Fantasy Football Articles & Model Guides

I've been excited to share with you this updated new study. It's about your relative payoff from drafting each different fantasy position. It's about the fantasy "return on investment", so to speak: Which position (QB/RB

I got a lot of requests to analyze kicker streaming. So I crunched a whole bunch new simulation programs, to try and find out whether it's better to hold or stream kickers. There were some mixed inputs from discussions h

In 2022, I started using my weekly game score forecasting models for a new purpose: generating advice for Survivor Pools. What’s different about the Subvertadown Strategy? Main features: The main input is from my own for

As covered in my Survivor strategy article, my algorithms seek to optimize the expectancy value of “#weeks survival”— in other words, the Expected Longevity. The meaning of that figure is easy to explain: it is the avera

Here’s a mathy post to improve your understanding of the game, with a stats breakdown that I had searched for but never found. The goal is to give some perspective on how predictable / unpredictable fantasy is, in order

Fantasy point distributions One way to visualize a position’s impact on your roster score is to look at the distributions of scores. (The below curves have been smoothed within 1 point.) It is also interesting to see the

Before this study, I was not completely convinced QB streaming was usually a good idea. Below I discuss some of the arguments and counter-arguments for streaming (vs. drafting+holding). TL;DR Of course I hope some of you

This post is about exploring how well projection models might improve the points we get from the QB position. Plus some commentary exploring how predictable/unpredictable QBs are. Some key questions were: What's differen

In extension of my "QB-Kicker Pairing" analysis, here is a look at all the fantasy positions. Why is this interesting? Especially for critical matchups, people often ask about the effects of using players from the same t

While making all my kicker model improvements, I have had my eye on this particular accuracy number. A correlation coefficient of 0.3 is sometimes considered an unofficial "cut-off" for what is reliably predictable. You'

Here is an illustration of how random fantasy football is. It is a high level view of the predictability of different football-related and fantasy-related variables. Kind of a summary of my research, with a good deal of

What factors of kicker performance are predictable-- and what is not? Fantasy kicker scoring is most of all linked to the production of field goals. I have made an analysis of different components, and how strongly they

The idea of focusing on this specific position pairing (QB+Kicker) was partly based on columns I had read years back (one example is here). For some time, I've been wanting to investigate this methodically, and I hope yo

Every year I see different people attempting to make a statistical model of some kind. In 2018 there were 8 of us providing D/ST rankings... and nearly all of us had a different approach. What I'm writing here is a 2- or

Randomness of Spreads: expected values vs. outcomes One of the "a-ha moments" for me was realizing how much error betting lines have, relative to results. You should pretty much expect Spreads to be wrong by 10 points ab

Randomness of O/U: Expectations vs. outcomes of game Total Totals are more random (harder to predict) than spreads: Interesting side note, that the O/U error is not centered at the mode, and there is a longer tail of ove

This is a reference for those of you who are: Wondering how much trust to put in accuracy grades (either from FantasyPros or from here on Reddit) Trying to make your own accuracy assessment of different ranking sources O

The point of this post is that I've put a lot of thought into making stats more understandable for fantasy. The focus here is correlation coefficients (CCs), which I often use to describe both positional predictability a

Explanation of Overfit (Cause of recent model updates) Some of you requested me to describe what is "overfit". So here's ...really probably... more than you really wanted to know! Anyway, I hope like 3 of you out in the

One of the deviously fun parts of developing predictive models is the feeling of defying what they said about stats in school: Correlation is not causation. With predictive modeling, we turn the idea on its head and bend