# Random forest model R² looks "too good to be true"

**URL:** <https://community.worldpop.org/t/random-forest-model-r-looks-too-good-to-be-true/61>\
**Category:** Tech Support\
**Tags:** modelling, statistics\
**Created:** [September 2, 2026, 1:13pm UTC](https://community.worldpop.org/t/random-forest-model-r-looks-too-good-to-be-true/61 "2026-09-02T13:13:45Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Graeme\_test](https://sea1.discourse-cdn.com/flex023/user_avatar/community.worldpop.org/graeme_test/32/27_2.png) [@Graeme\_test](https://community.worldpop.org/u/Graeme_test)\
**Post date:** [September 2, 2026, 1:13pm UTC](https://community.worldpop.org/t/random-forest-model-r-looks-too-good-to-be-true/61/1 "2026-09-02T13:13:45Z")

</div>

Hi WorldPop,

I’m building a random forest model for dasymetric redistribution similar to WorldPop’s approach, and I’m getting an R² of 0.97 on my validation. That seems suspiciously high. Am I doing something wrong, or is this plausible?