The Quarterback Streaming Challenge: Week 5

A look in at the current state of the Quarterback Streaming Challenge.

Welcome to the fourth installment of the Quarterback Streaming Challenge. As a quick recap: previous research of mine has found that quarterbacks average fewer points per game in weeks they were started than they do in weeks they were left on the bench. This finding challenges one of the underlying principles of quarterback streaming, which states that by assiduously selecting favorable matchups, one could get production from a player that exceeds his usual total.

Faced with this disconnect, I've designed a controlled experiment to test the hypothesis. I have invited 21 participants to try their hand at streaming quarterbacks based on a real fantasy football industry experts league. These participants are largely writers and contributors at various fantasy football sites— seven here at Footballguys, nine from elsewhere around the web— with five additional at-large spots awarded to fans.

This is primarily an attempt to measure the performance of the Late Round Quarterback strategy, as well as our ability to predict matchups without the benefit of hindsight. With that said, I'll be publishing a recap of how things stand every week so that it can serve as a roadmap for those of you who find yourself streaming in your own fantasy leagues.


I want to talk briefly about the concept of retrodiction. If you're unfamiliar with the word, (as my spell-checker apparently still is), its meaning is indicated by its roots, a mashup of retroactive and prediction. Retrodiction is the act of making "predictions" about the past.

While you may not have heard the word itself, you're undoubtedly familiar with some examples of retrodiction. Here's one you might have seen making the rounds last year: if you had started the quarterback who faced the New Orleans Saints every week, you would have finished the season with 4544 yards passing, 45 passing touchdowns, and just 9 interceptions at the position. Scoring systems can vary, but in most that would have been good for a top-3 fantasy finish.

(The problem, of course, is that starting the QB who faced the Saints every week would require owning Cam Newton and Carson Palmer, in which case you could have simply started them every week, instead.)

This retrodiction is offered as proof of the power of matchups. But the issue— and what makes it a retrodiction instead of a prediction— is that it was said after the results were in rather than before the season. If someone had offered "start quarterbacks against the Saints" as their strategy, that would have been a good one. But instead, about halfway through the year a lot of people took note of how bad the Saints' defense was and figured what they could have scored had they started quarterbacks against them.

A lot of fantasy analysis is retrodictive, "if I knew then what I know now" style analysis. Which is one of the reasons I wanted to structure this experiment the way that I did. If strategies such as "start quarterbacks against the Saints" are amenable to prediction as easily as to retrodiction, they will show up in this data. If they're not, they won't. Either way, we'll have a truer picture of what a Late Round Quarterback or quarterback streaming strategy actually looks like in practice.


On the theme of retrodiction, I would like to make two observations about the data so far. The first is the idea that good quarterbacks are always available late. This is entirely true; any of our participants could have selected Matt Ryan as one of their quarterbacks at the start of this experiment. (There's that retrodiction again.) It's also true that only five participants actually did, which is half as many as took Flacco (10) or Griffin (9), as many as took Marcus Mariota (5), and just one more than took Alex Smith (4).

The second is the idea that matchups are predictable. Carolina's defense is obviously a good matchup for a quarterback. There were plenty of signs going into last week that they were in trouble, and the fact that they just cut one of their top cornerbacks only seals it. It's possible that "you should have started your quarterbacks against Carolina" becomes one of the great retrodictions of 2016.

But of the five participants who rostered Matt Ryan, four benched him last week. The fifth waited to see if Cutler was going to play first before declaring Ryan his starter. What's obvious with hindsight isn't necessarily obvious with foresight, which is why getting good controlled data like this is so important when measuring a strategy's success.

Speaking of the strategy's success, at the moment our 21 participants are averaging 18.74 points per game, down just two hundredths of a point from last week. The twelve "normal" teams in the league we're tracking had a much worse week; their current 20.78 point per game average is a half point worse than it was last week, and as a result, our participants have narrowed the deficit to roughly 2 points per game, on aggregate.

Now, here are the hot additions for the week:

And here are the declared starters:

One last note of interest: three of the five Matt Ryan squads are benching him again this week, albeit against a much more fearsome defense to this point in the season.


More articles from Adam Harstad

See all

More articles on: QB

See all

More articles on: Strategy

See all