Mailbag: roster composition questions
Plus an answer to the audio question
Wanted to pop in for a quick Thursday post to share some fun discussions from the comment section on this week’s posts. But I also wanted to share there’s already an answer to the Stealing Signals audio dilemma!
Substack has an app that is new this year, called Substack Reader, and it has a text-to-speech option that is apparently pretty great. Several of you pointed this out to me and many noted the quality was very good. Some said they’ve been using it for some time, through the August writeups. So if you’re one of those people who consumes better through audio, this seems like your option. I tried the first few words of Part 1 this week, and it did seem pretty solid.
My understanding is this is only an iOS app available for iPhone and iPad right now. I know that’s not going to hit everyone and that kind of sucks. I also haven’t come up with a great universal solution and since this seems to work so well for some people, I’m planning to let this be the option for now and trust Substack will come out with apps to service other customers soon enough.
As for the questions from earlier this week, wanted to note that just as a reality, there will definitely be weeks where I can’t get to any questions, so apologies in advance for that. But I also had a blast reading all your thoughts and feedback this first week. This tweet is true and accurate, and it was prompted by going through the comments on the first week’s editions of Stealing Signals.
This is such a cool community. It can be tough for me to give quick answers, because I like to really engage when I do sink my teeth into a question. Also, as I’ve said before, if you’re asking a start/sit or some waiver preference questions, it’s probably because there’s a lot of overlap in the ranges of outcomes and it’s close. Sometimes you just have to decide which bet you prefer (or can live with).
Alright, here were a few fun questions from this week. I already re-posted part of my answer to Cordarrelle Patterson in the intro to Stealing Signals Part 2, but if you missed that, I probably should have been a little clearer on my Patterson stance (I’m definitely in on riding with him and seeing where he gets you, then figuring out any necessary replacement down the line; he’s at worst a great start to a Frankenstein RB).
From Don:
“Do you think Curtis Samuel has potential season long upside? He pretty much went undrafted (in my 15 round drafts) and I never heard any offseason discussion on him. Or is that trio likely to be too unpredictable? That's why I didn't draft Scary Terry or Dotson but Samuel did look really good.”
This was a popular question, and I realize waivers have already run, but I still want to clear this up because I didn’t anticipate Washington would be such a popular offense to target this week. I think it’s tough. Terry McLaurin has to get more involved going forward, and Jahan Dotson should only move forward, so I’m unsure Samuel can be quite as involved as he was Week 1.
I’m also skeptical of the pass rate stuff with them, because Carson Wentz had some legit terrible throws as I wrote about in Stealing Signals. He then got hot, and played very well for stretches, and I did a whole thing this week about how I thought Trey Lance was fine relative to expectations, so I’m not trying to seem super biased here. It’s just that we have a large sample with Wentz that suggests he can definitely get hot and be good for stretches, but the mistakes can really creep in. He put together a strong game, and they were behind and threw a lot late (as well as throwing a lot early), and all of that was very encouraging. But they wound up running 70 plays and had a heavy pass lean with a +8.6% PROE and I just think that was a little fluky.
Samuel’s role was indeed very encouraging, and he’s had his moments in his career, but I’ve felt like some of the questions this week have asked if I value him higher than we ever did even during his best stretches while in Carolina. And that’s sort of the point that I’m making — he seems back, and could get used more interestingly even than his best days in Carolina, but he’s never really been a priority WR throughout his career so it would be something of a shock if that came to fruition now. In some cases, though, that’s the best bet you can make on the wire, so in those instances it makes sense. The questions about “season-long upside” are getting ahead of things a bit, in my opinion, but I wouldn’t argue it’s impossible.
From Tristen:
Isn’t Elijah Moore’s usage more than a little disappointing? If we’re calling the Jets passing volume noise, doesn’t that mean we should be concerned with his 7 targets? Is it an overreaction to think we should adjust our priors given Wilson’s strong TPRR numbers and the decentralized passing distribution?
I get this point, but it could be applied to a lot of players — more pressingly for someone like Mike Williams — and my response is mostly that there’s a lot of target variance on the weekly level. Moore’s routes were fine. Seven targets is well within the range of maybe just being a slight down game.
If we see this TPRR continue over 3-4 weeks, that’s when I start to get concerned. Same goes for Williams and some other guys. For now, we rely on the larger sample back to last season and expect some positive bounce back in the short term.
From Eduardo:
In this article and your latest podcast episode you keep being skeptical about Michael Pittman because the Colts played in OT. However, Pittman didn't get a single reception in OT. So is overtime really that much of a detriment to evaluating Pittman? I know it affects his target share. Have you considered looking just at the first four quarters? I wonder whether you'd feel different if the Colts had only played four quarters. I guess it doesn't make sense to hold it against him that his team played into overtime when he didn't record a single reception.
I might be wrong on Pittman, but ignoring part of the sample would also be wrong. If he didn't get a reception for that other quarter that was super important in a close game, that's part of my issue with him. My argument is he's probably just not as good as top-10 level, but is rather slightly outside that level.
If they didn't play OT but he went the first quarter of his next game not recording a reception, that wouldn't be great. It would maybe put him on a path to a poor Week 2. My concern is he doesn't earn volume at the level of the best WRs in the league, and my argument about this game is it seemed that way mostly because they threw so many passes. He did have one OT target, but not catching a pass in overtime doesn't really refute my worries.
But I do want to keep harping that I’m not saying Pittman is bad. I’m saying that he had a very good year last year and played all 17 games and was WR18 in PPR. Because he played all 17 games, when you switch over to points per game for wide receivers who played at least half the season, he was WR26. He went WR9 over the final week of drafting at the FFPC. I just didn’t think his situation improved that much, or that there was an obvious reason for him to take a big next step, so I had him ranked WR17 as a fade. I acknowledged in Stealing Signals and on the pod this week that I’d have him in the WR12-WR14 range now. I still think my position is very reasonable relative to the amount of responses I’m getting that seem to think I’m just super biased against Pittman. I think he was overdrafted, that’s about it.
From Jeff:
Do any of you other readers know of a spot in the past where Ben has talked about how to use his info for making start/sit decisions in a given week? I am "Frankensteining" my RB2 spot so I have lots of similar guys and trying to use the best data to decide each week who to start.
I suspect this to be a very popular question so it was one of the reasons for the mailbag today. I typically rely on good projections, because on a weekly basis, I do trust good projections to account for all the factors. My favorite projector is Michael Leone at ETR, but I understand their subscription is out of some people’s price range.
But on a weekly basis, sharp projections that take into account role, projected game script (using Vegas lines), and other factors like weather can be really helpful. Just be careful with some defaults on sites that don’t have great projections. The key to why weekly projections are better than seasonal is there are a lot more known factors in the short term that can be accounted for, but good projections are only good if they are using all of that information that is available in an appropriate way. That takes considerable time, and there are definitely “projections” around the industry that don’t put in the work and aren’t much better than just going off feel.
From Justin:
If you had Trey Lance as your only QB in a single QB home league, would you grab a backup like Wentz or Mariota, not to start but as insurance? 5-6 bench leagues. Would have to drop Skyy Moore, Allgeier, Eno, McKinnon, R. White. These are combined options over 2 leagues.
The replies to this were so good that I have to give shoutouts.
From N8:
I'm in this situation in two leagues and am holding tight for now. I don't feel like the waiver wire guys are much better than the waiver wire QBs will be next week so I'd rather hold the lottery tickets.
From William:
Depends on how deep QB is but I wouldn't do it yet. You'll be able to find a Wentz/Mariota type of QB on waivers next week if you need to bench Lance but most of those guys are one injury away from becoming very valuable. You'll most likely see a couple more playable QBs hit waivers tomorrow and if you wait you get an extra week of info.
I added that those replies were right on the head, but that insurance is typically not worth the opportunity cost in fantasy. Let it ride and hope Lance has a big game next week. Cross the replacement bridge when you come to it if need be. (That’s specifically relative to the drops Justin would have to make, who are all good. If you had a worse player to drop and no obvious add, I’d get it.)
But the bigger-picture note is get comfortable with uncertainty. It’s a legit edge. Too many people waste resources to feel better about stuff like insurance. In single QB leagues, as long as you’re playing Lance this week, you’re fine without a backup because there’s an outcome where he plays really well and it looks great heading into Week 3. You can do similar at RB sometimes where if you have two really good anchors, you might not need a bunch of bench depth, or at WR where if you have a clear top four or five, you can get thin there.
And especially when it comes to stuff like stashing a DST with a great matchup one week early, I definitely understand the urge, but it better be a situation where the opportunity cost is very low and there’s not really a good alternative as a stash, because we aren’t even that good at predicting DST scoring so you’re holding that extra week for an uncertain outcome. I’m more likely to do this if the DST in question has a stretch of like three straight good matchups because that should smooth out some variance and you are actually gaining an edge over a multi-week stretch by giving up that one week of roster flexibility. But kicker is another one — deal with your kicker situation the week of your kicker’s bye, not before.
Alright, that’s all for today. I’ll have the first edition of “Input Volatility” either tomorrow or on Saturday to help with a few interesting start/sit options.
Thanks for the (second) shoutout for the comment earlier in the week. One more and I'll have to take my buddies advice and add it to my Twitter bio. I really appreciate your insights into the 'mechanical' aspects of fantasy like roster construction in home leagues.
<3