Recently there has been much discussion about the optimal strategy in superflex leagues, where teams have the ability to start a second quarterback. Does a heavy-QB roster build a predictable base with a high floor, or is the opportunity cost too great to pass on elite running backs and wide receivers? How do you go about drafting QBs in superflex leagues?
The addition of a second quarterback spot in the roster leads to a fresh debate in the fantasy football community, and it leaves open an opportunity for savvy drafters to establish an edge in the superflex format.
There is room for a tome’s worth of analysis on the superflex format, but for now I want to tackle an initial question: when should you draft your first quarterback? Let’s tackle that question in two parts.
It is hard to find reliable average draft position (ADP) data for superflex leagues, because they are still relatively uncommon and not covered up by any major fantasy outlets. One of the best sources of ADP is the Scott Fish Bowl, a massive league that has had a total of 50 drafts in the last two years.1 The drafts involve 12 teams, so all ADP data is based upon a 12-team format.
How many quarterbacks are drafted in the typical superflex league, by round? Fifty superflex drafts yield the following averages:
Round 1: 2-3 Quarterbacks
Round 2: 1-2 Quarterbacks
Round 3: 4 Quarterbacks
Round 4: 3-4 Quarterbacks
Round 5: 2 Quarterbacks
Round 6: 3 Quarterbacks
This year, drafters appear to be waiting longer on quarterback than they did last year, so you can safely shift those averages down slightly in your upcoming drafts.
Once we know where quarterbacks will be drafted in a superflex league, we next have to address the question of when to commit. Which round provides the best value on our first quarterback?
How Good Are We at Projecting QB Scoring?
The overall assumption, implicit in any set of rankings or projections, is that we have some ability to predict the future accurately. At least with quarterback, the fantasy community appears to identify tiers well, although we struggle to parse out what order to set within each tier.
You can see from the following chart that over the last four years we have done well predicting the quarterbacks with top-five potential, but after that handful the next five all look about the same, as do the rest after that.
The chart documents points per game (y-axis) against QB ADP (x-axis), based on how many quarterbacks were drafted before them in that specific year’s drafts.
The first five quarterbacks off the board have reached higher heights than any of the next dozen, proving that the community succeeds when it projects its elite quarterbacks. With the exception of QB4, the top five have also established better floors than the rest of the quarterback crowd. When you pay up for a top-five quarterback, you’re doing it to lock in that high-floor, high-ceiling combination.
Then, the next five or six quarterbacks have seemed to fit into a much tighter window, providing significantly less upside while maintaining a floor right around 20 points per game.
And after the top ten, the bottom begins to fall out, and volatility reigns. Those are the quarterbacks you stream, not the ones you take home to meet your parents. In their good matchups, each can provide a value, but outside the top ten you aren’t guaranteed a floor in terms of points per game averages. Depending on the year and the player, some mid-round quarterbacks hit, providing solid points per game across a whole season, but others struggle to return any value.
What’s the Takeaway?
In short, this data suggest you should be aiming for either the fifth or the tenth quarterback off the board. The first two tiers are each about five-deep and roughly identical across the board, and reason suggests you wait as long as possible if a group of players project for roughly the same value. If you want to attack the end of those top two tiers, look to the beginning of the third or the middle to end of the fourth round, where you’re likely to get QB5 or QB10.
This year, that means you’re looking for Cam Newton or Drew Brees in the late second or early third, or waiting to grab someone like Ryan Tannehill, who might have a very high floor this year, or Tony Romo at the end of the fourth.
There remain several counter-arguments to this article, among them the fact that it ignores opportunity cost and team composition entirely. There is fruitful ground to be sown in the field of superflex team composition theory, and for brevity’s sake I have not touched any of that here. This article is meant only to suggest, in a vacuum, where you should draft your first quarterback, based on recent history of quarterback scoring and ADP. One could respond that a team performs better on the whole if you ignore quarterbacks in the top ten and wait for the higher volatility options in the later rounds, but I have left such discussion for another time.
Additionally, I have not deviated from averages to discuss individual players who might outperform or underperform, relative to their ADPs. There is a place for that type of analysis — and I have done some of it elsewhere2 — but there is also great value in looking at broad, historical trends to find repeating patterns within the fantasy landscape. Within this article’s general framework, you should absolutely attempt to predict where you can find value or avoid pitfalls when it comes to specific quarterbacks.
- Fifty drafts creates a decent sample size from which to begin, but SFB’s other settings mean the data doesn’t perfectly apply to other superflex leagues. This year, the rosters require you to start 1 QB, 2 RBs, 3 WRs, and 1 TE, but you then have one superflex spot and three additional flex spots. That distribution changes the relative positional values, beyond where they normally fall. Additionally, scoring includes 0.25 points per carry. (back)
- I called Jay Cutler a steal and told you to stop drafting Matthew Stafford (back)