load up on three traditional non QB studs in the first 5 rounds. The table below is designed to show you just how much more valuable picks in the first 5 rounds are in superflex than they are in traditional one QB leagues. For picks 12, 24, 36, 48, and 60 (the end of rounds 1-5), we computed the equivalent pick in a standard league by adjusting for the number of quarterbacks taken. For example, there are 16 quarterbacks whose ADP is in the top 60 in superflex. In a traditional half-ppr league, on average 5 quarterbacks are off the board by this point. Thus, 11 more position players on the board at the end of the fifth round in superflex than there would be in regular half-ppr. That means if you aren’t taking a QB, pick number 60 in superflex is roughly the equivalent of pick 49 in 1QB leagues because 11 more quarterbacks than normal have been drafted. That turns your last pick in the fifth into the equivalent value of the first pick in the fifth. The full conversions are below:
Superflex Pick | 1QB Equivalent |
|---|---|
12 | 8 |
24 | 17 |
36 | 27 |
48 | 38 |
60 | 49 |
If you had the 12th overall pick, your first three picks are 12, 13, and 36. If you take non-quarterbacks with your first three picks then a QB in rounds 4 and 5, your first three picks look like the equivalent of picks 8, 9 and 27 in a traditional league. Your position players will look like those you would find with a mid first, a late first, and an early third. Then, if you draft quarterbacks at #37 and #60 overall you will likely end up with, for instance, Burrow and Cousins – two top 17 quarterbacks. Your quarterbacks will be very respectable but your position players will be absolute studs.
Filling in the Gaps
Let’s go a step further and figure out the best superflex draft strategy for each pick when we take two quarterbacks in the round 3-5 range. Our goal is a set of 12 statements of the form “The best average starting lineup at pick ____ is obtained on average by taking ‘RB-RB-QB-TE-QB’ in the first five rounds”. We go through one round at a time to figure out the best pick to make in each round at each position but with the caveat that we force two quarterbacks to be taken within rounds 3-5 and none before. If, for instance, we determine that an RB is optimal in round 1, then we compute the optimal round 2 pick given that we already took an RB in round 1. This is what is known as a greedy algorithm and should be expected to yield nearly optimal results in this setting. The optimal position order for the first five picks is:
Even though the specifics of the optimal draft order vary a little bit from pick to pick, the general trends remain largely the same. First, take a tight end in the first two rounds. If Kelce is there after #10 overall, he should probably be your pick. Otherwise, I would wait until the second round and try to draft Kittle or Waller.
At every single pick, it appears optimal to aim to draft a quarterback in rounds 3 and 5 but not in round 4. This is going to balance the ability to take studs at running back and tight end early while also still allowing yourself the ability to
In the first five rounds, running backs are king. In only one location in the above table does a receiver show up. In almost all cases, taking a wide receiver in the first five picks sacrifices over one point of expected, average performance. The depth of wide receivers means there is little value over replcaement to drafting an early receiver.
The Best Quarterback Based Superflex Draft Strategy
The most important question now, though, is “What is the best superflex draft strategy?”. While our table above tells you what the best average case strategy is when we aggregate over thousands upon thousands of drafts, this is not gospel. Every draft is different. Players will drop who shouldn’t. Players will get over drafted who shouldn’t. The most important thing is to be patient, use the above as a generic guide when drafting, and let the board come to you.
Let me tell you what I mean with an example. Suppose you go into your draft having read my articles and determine that you simply must take an early tight end. Your whole draft strategy is predicated upon an early tight end. Well, what happens if you have pick 20 and Kelce, Kittle, and Waller are gone by your pick? Obviously you need to adapt and change. Don’t take Kyle Pitts at #20 overall just because you feel like you need an early tight end. Instead, you need to recognize that because two of those tight ends went earlier than they probably should have, there is going to be extra value at either receiver or at running back. Every mistake someone else makes is something that you can directly capitalize on.
So, while the above superflex draft strategy is meant to highlight general trends, when it comes to draft day you just need to trust your gut. Remain pliable and adaptable. But seriously, though, take early tight ends and mid round quarterbacks.
