Welcome to the 13th installment of the “The Wrong Read,” an article series that reflects on recent podcast episodes and digs deeper into the aquifers that the podcast hosts and guests discover in attempt to provide a constant stream of fresh perspectives.1 If you’re worried about your Zero RB squad, check out this piece I wrote last week on some NFL trends in 2017. I don’t know if it will allay your worries, but I think it’s an interesting read.
With eleven weeks already in the books, you might already know whether your fantasy teams are going to make the playoffs, or whether you can start looking toward next year. So this might be a good time to start evaluating your 2017 teams. Peter Overzet appeared on both the RotoViz Highlight Reel and RotoViz Radio to do just that. For example, he mentioned on both shows and in his article recapping the season that he and his FFPC team co-manager and RotoViz Radio co-host Pat Kerrane would probably not be picking a tight end early in that format, even despite the TE-premium scoring.
It’s not just that their early TE did not return value, but they also learned that they were able to get serviceable weeks from waiver-wire TEs. Rather than making the top TEs more valuable, the TE-premium scoring makes all TEs more valuable, thus producing more stream-viable TEs. In a way, that is, it actually decreases the value of the top TEs. This is the sort of discovery that is often made by actually going through a season and evaluating things at the end, as Peter and Pat did.
Processes vs. Outcomes
Learning from your mistakes is perhaps one of the most important things you can do as a fantasy owner. It’s of course also important to learn from your successes. The problem is, how do you tell which are your successes and which are your mistakes? This is where the discussion of process versus outcomes enters. The thinking goes that because much of football, and therefore fantasy football, is governed by randomness, even the perfect process will not yield perfect results all the time.
For instance, there were a lot of reasons to think David Johnson was the best player going into the season and deserved to be taken first overall. But anyone who instead took Le’Veon Bell or Antonio Brown with the first pick is probably much happier. Does this mean taking Johnson first overall was bad process? Most people would still say it was a smart pick—that it was an unlucky bad outcome in spite of good process. Likewise, if you took Adam Thielen in, say, the fourth round, you’re probably pretty happy right now. But considering you could have gotten him several rounds later, it might not have been the best pick. This is, arguably, a good outcome despite bad process. Therefore, it is important when evaluating our 2017 season that we evaluate our processes apart from our outcomes.
Can Processes Be Separated from Outcomes?
There is an insight here well worth appreciating, but I have a problem with this line of thinking: I believe it presents a false dichotomy. Research in other fields indicates that evaluations based purely on outcomes play an important role, and that outcome-based decisions are often both easier to make and more satisfying.2
But even beyond the potential usefulness of outcome-based evaluations, a deeper question lingers: how do we recognize good or bad process apart from outcomes? Peter and Pat recognized that a TE-early draft strategy might not be optimal in the FFPC because their early TE, Jordan Reed, underperformed. Their process may indeed be flawed, but they only would have recognized this because of a poor outcome. If Reed had played most of the season and finished as the TE1 (a distinctly possible outcome), there would likely be no occasion to take a second look at their TE-drafting strategy.
And in the case of Thielen, are we really in a position to say that spending a fourth-round pick on him was bad process? It was a process that led to the desired outcome after all. And every year there are players who we think are grossly mispriced according to ADP (in fact, Thielen was one of those players in 2017). If you were so high on Thielen that you didn’t want to risk letting another team get him, and there was no one you liked more when your fourth-round pick came up, then you were absolutely right.
Process and outcome are thus tangled in such a way that if it’s true that good process sometimes produces bad outcomes and bad process sometimes produce good outcomes—and I think it is true—it may be impossible for us to really know which are the good processes and which are the bad processes. Not all good processes lead to good outcomes and not all bad processes lead to bad outcomes. So even looking at outcomes does not help us to distinguish between good and bad processes. Evaluating outcomes may be our best bet.
Process, Outcomes, and Symmetry
So was taking Johnson with the first overall pick bad process, or just bad luck? This question actually brings up a further problem with the process over outcomes attitude: it violates the principle of symmetry.3 That is to say, it tends to treat negative outcomes as unlucky, but positive outcomes as good process, rather than using either luck or process to explain both. To be sure, this isn’t universally or consciously the case, but the ease with which we reevaluate our process on the heels of negative outcomes almost ensures that it is both mostly and tacitly the case.
In principle, the notion of evaluating process rather than outcomes affirms that luck is part of the explanation for both good and bad outcomes. But in practice, we tend to recognize bad luck far more often. What this means, of course, is that we tend to assume we have good processes.
In light of this tendency, and considering how process and outcome are inherently tangled, I’m proposing a new way to go about evaluating your 2017 season: assume you always have bad processes. If you end up with a bad outcome, that should be expected. But if you end up with a good outcome, that’s just good luck. Is it true that you always have bad processes? Probably not. But if you assume your processes are bad, then you will always be looking to improve them, even in cases where they produce good outcomes.
If you picked Bell or Brown instead of Johnson, you got lucky, and you should be looking for ways to improve your process, since luck is unsustainable. If, on the other hand, you picked Johnson, you messed up, and you need to figure out where you went wrong. Obviously I’m being somewhat facetious, but the point—which I think everyone would agree with—is that you should always be looking to improve your processes, whether they produce good or bad outcomes, even if you don’t think they need improvement.
- My metaphors started getting odd a while ago—now I’m leaning into it. (back)
- Whether ease of decision making and satisfaction with our decisions are in fact things we ought to be aiming for in fantasy football is a topic for another day. (back)
- I’m borrowing this concept from certain sociologists of science—their basic premise in advancing the principle of symmetry is to say that both true and false scientific beliefs should be subject to sociological explanations. (back)