The drone-war public record points in the same direction from both sides.
Russian MoD claims say Ukrainian UAV losses exploded. Ukrainian Air Force reports say Russian long-range drone attack packages expanded sharply. The measures are different; the strategic signal is similar.
Trend evidence, not battlefield accounting.
Russian MoD data is a claimed Ukrainian UAV-loss series. Ukrainian Air Force data is a reported Russian strike-package series. The page compares timing, acceleration, and reporting emphasis, not absolute truth.
Drone-war trend comparison, with anomaly-resistant scaling.
Russian MoD UAV-loss claims and UAF Russian drone-launch reports normalized to each series p98. Robust scale is the default: the chart clips the visual y-axis around the high-percentile trend and marks clipped points at the ceiling, so isolated spikes do not flatten the rest of the series.
The cautious argument is stronger than a direct numerical comparison.
A claimed Ukrainian UAV loss and a reported Russian drone launch are not the same measurement. Treating them as interchangeable would overclaim the data. Treating them as two official/public reporting streams is more defensible.
On that basis, the finding is clear: both sources increasingly organize their public record around drones, especially from 2024 onward and into 2025-2026.
Russian MoD side
A cumulative claimed-loss stream. Its value here is not verification; it is the rising scale and prominence of UAVs in Russian official attrition claims.
UAF side
A strike-package stream. It records reported Russian drones and missiles launched, neutralized, suppressed, locally lost, or otherwise failing to reach targets.
Comparison logic
The comparison is about trend, acceleration, and narrative emphasis. The interactive graph defaults to robust scaling so a few spikes cannot make the rest unreadable.
Neutralized is broader than shot down.
UAF reports use several terms: shot down, suppressed, locally lost, did not reach targets, and other variants. The broader aggregate is labeled as reported neutralized only when the wording supports that broader concept.
- Telegram public preview pages are public source material, not a formal API.
- Combined Shahed, Gerbera, Italmas, and decoy wording is treated as a combined package unless the source gives a split.
- Same-date updates are aggregated cautiously to reduce double counting in the daily public-record series.
- Missile parsing is useful for broad mix context, but drone-package parsing is the stronger dataset.
Primary streams and context checks
Ukrainian Air Force Telegram
Primary source for reported Russian drone and missile attack packages.
Russian MoD English Telegram
Primary source for cumulative claimed Ukrainian equipment losses.
ISIS Online Shahed/Geran analysis
External context source for Russian Shahed/Geran deployment patterns.
ACLED Ukraine Conflict Monitor
External context source for broader event patterns in Ukraine.
Different metrics, same direction: drones have become central to official attrition narratives in this war. That conclusion does not require accepting either side's figures as verified losses; it only requires reading the public reporting streams as reporting streams.