Week 3 Signals from the Noise
Data driven breakouts and declines that impact your dynasty decisions · September 30, 2026
In week 1 I told you that week 3 is where signals start to become trends and if you didn’t have a knee jerk reaction to a week 1 or 2 flop, you’ll start to have better information. Welp…here we are. Week 3 is in the books, so what does the data say? Here is some of my data for you to consider.
First, some context. I run a data driven dynasty site where I create a metric called DPX, which is one metric that combines NFL stats and advanced stats for volume, efficiency and scoring into one signal. I have DPX Base which is last year’s full season production score (2025 in this case) and DPX Now which is this season’s production score (2026, through week 3). Comparing the two now, and as the season progresses, help you not only understand who’s overperforming and underperforming compared to last year, but why their underperforming (volume, efficiency, scoring). All figures through week 3 are pulled Tuesday morning.
After week 3, we can do this analysis safely and while it’s never final or perfect, it certainly provides some information to consider, especially when you’re managing dynasty rosters and you don’t want to make any rash decisions. Let’s start with the good news, here are players who have a negative DPX Base (performed under baseline in 2025) who are breaking out this year:
Isaiah Likely DPX Base: −0.16 ↑ DPX Now: 1.17
This is Likely’s time to shine. Yes, he lost Jaxson Dart, but Jameis Winston throws the rock pretty well, just as often to the other team as his own. Likely will be ok this year. From a dynasty perspective, this is the opportunity we were seeking for Likely as he gets out of Mark Andrews’ shadow in Baltimore. If you have him, keep him. If you can buy low because of the Dart injury, do so.
Matthew Golden DPX Base: −0.38 ↑ DPX Now: 1.70
Golden is making the second-year leap. His rookie year the WR corps in Green Bay was much too crowded, it’s been thinned out now and the main targets now are him and Christian Watson. This is a very strong signal from a player that had a top decile DMX draft score. If you drafted him, congratulations, you likely have a strong fantasy asset going forward.
Jalen Coker DPX Base: −0.07 ↑ DPX Now: 1.09
Coker has been the best WR on Carolina so far this year, but it’s only week 3. Tet McMillan will have his moments, but to get this kind of production from an undrafted player bucks the trends. He’s only 24, Bryce Young is starting to round out his form too, so from a dynasty perspective Carolina may actually be a pretty good offense to buy into for a few years. If he’s still reasonable or someone is selling high, Coker is a worthy target based on our data.
Bhayshul Tuten DPX Base: −0.57 ↑ DPX Now: 0.56
He’s not getting superior volume or scoring right now, but the efficiency is there and he’s the clear leader of this committee. He’s only 24, and his score change from last year to this year isn’t as dramatic as the others above, but I believe it’s a strong enough signal that from a dynasty perspective he’s someone you want to have on your team. It doesn’t hurt that Jax is performing very well these days, and he may have a long runway for success there from a dynasty perspective.
And here are players who had positive DPX Base (performed over baseline in 2025) who have fallen below baseline (Negative DPX Now) in 2026:
Drake Maye DPX Base: 1.16 ↓ DPX Now: −0.23
Wow, didn’t see this one coming. Honestly, in my humble opinion this scoring change between year 2 and year 3 makes Maye a hold rather than a sell in my book. He lost AJ Brown and he has a strong enough year last year to suggest this could be just growing pains. That said, the signal is pretty strong, if you have other QBs to start, I would do so at this point.
Rome Odunze DPX Base: 0.67 ↓ DPX Now: −0.06
Ben Johnson is a serious enough offensive mind to consider keeping Odunze, but he’s yet to breakout and it’s fair to say with so many other talented players on the team, he may never pay back the first round pick you used on him. If you can get something valuable for him, this may be a good time to do it.
Courtland Sutton DPX Base: 0.68 ↓ DPX Now: −0.21
Bo Nix is underperforming (while I’m not highlighting Nix on this list, he’s also gone from positive last year to negative so far this year) and Courtland Sutton is 30 years old. You may well be too late, but if he bounces back a little this season, and you can get something for him or add him to a trade that brings some youth or picks, do it before it’s too late.
Quentin Johnston DPX Base: 0.67 ↓ DPX Now: −0.25
This guy. Between an awful rookie year and a couple of productive scoring years, what do you do with him. Well, this score solidly says sell him if you can. With LAC’s offense being Mike McDaniel’d, it may be too late. But it doesn’t look like Johnston is getting any better than what he is today, a once in a while scoring threat with bad hands.
To finish up, some things to consider. Of the 66 players who qualified to be on these lists after week 2, only 61 qualify after week 3, and a third of the names changed between weeks. That is why you want to wait until at least week 3 to make decisions on these players. The first couple of weeks moves things in either direction, often quite a bit. This is not a comprehensive list, these are the players with some of the biggest variances between these two metrics that would have an impact on dynasty rosters. Bottom line: working with data helps you move away from media driven hot takes to actual performance against a baseline. Measuring all of the statistically significant measures rather than cherry-picking one to make a point, helps you understand the bigger picture. Data is not perfect, and judgement is also required to interpret what these signals mean and how you make decisions for your roster. Happy hunting!