1103: Market Score

**JJ Zachariason** (0:02)
This is The Late-Round Podcast with your host, JJ Zachariason.
What's up, everyone? It's JJ Zachariason. In this episode 1103 of The Late-Round Fantasy Football Podcast, thanks for tuning in.
Many of you are familiar with the Zat model. There's one at Running Back, Wide Receiver, and Tight End, and the model aims to project a player's fantasy output across his first three seasons in the NFL. That's been something that you've read and listened about for years now. And many of you know that draft capital is sort of the foundation of those models. Everything revolves around where a player gets drafted, and it pivots off of that based on various inputs, age-adjusted production, size, teammates' scores, program strength, and so on. A typical wide receiver drafted 50th overall might be expected to have a zap score of, say, 65
But if he has really good inputs, that number can be a lot higher. If the inputs are bad, then it can be lower. Those models have served me well through the years. They're helpful tools during your rookie drafts to help you make better decisions. But for a while now, I've been wondering why that logic stops with the NFL draft. Why create a model like that only for prospecting? Why not do something similar for our season-long leagues? That's what myself and Brandon Gidula have been working on this offseason. And for the first time, I'm talking about it with all of you. Introducing Market Score.
Now, before I dive deeper into this, I did want to remind you to check out the brand new lateround.com. You can become a member, become a subscriber straight on the website now, not Patreon. You can pre-order the 2026 Draft Guide if you'd like, or you can just look at how pretty this site is. Regardless of what you do, at least check it out, lateround.com.
Market score is a metric that lets us see if a player is overvalued or undervalued by the market. It looks at a player's average draft position, it factors in some previous season numbers, and it says, this player is expected to play above or below expected this year. Let me explain this a little bit further. I talk about average draft position expectation all the time on the show and in the draft guide. Essentially, when you do a little bit of regression analysis, you can find the number of points per game we'd expect a player to score at a particular ADP. A wide receiver selected 50th overall based on history might be expected to score something like 13 PPR points per game. That's his expected points per game. Now, obviously, if he was drafted earlier, that expectation would rise. If he's later, it would fall. Market score takes that number, it factors in some previous season analytics, and it tells us how many points above or below that number, we should expect the player to score. All based on his ADP.
So just like the ZAP model, this is centered around market value. But instead of draft capital, its average draft position. The obvious problem that pops up is that ADP is dynamic. It changes and it fluctuates. Draft capital, once it's set from the NFL draft, we're good to go.
And that's definitely going to be something I've got to talk about when I reference market score within my content. The formula might like a player in May or June, but if his cost really changes through the summer, then all of a sudden we could be looking at the player totally differently, at least through the lens of market score.
Now look, me talking about this, I'm sure you're curious about what market score says about players here in 2026 But before we get into that, I want to share sort of the historical numbers with this metric to really wrap your head around what this is doing.
Average draft position is a pretty good indicator of how good or bad a player is going to be. But just like draft capital, it is far from flawless. And just like draft capital, there are parts of the draft, mostly in the middle and late rounds, where average draft position becomes far less predictive. For example, middle round tight ends, something I've talked about for years now. They really just suck. As in, we're terrible at predicting which ones are going to be good. Tight ends pick 60 first, so just after round 5 through pick 120, or through round 10 of a 12-team league, there's zero correlation between where the tight end was drafted within that range and how many points per game that tight end scored. The R-squared is 0.7 percent, meaning ADP explains virtually none of the variation in fantasy scoring for tight ends drafted in that range. It's irrelevant. But with market score, we get that R-squared to about 14 percent, which isn't like going to solve all of our problems, but it directionally gets us into the right place.