X Algorithm Scorer
Bangermeter
Score Tweets With X's Open-Sourced Algorithm
Bangermeter is a free, open-source Chrome extension that scores every post in your X feed using the last ranking weights published by X. It runs locally in your browser, collects no data, and requires no account. The extension provides a score badge on every tweet, allowing users to understand the content appeal and earned engagement of their posts without any data collection.
About Bangermeter
Bangermeter is a free, open-source Chrome extension that scores every post in your X feed using the last ranking weights published by X. It runs locally in your browser, collects no data, and requires no account. The extension provides a score badge on every tweet, allowing users to understand the content appeal and earned engagement of their posts without any data collection.
Upvotes over time
Last 24 hoursTotal upvotes
53
Last 24h
—
Peak hour
—
No upvote activity in the last 24 hours.
7 comments
A must-have tool for creators and marketers! Empower your X experience & beat your competition.
Hey all, if you're reading this PLEASE drop a 5 star review if you liked the tool on the chrome store. It helps me make more tools like this in the future! https://chromewebstore.google.com/detail/bangermeter-%E2%80%94-x-algorithm/hecnkfmbffaccdljhpkcoefickilfgmh
WHAT'S NEW IN 0.9.5 • Corrected a number in the breakdown panel. It said For You hard-filters 670 accounts reported to Brazil's Electoral Court for the 2026 election. The real figure is 665. The correction was made in the source before 0.9.4 shipped and missed that build by a few hours; this release carries it. If you read 670 anywhere, that was us, not X. • Three panel and tooltip strings used British spellings. Fixed, including one inside a passage quoting X's own wording.
Revision on the main screenshot - the math has changed since that was uploaded. Replies are NOT worth 27 likes. See documentation on my website or github for more info.
WHAT'S NEW IN 0.9.4 • Scores are now calibrated against real data. The reference rates were measured from 158 real timeline posts instead of estimated. If you used an earlier version, expect your engagement scores to read roughly 15 points lower — the old scale put a typical post at 67 when it should have been 50. Nothing broke; the midpoint moved to where it belongs. • Updated for X's August 14 clarifications. X added comments to the ranking code confirming that weights multiply a predicted probability, not a count, and explaining that Report carries a large coefficient because reports are over 1000x rarer than likes — not because it's a bigger punishment. All 26 weights re-verified against the update; none changed.
• Two facts about transparency, now in the breakdown panel: engagement only counts toward ranking if the post reached the reader through their Home Timeline (sending your own link to a group chat does nothing), and mass report campaigns don't straightforwardly bury a post, because predictions are personalised per viewer. • The draft meter in the compose box no longer covers the Post button, and clears itself once you've posted. • 120 automated tests, up from 33. Every published weight is checked against X's source individually.
Bangermeter has been updated for the algo drop X made yesterday (8/13/2026) It now holds all the currently released weights! Try it out today.