Streaming QBs in fantasy: How to do it, what to look for

It's easy to say you'll stream QBs. It's a lot harder to do successfully. Scott Barrett brings you all the research on the practice.

| 2 months ago
(Brett Carlsen/Getty Images)

(Brett Carlsen/Getty Images)

Streaming QBs in fantasy: How to do it, what to look for

In college, I would have to write these manifesto-length research papers that would take me days just to get started writing. When a follower on Twitter recently asked me “What should I be looking at when targeting quarterbacks to stream on a weekly basis?”, I became overjoyed with the possibility of writing an article hundreds of words more than what my editor deemed appropriate. (Editor’s note: Thanks a LOT, follower on Twitter.) In part due to this question, and in part inspired by this article by Rotoworld’s Evan Silva, I’ve sought to out to determine just that – and finally crack “the streamer’s dilemma.”

If you’re not familiar, streaming is the strategy of playing different waiver-wire quality players (in this case quarterbacks) each week based on matchups. Ideally, you’re supposed to draft quarterbacks late (while loading up on other positions) and then play different quarterbacks each week while targeting porous defenses. This does not necessarily continue throughout the fantasy season. If at a certain point a quarterback you acquire is on a hot streak or has materialized into an every-week option, then you can continue to start him each week. This is considered an important part of the streaming strategy by its proponents, but today I’m delving primarily into what to look for when playing the more bottom-barrel QBs on a weekly basis.

In order to determine what exactly we should be looking for, I compiled a Karlos Williams-sized excel spreadsheet and an Eddie Lacy-esque amount of data. (Ed.: Oh lord.) Going back three years, I’ve included every QB game where a QB played at least one snap. Going back one year, I have the same data along with opponent DVOA data and Vegas lines for each matchup.

The sum of the parts

What I’ve found is that, while there might not be any one magic formula for streaming QBs, we can compound a number of small positive trends to find QBs likely to outscore their weekly average by a significant amount. While I’m not sure we may be able to match elite QB1 numbers on a consistent basis, this shows how we can maximize the potential of our lesser-tier options. I also believe my findings will also be especially useful in DFS and Superflex/2QB leagues.

To read the entire article, please login or sign up for a PFF Fantasy Subscription

Not a PFF fantasy subscriber? Compare all of our packages and features here.


  • EW Truett

    Is there a link to download the dataset? Thanks!

    • Jenniferehager2

      <<t:i. ★★✫★★✫★★✫★★✫★★✫★★✫★★✫★★✫★★✫★★✫★★✫★★✫★★✫★★✫★★✫★★✫★★✫★★:::::::!!bz665a:….,..

  • evo34

    Your correlations are YTD QB stats vs. current-game QB fant. pts., or are they current-game stats vs. current-game fant. pts.? If the latter: a) the data is not useful at for prediction; b) the reason why att. is no weakly correlated to production is that teams having a bad game will end up throwing more, esp. late game in bad situations. If you know that a team is having a bad game, it’s fair to guess the QB’s efficiency numbers will be bad.