Value-based drafting for Quarterback in the 2018 Fantasy Football Season
According to our machine learning models and the Fantasy Outliers team, these Quarterbacks could return higher value than ADP and ECR in the 2018 Fantasy Football season.

Quarterbacks were our most accurate position group last year. In fact, we beat ESPN’s weekly and yearly projections for quarterback more than any other position group. The overall rankings derived from our Total Points projections last year were more accurate than ESPN’s 72% of the time, and were directionally accurate 84% of the time. Our within position rankings were tied or better than ESPN’s for 12 out of the top 15 players last year with at least two years of experience. Russell Wilson was our #1 QB going into the season and he ended as the #1 last year (he was #5 in Expert Consensus Rankings). So while it’s hard to predict in the NFL, we do now have a track record.
Our models comprise of 140 different machine learning algorithms that learn from data across a wide range of sources. Every year, we make predictions for: Points per Opportunity, Opportunities per Game, and Games Played. From these, other metrics like the ones we’ll discuss below can be derived. That said, our dataset does not have all the data. That’s why we think combining our models’ outputs with human opinions will lead to better results than either by itself. Hence, a few (not many) of the projections below have been hand adjusted.
We will show you:
- Weekly Value: This compares individual Points per Game projections to the average Pts/Gm projection of the top 20 projected players at a given position. It can be interpreted as average points won/lost versus position (against the top 20 players).
- Total Points: This is simply our Points Per Game projection multiplied by our Games Played projection. Last year, we predicted 27% of the variance in games played (not bad), and our total points projection beat ESPN’s pretty handily.
So without further ado, let’s get to it…
Fantasy Outliers’ Weekly Value Projections for Quarterbacks
This Average Weekly Value chart shows how many points our models think a player is going to win/lose your team against the top 20 starters at quarterback. If you’re in a 12-team/1QB league, the numbers will be a little lower, but the relative rankings should be similar.

- No surprise on Aaron Rodgers. We can all agree on that (at least, us and Expert Consensus Rankings/ECR). But he’s ‘almost’ twice as valuable as the next guy, in Matt Ryan.
- Russell Wilson (ECR #2) doesn’t even show up on here, as our models are very low on him this year.
Fantasy Outliers’ Quarterback Total Points Projections
These projections are Points/Game * Games Played, so they take into account injury risk. While our models are high on Aaron Rodgers when he’s healthy, he goes from #1 to #6 here (and we had to adjust it to get it that high) since they see some increased injury risk for him.

- Dak Prescott and Marcus Mariota at the top? If that turns out to be true, those could be two great value finds.
Finding Quarterback Draft Value versus Expert Consensus
Here is a little graph I like. Top 20 ECR rankings on the y-axis and corresponding Fantasy Outliers Weekly Value rankings on the x-axis. Anyone in the top left of the graph is a potential value find. Green means they have a projected positive value versus top 20 players at their position.

- Andy Dalton doesn’t even show up on this graph since he is out of the Top 15 in ECR (#21). Our models have him at 11th Pts/Gm and 14th Total Points. On top of that, as a Bengals fan, I see some solid offseason moves, better offensive line, and better receiving core. I don’t like throwing random stats around, but he did have the highest passer rating in the preseason. He could be a sleeper.
- Otherwise, I like all of these guys as value finds — Matt Ryan, Dak Prescott, Marcus Mariota. Maybe Carson Wentz or Andrew Luck if they slide to you. Phillip Rivers and Blake Bortles are roughly replacement level guys you can probably get towards the end of your draft.
Top 30 Quarterbacks Points per Game Projections
This is a screenshot from our interactive tool, built by Ray Harris.

For the full experience, our 2018 Fantasy Football Season Projections are now available for free in this Glorious Google Sheet.

To stay in touch, please, join us by following us on Twitter (@fantasyoutliers) or subscribing to our weekly newsletter. We’re a small team, so let’s grow together!
Model performance: How we used machine learning to beat ESPN’s yearly and weekly projections in the 2017 fantasy football season
- Can machine learning help improve your fantasy football draft — Comparison of Fantasy Outliers’ yearly models’ 2017 fantasy football draft performance versus ESPN and Expert Consensus Rankings
- How Artificial Intelligence (AI) beat ESPN in Fantasy Football — Summary of results of Fantasy Outliers’ weekly predictive models vs. ESPN during Weeks 6–16 of the 2017 NFL regular season
- Hey everyone, we beat ESPN — Part 1. Background and Methodology — Comparison of fantasy football weekly model performance of Fantasy Outliers vs. ESPN during Weeks 6–16 of the 2017 NFL season for Standard and PPR scoring formats
- Hey everyone, we beat ESPN — Part 2. Results — Comparison of fantasy football weekly model performance of Fantasy Outliers vs. ESPN during Weeks 6–16 of the 2017 NFL season for Standard and PPR scoring formats

