# Anomaly Scores Differ on Same UI

**URL:** <https://discuss.elastic.co/t/anomaly-scores-differ-on-same-ui/383251>\
**Category:** Kibana\
**Created:** [November 5, 2025, 5:46pm UTC](https://discuss.elastic.co/t/anomaly-scores-differ-on-same-ui/383251 "2025-11-05T17:46:31Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![erikg](https://avatars.discourse-cdn.com/v4/letter/e/91b2a8/32.png) [@erikg](https://discuss.elastic.co/u/erikg)\
**Post date:** [November 5, 2025, 5:46pm UTC](https://discuss.elastic.co/t/anomaly-scores-differ-on-same-ui/383251/1 "2025-11-05T17:46:31Z")

</div>

Hello,  
I've been leveraging the Anomaly Detection jobs in Elastic. I think this might be a visual bug but I see an influencer with an anomaly score of 98:

 ![image](https://us1.discourse-cdn.com/elastic/original/3X/1/7/17968d0410499ef1a7abf49681442c0cdac7a9d5.png)

But down at the bottom it reveals a score of 77:

 ![image](https://us1.discourse-cdn.com/elastic/original/3X/d/e/dede123e3f0aa7a92b24290e0cb741587936ef76.png)  
Even the alert section of Anomaly Explorer shows 77:

 ![image](https://us1.discourse-cdn.com/elastic/original/3X/e/d/ed2fc75107ca7d5fd154026fc5d8d0f3794fc233.png)

---

<div class="post-metadata">

**Author:** ![Tortoise](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/tortoise/32/147587_2.png) [@Tortoise](https://discuss.elastic.co/u/Tortoise)\
**Post date:** [November 9, 2025, 6:38am UTC](https://discuss.elastic.co/t/anomaly-scores-differ-on-same-ui/383251/2 "2025-11-09T06:38:33Z")

</div>

Hello @erikg

I do not think there is a visual bug :

I was checking in demo application and i see similar type what you have highlights like if we check advertservice shows 87 as the anomalyscore but the individual values are more when we check for that time duration. I believe 87 is the normalized max score for that bucket

 ![image](https://us1.discourse-cdn.com/elastic/original/3X/9/9/996840e4d052491bcfa62f542bd813c7115b88e4.png)

If we take 1 bucket :

 ![image](https://us1.discourse-cdn.com/elastic/original/3X/9/5/95db71639e5bd4d17facb89f4dfebc7fd94456c5.png)

Documentation :

> **[Anomaly score explanation | Machine Learning in the Elastic Stack \[8.19\] |...](https://www.elastic.co/guide/en/machine-learning/8.19/ml-ad-explain.html)**

I am also not sure how it is calculated 🙂

Thanks!!