June 7, 2026

TL;DR— The below explains all the reasons why I designed a new option for you to re-order player rankings, when you customize your cheat sheets [<- link to the free tool right here].  I explain how the method solves a long-standing problem, while it preserves the useful ability to measure player "draft value".  But to get it working properly on the site, you need to order the players as if in 0.5ppr, to get the right output generated, when it converts valuations into your specific scoring settings.

Introduction

Whether auction draft or snake draft, there’s long been a detail issue with customizable draft sheets (including the old BeerSheets): Correct player ordering, within the correct value scaling.

Let me explain why this has been an issue in the past.

Why Valuation is useful

First, a simple re-cap of the context: Remember that the key benefit of Value-Based Draft sheets is that you obtain reasonable numerical "values" for very player, tailored to your settings.  Obviously, many of us care about this, because making draft decisions is easier if you can assess player “worth”. A key example is how useful it is to get auction values in dollar terms ($100, $85, $79, etc.). In contrast-- even in snake drafts-- it’s more difficult to make your pick decisions if you only have plain old ordinal ranking (who’s #1, #2, #3…).

The parts we can trust

Some steps of the player value process are easy to trust -- and some steps are not. If you care about achieving reliable calculations of player “worth”, let's start with the trustworthy steps, before the problematic ones:

• Reliable process: calculating player “$Values” from expected fantasy point inputs.  This conversion is based on rational theory.  The calculations process those points according to your league settings.

• Also reliable: estimating the fantasy points themselves for each player.  fantasy points always take this input data: #yards (receiving, rushing, or passing), expected #TDs, #receptions, #interceptions, etc. To get forecasted points, you simply use forecasted data = the “projections”.

However… the “Projections” themselves are less reliable. 

Then-- as been discussed on this subreddit in years past-- things start to look fishier when you consider the range of available projection sources.
There’s a much small number of experts who contribute reliable “projections”.  

See, not every “expert” publishes projection details (like expected #TDs, #receptions).  Most of them only reveal their final player “rankings” (#1, #2, #3…).  Regardless of whether experts start their process by first generating projections, very few good projections get publicized.

Maybe it shouldn't be a big deal. Maybe a small sample size would be okay, if the quality was great. However, it feels like an evident problem when the calculations end up producing a different player order, compared to the rankings we expect. 

For one simple example, from 2025: projection sources averaged to give Kittle more fantasy points McBride.  That was confusing, since most of us considered McBride the clearer #2 pick, just like the ECR.  As a side note, I dug into it: they were projecting Kittle for more TDs, so a higher ratio of TDs to yards.

Solving the problem

I wanted to address these discrepancies in rankings vs. projection. When I accepted the challenge (in this linked reddit post) of creating a draft tool on my site-- to try and fill a void that people felt-- I researched all the past Reddit criticisms (from earlier years of BeerSheets, now only visible in WayBackMachine). I found many comment threads that attacked the problem of trusting projections.  Commenters considered projection-makers to be imprecise-- warning that player valuations couldn't be taken too seriously. 

This is why I designed my draft tools to automatically modify the underlying projections.   Doing so brings the projections into alignment with rankings-- and a lot of us put more trust in aggregate rankings (or specific expert rankings). 
Maybe it helps if I’m explicit about the assumptions I'm applying, in this set-up:

• Projection makers are trusted to provide the relative ratios of expected player stats: TDs : yards : receptions, etc.  (Some players have double the TD:rec ratio, compared to others)

• Projection makers are also trusted to provide the magnitude of the general curve of fantasy points vs. rank#. (So the relative value of WR#2 compared to WR #10.)

• But Projection makers are not trusted to provide the precise positional player ordering.

  • Instead, rankings are used to re-calculate the player’s fantasy output, which is then used to re-scale that player’s projections (but with the same ratios of stats).

How to Use the New "My Rankings" Feature

To use the tool practically, you have 2 options when grabbing your Hold-My-Beer-Sheets:

1. By default, the sheets will use projections that the tool aligns with ECR (0.5ppr).

2. And, New this year: You can modify your own rankings (0.5ppr) to generate custom, aligned projections!

By clicking “My Rankings”, you can drag-and-drop the players.  So, if you have your favorite Fantasy Expert, you can modify all the players into their positional order.

All you need to know is this:

• THE ORDERING THAT YOU INPUT MUST MATCH TO RANKINGS IN TYPICAL 0.5 PPR, EVEN IF THAT’S NOT YOUR LEAGUE SETTING.  This is important, because the underlying fantasy point scale is built to calibrate against 0.5ppr.  Then, the tool is built to convert the projections to your own scoring settings—which can very well result in a different player ordering.

Summary

I think you can have more faith in Draft Cheat Sheet valuations that match the player rankings you expect (whether ECR or your own).  This was something I believed was critically important, to make more sensible draft sheets. 
So now, Hold-My-Beer-Sheets (the free 1-pagers) and TapThatDraft (the web-based tool) always carry out 2 combined customizations: (1) valuation according to specific league settings, (2) valuation according to the player ordering that you trust.

Hope you enjoy!
And as always let me know if you have questions.

/Subvertadown