Skip to content
Language
Menu
The open jaws of a tyrannosaur skeleton seen from below, both rows of teeth filling the frame against black.

Why is your EPEF wrong if thinning counts as mortality?

Thinning removes LIVE birds from the shed. If mortality is computed as birds placed minus birds alive, those birds count as dead: a flock at 4.75% real mortality can display 23.97%. The error breaks no screen, which is why it survives for years.

No form, and not out of generosity#

The full text is open below and in markdown. I do not ask for your email in exchange, for two reasons worth saying out loud:

  1. A document behind a form is invisible. No search engine indexes it and no answer engine cites it. Gating it would spend the best argument I have on collecting addresses.
  2. Asking for the data saves me no obligation. Keeping it at an email provider instead of in my own database does not change who answers for it. The only thing that genuinely lowers the bar is not needing it.

If you want to talk after reading, the demo page is one click away.

What is the argument, in short?#

Thinning removes live birds from the shed. If cumulative mortality is computed as birds placed minus birds alive, those birds count as dead.

ItemBirds
Placed20,000
Removed alive when thinning3,844
Actually dead950
Mortality, naive calculation23.97%
Real mortality4.75%

Nineteen percentage points on the same flock, the same days and the same records. And since survival sits in the EPEF numerator, the index falls in the same proportion: a flock at 360 shows up at 287.

The error breaks no screen. It produces a number that looks normal and decisions taken on it, which is why it survives for years.

How do you check it in your data this week?#

  1. Find a flock with a staggered harvest and look at its mortality. Above 15% with no documented health event, it is almost certainly counting the thinning.
  2. Add up the exits: placed, minus recorded deaths, minus birds removed alive should equal the final harvest exactly.
  3. Compare two twin flocks, one thinned and one not. If the thinned one is worse on mortality and EPEF but identical on conversion, you have your answer.

Where is the full text?#

The complete development — why EPEF amplifies the error, why it gets copied to five different places, and what has to change in the data model rather than just in the formula — is in the article, and also in markdown for anyone who prefers to read it outside a browser.

Read the full articleDownload as markdown

Want to see it on your own flocks?

Leave me your details and pick how to continue: book a date and time, or send me the same information on WhatsApp.

Authorisation for the processing of personal data

Data controller: Santiago Blandón Ferro — Armenia, Quindío, Colombia. By ticking this box you authorise the controller to process the personal data you provide in this form, for these purposes and no others: 1. To contact you and answer your enquiry. 2. To arrange and hold a product demonstration. 3. To send you information about the product and about the published technical material. 4. To keep the record of this authorisation as evidence that you granted it. Your data will not be sold, rented or transferred to third parties for their own purposes. The processing relies on providers that host the website and the database (Vercel, Render and Supabase, in the United States), which act solely on the controller’s instructions and may not use the data for their own purposes. The United States is on the list of countries with an adequate level of protection under Circular 005 of 2017 of the Superintendencia de Industria y Comercio. If you choose to book online, your data is used to create the meeting in the controller’s calendar (Cal.com and Google, acting on the controller’s instructions). If you choose WhatsApp, the message is sent by you, from your own account, and nothing is forwarded on your behalf. Providing this data is voluntary. Sensitive data and data of children or adolescents are not collected, and you are asked not to include any. As the data subject you may: know, update, rectify and delete your data; revoke this authorisation; request proof of the authorisation granted; be informed of the use made of your data; and lodge complaints with the Superintendencia de Industria y Comercio. To exercise them, write to santiagoblandonferro@gmail.com or call 311 663 9606. You may also use SIC Facilita, the Superintendencia’s own platform. Enquiries are answered within 10 business days and complaints within 15 business days, extendable as provided in Articles 14 and 15 of Law 1581 of 2012. Version 1.1.

There is no newsletter, no automated sequence, and your email is not shared with anyone.