HLTV Updates Rating 3.0 With New Accuracy Improvements

HLTV has released a new accuracy update for Rating 3.0. The revised system uses data from 29,030 CS2 maps, including 9,029 played on LAN, and applies the changes to previous player statistics.

The update does not introduce Rating 3.1. Instead, it improves the calculations behind Round Swing, eco-adjustment and map-specific win probabilities.

What Changed in Rating 3.0?

HLTV updated the win probability data behind Round Swing and eco-adjustment using every CS2 match in its database through September 29, 2026.

Cache now has its own map-specific data. Previously, the system used general averages because the map did not have a sufficient CS2 sample.

Kills made when a bomb explosion can no longer be prevented will not increase Round Swing. This applies when fewer than 10 seconds remain, or fewer than five seconds if the CT player has a defuse kit. Such kills are treated similarly to exit frags because they no longer affect the probability of winning the current round.

HLTV also changed how post-plant impact is distributed, removed the possibility of receiving negative Round Swing after a kill and stopped stand-ins from earning Round Swing for rounds they did not play.

How Much Did Player Ratings Change?

The effect is limited for most players. Across 6,687 maps played between July and September 2026, the average change to a single-map rating was only 0.021.

However, 9.4% of player-map ratings changed by at least 0.05, while 1% moved by 0.10 or more. The largest recorded adjustment reached 0.31.

Cache received the biggest overall changes because it now uses map-specific data. Inferno was also affected due to its high number of post-plant situations and the revised bomb timer logic.

The updated system should provide more accurate context ahead of the HLTV Awards and the Top 20 Players of 2026 ranking.

Statistics can change when the model receives better data. Match probabilities should work the same way.

Follow upcoming CS2 matches on Winio and compare team form, map strength and AI-powered predictions before every series.