---
title: "Discussion on \"De-overlap before you believe any walk-forward result\" | Hashnode"
description: "Discussion on \"De-overlap before you believe any walk-forward result\". The same data scored two ways: 2.4 standard errors, then 1.4. The effect grew and the significance collapsed. De-overlap before you believe a backtest."
url: "https://hashnode.com/posts/de-overlap-before-you-believe-any-walk-forward-result/6ace1fdb3a2487de9a475f39"
---

[Feed](https://hashnode.com/feed) /

Discussion

# [De-overlap before you believe any walk-forward result](https://openred.hashnode.dev/de-overlap-before-you-believe-any-walk-forward-result?utm_source=hashnode&utm_medium=feed)

The same data scored two ways: 2.4 standard errors, then 1.4. The effect grew and the significance collapsed. De-overlap before you believe a backtest.

Oct 13 · 2 min read · openred.hashnode.dev  
Read the article

-   [#data-science](https://hashnode.com/tag/data-science)
-   [#statistics](https://hashnode.com/tag/statistics)
-   [#machine-learning](https://hashnode.com/tag/machine-learning)
-   [#python](https://hashnode.com/tag/python)

## Written by

[Michael Hairetis](https://hashnode.com/@michaelhairetis)

@michaelhairetis · Marietta, GA USA

  
[View profile](https://hashnode.com/@michaelhairetis)

sharing my software engineering thoughts and ideas

Full stack engineer. 25+ years of production software engineering across e-commerce, workflow automation and agentic ai

## Responses 1

Markdown supported · ⌘↵ to post

[Arnold Holm](https://hashnode.com/@stratcorealpha)

[Oct 14](https://hashnode.com/posts/de-overlap-before-you-believe-any-walk-forward-result/6ace1fdb3a2487de9a475f39/comment/6acf699acbdfbef752afa663)

Your 912-to-163 example separates effect size from certainty well. One extra check I would keep is the anchor offset: repeat the non-overlapping selection for each feasible starting offset, then show the range of edge estimates rather than choosing the best offset. A single 30-minute grid can accidentally select favourable observations. This is a sensitivity check, not a set of independent tests.
