Operational Fires in the Age of Punishment

For over a century, military professionals have treated operational fires as the backbone of modern campaigns—shaping the battlefield, degrading enemy formations, and setting the conditions for maneuver. But what happens when that logic breaks down? What if long-range strikes become less about shaping operations and more about shaping narratives—used not to support a ground advance but to terrorize civilians and coerce political outcomes?

That is the story emerging from Russia’s war in Ukraine. Despite inheriting a military doctrine steeped in deep battle, reconnaissance-strike complexes, and precision noncontact warfare, Russia has consistently failed to employ operational fires in a way that reflects this legacy. Instead, firepower has become unmoored from maneuver. Russian missile salvos and loitering drone attacks increasingly appear to serve a punitive, strategic coercion logic rather than a campaign to dislocate Ukrainian defenses or synchronize effects across domains. Chaos reigns along an extended front line defined by human wave attacks, thousands of small first-person view (FPV) drones, and artillery fire vectored by strike and reconnaissance unmanned aerial vehicles (UAVs).

Drawing on an original dataset compiled by a Ukrainian researcher, the CSIS Futures Lab wrangled, cleaned, transformed, and augmented the raw data to use for analysis. This analysis critically evaluates the alleged disconnect between Russian operational fires and battlefield intensity. Using daily estimates of Russian personnel losses as a proxy for combat tempo, the Futures Lab analyzed their relationship with three types of firepower activity—(1) ballistic and cruise missile salvos, (2) strike and reconnaissance UAVs, and (3) Shahed drone attacks—to test whether Moscow was using fires to set conditions for ground advances. The findings are striking:

• There is no statistically significant relationship between Russian cruise and ballistic missile strikes and increased battlefield intensity three days later. In fact, the data shows the opposite: Larger missile salvos are associated with lower Russian personnel losses—suggesting that fires are either poorly targeted or employed for reasons unrelated to frontline outcomes.
• Reconnaissance and strike UAV activity, on the other hand, is strongly associated with battlefield intensity, supporting the idea that the security zone remains an active and lethal domain of modern war—a space where human-machine teams fight for information and survival.
• Finally, Russia’s widespread use of Shahed drones appears associated with rising battlefield intensity. Our model shows an increase in Russian personnel loss but this is likely a spurious effect explained not by operational integration of firepower in fighting but likely by material constraints—such as shortages of precision-guided munitions—and a shift toward a desperate strategy focused on punishment rather than achieving tactical breakthroughs.

Together, these findings point to a dangerous evolution in modern war: the rise of coercive fires unlinked from operational design and increasingly aimed at civilian populations. The war in Ukraine is not just a land campaign—it is an algorithmic war of attrition, fought with drone swarms and missile salvos designed as much to terrorize as to degrade military capability.

To meet this challenge, four policy recommendations emerge:

1. Sustain and expand air defense support for Ukraine, including both kinetic interceptors and enhanced intelligence sharing to reduce sensor-to-shooter timelines.
2. Disrupt the global supply chains enabling Russia’s long-range firepower, especially dual-use components from Iran, North Korea, and China.
3. Reimagine NATO’s concept of operational fires, integrating lessons from Ukraine into new doctrines that reflect dynamic targeting in the security zone and that leverage human-machine teaming.
4. Develop a counter-coercion playbook, preparing allied societies for an era of low-cost strategic strike campaigns designed to break will, not win battles.

Russia’s failure to employ fires operationally should not invite complacency—it should serve as a warning. The tools of war are evolving faster than the doctrines that guide them. It is time the United States caught up.
 

Operational Fires in Historical Perspective

In military doctrine, fires traditionally support maneuver: Armies use everything from artillery fire to long-range ballistic and cruise missile strikes to set conditions for ground forces to close with the enemy. In modern war, this concept of fire in support of maneuver extends to multiple domains and includes everything from space-based sensing to cyber and electronic attacks alongside a mix of aerial interdiction and close air support. According to U.S. joint doctrine, fires “assist air, land, maritime, space, cyberspace, and special operations forces to move, maneuver, and control territory, airspace, space, cyberspace, the electromagnetic spectrum, and key waters and to influence populations.”

In this framework, longer-range weapons systems like ballistic and cruise missiles set conditions at higher echelons, usually associated with divisions and corps. The intent is to strike deep targets linked to command and control (C2), key logistics nodes (e.g., fuel and ammunition), and troop concentrations, often up to 72 hours prior to attacks along the front, and coordinate these activities with planned close and rear military operations. These shaping fires are designed to degrade the enemy’s ability to generate combat power. The result is that the targeted enemy force struggles to coordinate (i.e., principle of mass) at the right time and place while also stumbling to conduct counterattacks (i.e., principles of objective and maneuver).

This concept has extended across different military organizations since the advent of longer-range artillery and aviation fires in the early twentieth century. Furthermore, some militaries—especially the Soviet and now Russian formations—are built to be fire-centric and place a higher premium on killing with fires than closing with the enemy.

In World War I, Russian military planners relied on related concepts like deep operations—a set of coordinated, long-range actions designed to shape where and how the enemy can respond, often by diverting, disrupting, delaying, or destroying key forces—as a solution to countering the firepower-intensive fronts of integrated trenches in World War I. That concept extended into the Soviet and even post-Soviet period, in which the integration of precision strike capabilities led to the rise of a reconnaissance-strike complex and noncontact warfare. In Russian military doctrine, the reconnaissance-strike complex refers to the “coordinated employment of high-precision, long-range weapons linked to real-time intelligence data and precise targeting provided to a fused intelligence and fire-direction center.” In turn, noncontact warfare is a Russian military concept describing battle waged by long-range precision fires. In both, the combination of depth and precision combine to hold the enemy at risk and set conditions for major operations.
 

Operational Disconnect in Ukraine

Given this mix of enduring practice and strategic culture, one would expect to see operational fires playing a critical role in the war in Ukraine. Yet, a mix of battlefield anecdotes and statistical observations suggest the contrary.

Russia used operational fires only for a short, intensive window to open its invasion. In all likelihood, Moscow was targeting preregistered (i.e., fixed) installations designed to blind Kyiv but preserve the critical infrastructure needed for an occupation. Of note, this opening salvo included a large-scale cyberattack against Viasat designed to limit Ukrainian satellite communications. In addition, Russian planners appear in hindsight to have suffered from overconfidence, limiting their firepower strike campaign because they assessed the war would last days, not years.

Statistical observations reveal a more enduring disconnect between operational fires and ground combat that extends beyond the flawed military planning of the opening days of the war. The Futures Lab leveraged a mix of data from Ukraine on daily battlefield casualty estimates and firepower strikes. Firepower strikes reflect the mix of cruise and ballistic missiles alongside reconnaissance and strike UAVs and one-way attack drones like Shaheds that Russia uses to bombard Ukrainian cities daily. These attacks can be further divided between weapons systems associated with operational fires (e.g., cruise and ballistic missiles), daily tactical activity (e.g., intelligence and strike drones used along the front line), and more strategic weapons used to terrorize the population and pressure political leaders (e.g., Shaheds launched in daily salvos largely against nonmilitary targets).

Since the start of the Russia-Ukraine War on February 24, 2022, an average of 10.64 cruise and ballistic missiles have been launched per day, with a median of 3 and a maximum of 127 in a single day. The use of other intelligence and strike UAVs (non-Shahed drones) is significantly lower, averaging about 2 per day with a daily max of 17, with some days seeing no launches at all. These lower levels reflect that these assets—likely larger drones used to support artillery fires, collect intelligence, and strike point targets—fly at higher altitudes than quadcopters and FPV drones. In contrast, Shahed long-range loitering UAVs are deployed in much greater numbers, with a daily average of 39.29, a median of 20, and a peak of 321 in one day. As stated previously, Russia appears to be using these systems following a strategic coercive logic of punishment designed to pressure Ukraine’s political leaders by holding the population at risk.

Closer examination also reveals interesting patterns that suggest Russia is not employing operational fires in the war in Ukraine. Rather, fires appear more dynamic and tethered to shifting conditions along the front line, indicative of a suboptimal attritional approach to joint fires and a separate strategic campaign of terror.

The Futures Lab evaluated Russia’s use of operational fires by introducing a time-lagged variable to capture delayed effects on battlefield intensity. If Moscow employed operational fires to set conditions for its ground campaign, one would expect to observe a positive relationship between increases in operational fires and increased intensity along the front line three days (i.e., 72 hours) later. This hypothesis suggests the type of sequence accounted for in doctrine, in which longer-range operational fires hit C2 and logistics targets in depth to set conditions for sequenced tactical actions along the front line. This sequencing of battles and effects is the essence of modern operational art.

To measure combat intensity, the Futures Lab used official estimates of daily Russian causalities. These losses included soldiers who were killed in action or wounded. One would expect that major offensive pushes against fixed, fortified front lines like those seen in Ukraine would be accompanied by higher casualty levels. This assumption creates the logic of the dependent variable (i.e., daily Russian personnel losses).

To measure firepower strikes, the Futures Lab adapted its firepower strike tracker data, separating and recoding categories and weapons into three groups: operational firepower strikes (i.e., cruise and ballistic missiles); security zone drone activity associated with the reconnaissance/counter-reconnaissance fight (i.e., reconnaissance and strike UAVs, often Group 3 or above); and Shaheds (i.e., long-range, one-way attack drones). If Moscow is employing its operational fires, one would expect to see a 72-hour lag between increased operational firepower strikes (i.e., ballistic and cruise missiles) and increased intensity along the front line. For additional details, including regression formulas and a breakdown of variables and limitations, see the appendix.

As seen in Model I from Table 2, there is no statistically significant relationship between operational firepower strikes using cruise and ballistic missiles and combat intensity three days (i.e., 72 hours later).¹ In fact, the coefficient (-0.333) is negative and in the opposite direction than expected, implying that increases in ballistic and cruise missiles are associated with decreased Russian personnel losses three days later. This indicates that for each additional missile launched, daily Russian personnel losses are estimated to decrease by 0.333 three days later.

To further test this relationship, the Futures Lab created a second and binary variable—large salvos—that only measured when a day’s combined cruise and ballistic missile strikes were over the first standard deviation. This measure acts as a robustness check but ends up yielding the same results. When analyzing operational firepower strikes above the first standard deviation, the model predicts an expected decrease in Russian daily personnel losses three days later. The negative coefficient runs counter to expectations. Specifically, days characterized by a large missile salvo (i.e., more than one standard deviation above the mean, approximately 20 missiles) are associated with a statistically significant decrease in battlefield intensity three days later, corresponding to approximately 212 fewer Russian personnel losses.²

This implies that large salvos are not associated with an increase in the proxy measure of battlefield intensity. The negative relationship holds but loses statistical significance when controlling for other strike and reconnaissance UAVs (Model II) and Shaheds (Model III). An alternative explanation could be that the attacks are so effective that they act as a substitute for Russian use of ground forces—saving the equivalent of two infantry companies. However, this hypothesis is not supported by battlefield accounts of Russia’s continued use of human wave attacks and operational approach of constantly pressuring key sectors along the front to gain territory. These operational patterns are not captured in the model and are instead absorbed into the error term, as indicated by the models’ generally low explanatory power, apart from some improvement in Model III. Additional threats that affect and limit inference are discussed in the appendix.

Another finding from Model II in Table 2 is that there is a positive relationship between strike and reconnaissance UAVs intercepted and increased combat intensity on the same day. Each additional reconnaissance or strike UAV launched on the same day is associated with an increase of approximately 13 daily Russian personnel losses, and the result is statistically significant. This finding is consistent with the concept of a security zone fight, where smaller, mobile units—modern skirmishers, including UAVs—carry out a reconnaissance and counter-reconnaissance fight in front of any formation. Their struggle is a fight for information, with each side trying to blind the other so they can gain an advantage and preserve combat power. In legacy Russian and Soviet doctrine, the security zone extends between 15 and 50 kilometers in front of the main line of troops. As a result, the positive relationship makes sense, as one would expect to see increased losses of UAVs along the front associated with increased battlefield intensity.

Lastly, Model III in Table 2 also shows that coercive, counter-value weapons linked to punishment strategies like the Shahed are associated with increased battlefield intensity. For each additional Shahed drone launched, Russian personnel losses increase by approximately three soldiers per day. The results are statistically significant. Here, the findings are likely spurious. The confounding factor at play is how Russia has had to adapt its warfighting strategy to material constraints and political imperatives to end the war. Both casualty levels and Shahed strikes have been increasing over time as a result of material losses and difficulty generating cruise and ballistic missiles at the same rate as low-cost one-way attack drones. In other words, Russia’s strategy by default explains the increase in both battlefield intensity and Shahed strikes in Ukrainian cities. Russia is desperately trying to break Ukrainian front lines and the will of its people to end the war on favorable terms. The result is daily human wave attacks and air raid sirens across Ukraine.
 

Policy Recommendations

Analyzing battlefield data from Ukraine provides insights into policy options for countering Russia and strengthening NATO. In particular, four recommendations emerge:

1. Sustain and expand air defense support for Ukraine.
   The United States and Europe should continue providing Ukraine with integrated air defense systems, combining interceptors with layered sensor networks and intelligence-sharing arrangements. This includes expanding counter-UAV capabilities and deploying mobile systems to protect critical infrastructure and civilian populations. If the United States reduces its footprint due to domestic political constraints, Europe must lean forward. Russia’s campaign of strategic punishment—manifested through relentless Shahed drone and missile attacks on cities—is unlikely to abate. Supporting Ukraine’s ability to defend its skies is both a moral imperative and a strategic necessity.
    
2. Disrupt the global supply chains enabling Russia’s war machine.
   The sheer volume of missiles and drones launched daily by Russia suggests a failure in existing sanctions regimes. Washington and Brussels must go beyond traditional sanctions and pursue targeted technology denial strategies. This includes cracking down on the transfer of dual-use technologies and components—especially microelectronics and drone parts—from Iran, North Korea, and China. These actors are now vital cogs in a distributed arsenal supplying Russia with the means to sustain its long-range fire campaigns. Export enforcement teams should be embedded with intelligence services to map and dismantle these networks in real time. This effort must also extend to third-party cutouts and shell companies acting as proxies.

3. Reimagine operational fires and human-machine integration for NATO.
   The Ukraine war reveals that legacy doctrine linking fires to maneuver may be insufficient in modern war. NATO must invest in new operational concepts that reflect the evolving character of fires—where intelligence, surveillance, and reconnaissance (ISR) drones, loitering munitions, and long-range strikes are used to shape battlefields, but in a manner that supports both operational maneuver and more dynamic targeting in the security zone. This reimagining should include red teaming adversary use of coercive fires, as seen in Russian strikes on civilian targets, and testing new fires—human-machine integration (HMI) constructs that blend machine reconnaissance, networked kill chains, and AI-enabled decision cycles. The security zone fight of the future will be a contest between drone swarms, decoys, jammers, and algorithmic kill webs. NATO must ensure it dominates this battlespace.
    
4. Build a counter-coercion playbook for strategic strike campaigns.
   Russia’s daily use of Shahed drones points to the resurgence of an old form of warfare: punishment. Future adversaries—state or nonstate—will increasingly use low-cost, long-range systems to hold civilian infrastructure at risk and manipulate domestic political will. NATO should prepare now by developing counter-coercion strategies that include hardening infrastructure, distributing critical services, integrating national warning systems, and creating escalation management protocols. Strategic messaging must be part of the defense plan, helping shape public resilience under sustained aerial and psychological assault.
    

Conclusion

Russia’s war in Ukraine is not just a throwback to World War I trench warfare and Soviet deep battle—it is a live demonstration of how authoritarian regimes adapt old ideas to new technologies. Instead of using fires to support maneuver, Moscow uses them to punish civilians, shape perceptions, and degrade morale. This is a war of attrition by algorithm, where drone salvos replace doctrine and coercion substitutes for combined arms.

The United States and its partners cannot afford to watch this play out from the sidelines. Support for Ukraine is more than a moral cause—it is a strategic laboratory. Every Shahed intercepted, every sensor spoofed, every Russian tactic studied gives NATO an edge in tomorrow’s conflicts. But only if the United States and its allies act—fast, forward, and together.

Benjamin Jensen is director of the Futures Lab and a senior fellow for the Defense and Security Department at the Center for Strategic and International Studies (CSIS) in Washington, D.C. Jose M. Macias III is an associate data fellow with the CSIS Futures Lab.

This report is made possible by general support to CSIS. No direct sponsorship contributed to this report.

Please consult the PDF for appendix and references.