PFP Lab Notes

PFP Lab Notes

Archives
Log in
Subscribe
November 15, 2026

PFP Lab Notes #6: units fell, the close didn't

Last week three of our rules were over +10 units, and we said that wasn't proof yet.

This week showed why.

Units moved a lot. N2, our NHL overs rule, went from +16.6 to +7.1. S2, our soccer steam rule, from +12.9 to +7.9. B4, our NBA unders rule, from +11.9 to +10.1.

The close barely cared. N2 still beats the closing line on 91 of 121 picks where the price moved. S2 on 113 of 185. B4 on 88 of 148.

A bad week of results takes units. It doesn't take a good price. That's why the close is the scoreboard we trust, and why none of these rules is proven yet. Beating the close isn't enough on its own either. It has to beat it by enough to cover the book's cut.

Joined the board this week

  • Kairos Edge, NHL and NBA builder
  • +CLV, NHL player props (shots and saves)
  • Castri2k, NBA player props

Every pick they post from their start date on is graded against the close, in public.

Also new

  • Name your book, get graded at that book. If a pick says FanDuel, we check FanDuel's close. No book named means DraftKings.
  • Want your picks graded? It's free: pixelforgeplays.com/builders

Next week: one of our posts went viral. Did the market care? We wrote the test down before we looked.

Every pick stays up, losses included: the Board

Casey

Don't miss what's next. Subscribe to PFP Lab Notes:
← Newer What I learned after grading other people's picks Older → PFP Lab Notes #5: three rules over +10 units. Not proof yet.
Twitter
pixelforgeplays.com
Powered by Buttondown, the easiest way to start and grow your newsletter.