🔍
 
 
Display
Native
Choice Cutout
-
Select Cutout
-
Pork Carcass
-
CH/SE Spread
-
Beef Loads
-
Pork Loads
-
← home
The read, scored

How often is the read right?

Every tool can draw a chart. The question that matters is whether the call holds up. So we backtested the exact CHEAP/EXPENSIVE signal you see on the scanner, walk-forward, no lookahead, across 306 cuts and 82,286 historical windows (2006-04-09 to 2026-03-22). The number that proves the point is shown next to the number you'd get guessing blind.

When MeatRead flags a cut CHEAP · 4 weeks out
62.4%reverted up, vs 52.3% guessing blind · avg move +6.1% vs +1.3% base · n=16,207

4 weeks forward

SignalHitBlindEdgeAvg moven
CHEAP or BELOW NORM
reverts up
62.4%52.3%+10.1 pts+6.1%16,207
CHEAP (strict, ≤ -12%)
reverts up
64.7%52.3%+12.4 pts+8.1%9,539
EXPENSIVE or ABOVE NORM
reverts down
52.4%47.7%+4.7 pts-0.8%44,282
EXPENSIVE (strict, ≥ +12%)
reverts down
54.0%47.7%+6.3 pts-1.2%35,615

8 weeks forward

SignalHitBlindEdgeAvg moven
CHEAP or BELOW NORM
reverts up
65.5%52.9%+12.6 pts+8.9%16,603
CHEAP (strict, ≤ -12%)
reverts up
68.2%52.9%+15.3 pts+11.3%9,820
EXPENSIVE or ABOVE NORM
reverts down
52.7%47.0%+5.7 pts-0.8%44,927
EXPENSIVE (strict, ≥ +12%)
reverts down
54.6%47.0%+7.6 pts-1.5%36,141
Read this straight
  • CHEAP is the strong call and the one a buyer acts on: it beats the blind base rate by a wide margin and the average move is real.
  • EXPENSIVE barely beats blind. We show it anyway. A scorecard that only prints its wins is a brochure, not a receipt.
  • Walk-forward, no lookahead. The seasonal average at each test week uses only the five prior calendar years, identical to the live scanner. Pooled across every eligible cut. Hit rate shown against the unconditional base rate over the same anchors.
  • This is a backtest of past behavior, not a guarantee of the next print. It is not trading advice.
Start free →See today's signals
Backtest generated 2026-05-16. Source: USDA AMS daily prints (protein market dataset). Methodology mirrors the live scanner signal exactly; the seasonal average at each test week uses only the five prior calendar years.