Stealing Signals, Week 9, Part 1

Plus why it's OK to cut good players, and more roster management thoughts

One of the biggest things I’ve learned over the years, that I still struggle with, is you have to be OK cutting players. Particularly, you have to be OK cutting players even if their value is not at zero, and even while accepting that some percentage of the time they will have a good year anyway.

One of the biggest things that I’ve taken from fantasy football, and from the sport of football in general, is not being afraid to be aggressive. And I think it’s most damaging to success in fantasy when we talk about in-season roster management, specifically in terms of trading away or cutting players.

For so many new-ish players, you’ll see them in one of two camps: 1) Too afraid to cut any player or do anything with their roster (a camp some people stay in forever); 2) Far too loose. For 2), the inevitability seems to be getting bit and then straying into camp 1). I do think over time that people tend to improve, but the endowment effect is real. For those who don’t know, it’s one of the many cognitive biases that messes with your ability to win in this hobby, and this one in particular refers to how people are “more likely to retain an object they own than acquire that same object when they do not own it.” Players aren’t objects, but the fantasy application is still clear that we overvalue the guys on our teams.

This tightness with roster management is a massive issue. You’ll never realize it, and you’ll always put far more weight on the add, but good cuts can be the best moves you can make in a season. The roster flexibility of freeing up a slot to churn for a block of the season is what leads to player hits.

What you will remember are the misses. Last year, I had FFPC teams where I cut Trey McBride. It doesn’t matter that through Week 7, he hadn’t played 60% of the snaps and was averaging 2.1 receptions and 24 yards per game. I remember that shit.

This year, in my league of record, I drafted Tyrone Tracy and was churning a different roster spot for Justin Fields to cover QB since I went single-QB with Jordan Love who got hurt in Week 1, so I didn’t have another cut. I put in an enforcement bid on Isaiah Likely for $123 and beat out bids of $115, $103, and $75, losing Tracy in the process. I unceremoniously cut Likely after Week 4, when it became clear he wasn’t truly breaking out.

I share these examples to say very clearly that these are not mistakes. If you believe that to be true, I challenge you to question whether it’s the Endowment Effect impacting your thought process.

We don’t look back on draft day and seriously consider every double-digit-round pick that we didn’t make as some type of mistake. Of course players have ranges of outcomes and paths to usefulness. But there’s a reason Tracy was going later than most, and I drafted him in Round 14 of a 16-round draft where my last pick was a kicker and my Round 15 pick (Luke Musgrave) was the other Week 1 cut to grab a necessary QB2. In other words, something in Week 1 would have had to convince me to cut some player I drafted over Tracy, or else I would have had to say I was unwilling to make a bet on Likely.

In this league, my TE happens to be Dalton Kincaid, so if Likely would have built off his 12-9-111-1 Week 1 line and gone on to be a top-five TE like was pretty clearly feasible when we only had that one data point, I’d probably be starting him. It’s why I drafted Musgrave; I felt like an upside TE2 behind a situation-based bet with questionable talent-based breakout upside in Kincaid was a reasonable use of a roster spot, because my team didn’t have a clear path to differentiating TE play even as I used a reasonably high pick on Kincaid.

That seems like a tangent, but it’s central to the point I want to make today. When you’re balancing your bench, and you’re considering cutting players who do have value — and do have paths to upside — the biggest thing you need to be honest with yourself about is your team’s needs. In the example I’m giving, I absolutely had to devote a second roster spot to QB2. All of my other picks from Round 10 to Round 13 in this league were upside RBs like Tracy, and I’m still rostering Bucky Irving, Blake Corum, and Kimani Vidal. The decision I had to make when I needed to carry a second QB was whether a TE2 was worth carrying, or I wanted to continue rostering five of those types of RBs. This is why I ultimately split the difference with my bid and said I’d do it if the price was right, but I don’t think winning Likely for far cheaper than I saw him go almost anywhere else was a mistake, and I don’t think Tracy being the play of those five RBs whose value improved the most in the time since was reflected in his value at that time.

I still liked him, which is why I drafted him in the first place. But a big issue for him was the Giants got beat 28-6 at home in Week 1, in addition to him having 3 touches for 7 yards. Anyway, this is a tangent, because now I’m trying to justify (to myself) that he was the RB of the five I was rostering that I could justify cutting. That would be part of this I could regret, if I wanted to torture myself — everything about roster balance and positional allocation is truly believe in, but maybe I should have just cut another RB right away. (Tracy actually lingered unadded in this league for a couple weeks, and that’s truly where I regret not getting him back as I churned. But that’s what’s so funny about the Endowment Effect — as soon as he was no longer on my roster, he wasn’t a priority to me, and I was instead churning roster spots with the flavor of the week, like Jauan Jennings and Tucker Kraft.)

Setting aside this specific Tracy example and getting back to the broader point, when I said we don’t look back on every double-digit-round pick we didn’t make as some big mistake, that’s not because we didn’t believe that player couldn’t be good. We just had to make a decision when we were on the clock, and we can’t expect perfect accuracy. I think when you frame it as a draft recap, everyone does pretty well in understanding the lesson — yes, hindsight is 20/20, and we all wish we had the perfect picks in every round, but that’s just not practical.

But when we look back at an add/drop, it feels more personal, because of the Endowment Effect, and because we already had the player on our roster. But the fact of the matter is every add/drop should be judged the same way a last-round draft pick should be judged, which is that we have a lot of uncertainty about multiple profiles and we’re picking the one we like the best. At the time I dropped Tracy and added Likely, for example, the RB had 7 yards and the TE had 111 and a TD, and if I was drafting for rest of season, Likely would have been a no-brainer. The FFPC high stakes stuff runs drafts after those opening games and Likely was going in the mid-to-high single-digit rounds. Tracy was going undrafted by the end of draft season in some cases because of that huge injury scare he had in camp where it sounded like he was definitely done for the year. It wasn’t even 100% clear how healthy he was entering the year, and after Week 1 the feeling was that we had been lied to about every August injury (read: Christian McCaffrey, Puka Nacua, et al.).

The point of this lesson is that at any given moment in time, you have to be willing to optimize your roster for a blend of short-term and long-term potential. One mistake I do see — and have made — is over-emphasizing future upside with too many of those RB types I was just discussing, where I’m not optimizing for short-term points enough. As the season goes along, that increasingly becomes an issue, and we have to increasingly get OK with just cutting bait on dudes we don’t want to cut.

At the same time, in cases where we do have the short-term scoring figured out, we also have to be real with ourselves about what the waiver wire options add to our rosters, and whether we can just find that later if we need it. The last paragraph might argue for just cutting bait on Vidal and Corum at this point in the league I’m talking about, but I haven’t done that yet because I haven’t actually needed those roster spots since the starting lineup is fine, and those still are the best use of the back-end roster spots when you’re in that situation, because RBs like that — like Tracy — are clearly the type of player who can gain the most value in a single week at this point of the season. I did add Jalen Coker to this team, and I was excited that he might be the rare WR success story in a similar mold, and instead the Panthers didn’t prioritize him with a strong route share, which was a reminder that he’s a UDFA and it’s just a difficult path.

So the point is there’s a balance, which is why I don’t always tackle this topic, because it’s not necessarily easy to parse. There are a ton of if/thens, and it’s all tacit knowledge. But at the start, the biggest and clearest note I can offer is you do have to be willing to let players go, and just because you do does not mean you don’t think they can succeed. And you shouldn’t regret that if they do. It should be freeing, frankly. On one hand, you understood a good bet in the first place. On the other, you’re going to have misses, because this is all probability-based, and misses should allow you to fire away more. That fear of failure is always worst before you’ve actually failed. It’s why the least accomplished of us seem to have the least shame as they spout the worst opinions everywhere.

But all of this is an especially necessary lesson in a 2024 season where we’re all just trying to get by a little bit. A big part of this intro was catalyzed by me seeing a season-long FFPC roster that has Nico Collins and D.K. Metcalf on the shelf, and is rostering nine WRs with multiple rookies like Adonai Mitchell, Jermaine Burton, Troy Franklin, and Coker. It’s only rostering five RBs, despite not having clear answers at RB.

That’s a mistake. That’s the point I want to drive home here is the answers to what you should be doing with your bench are again this balance of short-term and long-term potential.