Three Ways to Execute: Clustering T-Side Attack Formations on de_ancient in Professional Counter-Strike 2

This paper is the first to study offensive formations in professional Counter-Strike 2. Using the bomb plant as an anchor, we analyze T-side position trajectories in a 20-second pre-plant window across 459 rounds that reached a bomb plant, drawn from 57 professional matches in 19 Tier 1 tournaments in 2026. We build a quantitative pipeline for studying offensive formations: for each round, we extract eight formation features (e.g., spatial spread, dispersion, centroid displacement, and speed) and cluster a filtered 393-round subset with K-Means. We identify three stable offensive patterns on de_ancient—Spread Control, Executed Push, and Fast Rush—and find no evidence that pattern choice depends on the bombsite. We report two statistical findings. First, the offensive pattern is significantly associated with economic state (χ²(6) = 27.36, p = 1.2 × 10⁻⁴, Cramér’s V = 0.187): pistol rounds tend toward Fast Rush, eco rounds toward Executed Push, and full-buy rounds toward Spread Control. Second, the occurrence of a firefight within the window is strongly associated with pattern membership (p = 1.1 × 10⁻⁸): 89.4% of rounds with a firefight are Executed Push, whereas 43.8% of clean rounds are Spread Control. Preprint. Under review at the 2027 IEEE Conference on Games (CoG).

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-14
DOI
https://doi.org/10.5281/zenodo.24223778
Primary Topic
Artificial Intelligence in Games
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Three Ways to Execute: Clustering T-Side Attack Formations on de_ancient in Professional Counter-Strike 2

Yixiao Wang
Zenodo (CERN European Organization for Nuclear Research)
Artificial Intelligence in Games
preprint

Three Ways to Execute: Clustering T-Side Attack Formations on de_ancient in Professional Counter-Strike 2

Yixiao Wang
preprint en

Abstract

This paper is the first to study offensive formations in professional Counter-Strike 2. Using the bomb plant as an anchor, we analyze T-side position trajectories in a 20-second pre-plant window across 459 rounds that reached a bomb plant, drawn from 57 professional matches in 19 Tier 1 tournaments in 2026. We build a quantitative pipeline for studying offensive formations: for each round, we extract eight formation features (e.g., spatial spread, dispersion, centroid displacement, and speed) and cluster a filtered 393-round subset with K-Means. We identify three stable offensive patterns on de_ancient—Spread Control, Executed Push, and Fast Rush—and find no evidence that pattern choice depends on the bombsite. We report two statistical findings. First, the offensive pattern is significantly associated with economic state (χ²(6) = 27.36, p = 1.2 × 10⁻⁴, Cramér’s V = 0.187): pistol rounds tend toward Fast Rush, eco rounds toward Executed Push, and full-buy rounds toward Spread Control. Second, the occurrence of a firefight within the window is strongly associated with pattern membership (p = 1.1 × 10⁻⁸): 89.4% of rounds with a firefight are Executed Push, whereas 43.8% of clean rounds are Spread Control. Preprint. Under review at the 2027 IEEE Conference on Games (CoG).

Zenodo (CERN European Organization for Nuclear Research)
Artificial Intelligence in Games
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon