Beating the System: A Quarterback Heuristic
This is the second in a series on DLF looking for heuristics we can use for fantasy football. The first covered IDPs. This week, we’re starting to cover offensive players with quarterbacks first on the list.
To recap, a heuristic is any approach to problem-solving or self-discovery that employs a practical method, not guaranteed to be optimal, perfect, logical, or rational, but instead sufficient for reaching an immediate goal. Where finding an optimal solution is impossible or impractical, heuristic methods can be used to speed up the process of finding a satisfactory solution. Heuristics can be mental shortcuts that ease the cognitive load of making a decision.
In short, they’re useful but not perfect tools that help us be mostly right much faster than it takes to be totally right. This is super useful in fantasy football because every week we’re reacting to another huge amount of data and we need to make our decisions and make trades and waiver claims off the back of the available information.
Identifying a quarterback heuristic
The opportunity with quarterbacks arises because of the different ways they can score points. Passing yards are the backbone of the position but depend a lot on game script. QBs of teams that win easily see less action due to the team running more. Because of the fluctuation, they’re not that useful as a heuristic. Here are Drew Brees’ weekly passing totals from 2018 to illustrate the point:
Similarly, passing touchdowns are very volatile and fluctuate wildly week on week. With such a low base number we simply cannot expect to see enough stability to be that useful. Coupled with the fact that they’re (arguably) overweighted in terms of fantasy points, it’s a bad metric for our use here. To illustrate the instability here are all quarterbacks’ passing TDs by week for 2018:
You don’t need to be able to pick out the individual data points to be able to see how much it fluctuates for players.
Rushing yards are obviously not a candidate for predicting QB scoring reliably but here are the weekly rushing yards for Josh Allen in 2018 anyway. It’s just funny how people think he’s a reliable rusher when it all happened in just four games.
Actually, when you boil it all down, the best correlating stat for QBs over a season to their total fantasy scoring is just combined drop-backs and rushing attempts. This factors in the number of passes they attempt but with sacks taken also counted – and also adjusts for rushing ability.
Here’s how this stat correlates against fantasy scoring for the last decade of data:
The R squared figure of 0.917 indicates a very strong relationship which is logical. The more times a QB throws or runs the ball, the more points he’s likely to score. We’re not looking for the most perfect of all statistical formulae here – we’re looking for a useful heuristic. This chart tells us that the link between the number of dropbacks plus rushes is a really good indicator for how many total points that player will score for a season.
If we look at the same datapoint on a weekly basis for 2018, this is what we see:
After week one, the figure was already at 0.74. That’s good but not great. The reason for this is because of injuries and changes in depth charts. After two weeks, Tyrod Taylor, Joe Flacco, and Sam Bradford were starters whilst Baker Mayfield, Lamar Jackson and Josh Rosen had played a combined 37 snaps.
But by the end of the season, the correlation was far better – as we’d expect. By that time, we knew exactly how much every player had played for the season and what they’d scored and (as per the chart above) we know those are strongly linked.
Realistically after the first four weeks of the season, we can get a really good idea of how QBs will score simply by looking at their dropbacks and rushing attempts.
It’s worth noting that on a limited scale, drop backs plus rushes (we need a snappy acronym for that) is not a strong correlator. This chart shows the correlation in each week of 2018: 
These are all pretty low values (with some absolutely terrible weeks) which shows us that when looking at a limited sample size (32 in each week is small), we cannot really expect drop backs plus rushes to equate to overall scoring – even though we can rely on it over a season’s worth of data.
Conclusion
There’s a fascinatingly counter-intuitive bit of info at the heart of this article. One particular, easily-tracked stat gives us a really good idea of how quarterbacks will end up scoring at the end of the season. It’s not perfect and there will be variance but it’s definitely useful enough for us to base in-season decisions on it. The very simplicity of the single stat means that we can be more agile and faster than some of our league mates which will helps us streamline our decision-making.
This stat should absolutely not replace any of the smart, complex analysis that exists in the fantasy world but it’s another helpful shortcut to making good decisions. In the more complicated leagues that are increasingly common these days, we need to be able to be more agile and decisive than the people we’re competing against. Heuristics offer that benefit and this one is probably useful.
Thanks for reading.
- Ten IDP Fantasy Football Stats You Need to Know after Week 16 - December 29, 2023
- Ten IDP Fantasy Football Stats You Need to Know after Week 15 - December 22, 2023
- Ten IDP Fantasy Football Stats You Need to Know after Week 14 - December 14, 2023
