Armor Exchange Ratios at Kursk

A friend recently sent me part of an article from a foreign magazine that pulled data from my Kursk book, which in translation reads:

The German tank losses in the battle at Kursk has been debated. Until the 1990s, Soviet propaganda figures dominated, which claimed that the Germans lost 2,952 tanks and 195 assault guns. Subsequently, the pendulum swung back, as research in German archives showed that no more than 278 German tanks were lost during the battle at Kursk 5-23 July.

By far the most thorough study of the battle has been conducted by Chris Lawrence and is presented in his 1,600 page book about Kursk, which was released in 2015. Lawrence shows that previous figures, which suggest that ten Soviet tanks were lost for each German tank lost, is a comparison of apples and oranges….

….Lawrence finds that if German tanks included in the same way as Soviet losses, von Mansteins formations lost 1536…tanks and assault guns 5-18 July. The opposing Soviet forces lost 2471 tanks and assault guns, a much less lopsided ratio.

I took out the parts that of that quote that I didn’t want to debate or that gave the wrong impression.

I have also been bothered by other published comparisons of tank losses, where the author focused on total German tanks destroyed vice total Soviet tanks destroyed. This, of course, produces a very lop-sided exchange figure. This is not a valid measure of combat effectiveness. What would be a valid measure is total tanks destroyed and damaged compared to total tanks destroyed and damaged. I can talk for awhile about the differences in the German and Soviet repair systems and philosophies, but to try to shorten the discussion, lets just say that the Germans refused to write off any tanks if possible, whereas the Soviets often willingly wrote off tanks because they had spare tanks in their units, a steady flow of tanks from the factory, and I suspect a lack of repair parts (a fundamental flaw in the Soviet system, both military and civilian). And, of course, in many cases the Germans held the field of battle. In many respects you are comparing apples and oranges.

For my book I ended up comparing total destroyed, damaged and broken down over night compared to total destroyed, damaged and broken down over night. As my primary trusted source of tank losses was the ready-for-action reports for both sides from day-to-day, I ended up picking up those that broke down and were not returned to duty the same day along with those destroyed and damaged. It was pretty typical for the Germans to report 1 tank destroyed for around every 10 tanks not ready-for-action. When the Soviets reported total destroyed for a day (which they often did), it was not unusual that the number ready-for-action the next day indicated more were damaged or broke down. Now, broken down tanks may make up 10 to 20% of the total losses for a day, but total losses for the day (regardless of cause) is clearly a more valid measure of combat effectiveness than total destroyed.

So, I do end up with a very different comparison of the exchange ratio of armor compared to some authors. Also, I count tank-like vehicles (like Sturmgeshuetz and Marders) as tanks. I also count any German light tanks and command tanks as tanks. Most comparisons I gather count Soviet assault guns and light tanks in their figures, so they need to make sure that both sides forces are counted by the same rules. This results in my having higher figures for German tanks than some other sources do, among many other differences. Now I do take the time to break down the counts by exact vehicle type in my appendices, so anyone who wants to calculate otherwise can do so. Note my figures include tanks lost to mine damage, which very rarely destroys tanks. Using total destroyed (or more to the point, destroyed totally) simply ignors the effects of mines on the battle.

The exchange ratios for armor are discussed in my book on pages 640-641, 744-745, 809-811, 1021-1022, and 1209-1211. The figures of 2,471 Soviet tanks destroyed, damaged or broken down and 1,536 German tanks destroyed, damaged and broken down comes from page 1210, among other places (pages 1338, 1339, 1340, 1367 and 1368). This is a 1.61 armor exchange, although the majority of tanks were probably not taken out in combat between tanks.


P.S. Picture is labeled: “Crew of a Wehrmacht repair unit working on a Panzer III.” Source:

Share this:
Christopher A. Lawrence
Christopher A. Lawrence

Christopher A. Lawrence is a professional historian and military analyst. He is the Executive Director and President of The Dupuy Institute, an organization dedicated to scholarly research and objective analysis of historical data related to armed conflict and the resolution of armed conflict. The Dupuy Institute provides independent, historically-based analyses of lessons learned from modern military experience.
Mr. Lawrence was the program manager for the Ardennes Campaign Simulation Data Base, the Kursk Data Base, the Modern Insurgency Spread Sheets and for a number of other smaller combat data bases. He has participated in casualty estimation studies (including estimates for Bosnia and Iraq) and studies of air campaign modeling, enemy prisoner of war capture rates, medium weight armor, urban warfare, situational awareness, counterinsurgency and other subjects for the U.S. Army, the Defense Department, the Joint Staff and the U.S. Air Force. He has also directed a number of studies related to the military impact of banning antipersonnel mines for the Joint Staff, Los Alamos National Laboratories and the Vietnam Veterans of American Foundation.
His published works include papers and monographs for the Congressional Office of Technology Assessment and the Vietnam Veterans of American Foundation, in addition to over 40 articles written for limited-distribution newsletters and over 60 analytical reports prepared for the Defense Department. He is the author of Kursk: The Battle of Prokhorovka (Aberdeen Books, Sheridan, CO., 2015), America’s Modern Wars: Understanding Iraq, Afghanistan and Vietnam (Casemate Publishers, Philadelphia & Oxford, 2015), War by Numbers: Understanding Conventional Combat (Potomac Books, Lincoln, NE., 2017) , The Battle of Prokhorovka (Stackpole Books, Guilford, CT., 2019), The Battle for Kyiv (Frontline Books, Yorkshire, UK, 2023), Aces at Kursk (Air World, Yorkshire, UK, 2024), Hunting Falcon: The Story of WWI German Ace Hans-Joachim Buddecke (Air World, Yorkshire, UK, 2024) and The Siege of Mariupol (Frontline Books, Yorkshire, UK, 2024).
Mr. Lawrence lives in northern Virginia, near Washington, D.C., with his wife and son.

Articles: 1516


  1. There are multiple ways of comparing them, often debated, but two variants come into my mind: Either irrecoverables, “evacuated” term vs the German “Totalausfall” term. Note that they are not exactly the same but deliver a good approximation (a vehicle was rarely sent back to homeland repairs or totally scrapped off, but this included also vehicles which were lost on the battlefield, burned out/abandoned etc.). This offers a more transparent picture for the later phases of the war, when the Soviets controlled the field of battle and could recover their vehicles effectively, many vehicles were sent into the battle twice or simply disassembled.
    Example, for 1943, 32,935 tanks and spgs evacuated from combat, 1,460 stuck in bogs/mud, 9,344 fixed at homeland repair depots, dismantled N.A. (but my assumption is around 16,000). The Soviets lost 32,935 AFVs in 1943, the actual figure to be compared to the German number of 9,092 tanks and spgs for that year respectively. That is an exchange rate of 3.62 and it does not even include SdKfzs or PSWs, BA-64s etc.
    Material losses can stand in relation to personnel losses as well.

    The other one is close to the method you have listed in the post, but it does have a few inconsistencies and does not always reflect losses as a consequence of direct combat (between units of some sort). Furthermore, operational readiness figures are not always a reliable information on this matter, this has been pointed out by Dr. Zetterling before.
    Also from: Niklas Zetterling & Anders Frankson (1998): Analyzing
    World War II eastern front battles , The Journal of Slavic Military Studies, 11:1,
    p.191: “That the number of tanks temporarily disabled compared to total writeoffs
    is greater during offensive operations than during defensive operations
    is of course a consequence of the likelihood of a damaged vehicle being
    towed to a workshop. Another explanation is mines, which rarely destroy
    a tank, but can disable them sufficiently to require repairs in workshops. It
    is much more likely that an attacking tank will hit a mine.”

    Also, listing the original TsAMO documents we can comment on the losses between 5-23 July and the 5-18 July for the Voronezh front.

    Losses of armor 5-15.7.43:
    13 T-60s, 190 T-70s, 844 T,34, 21 KV, 48 Churchill Mk-IV, 41 M3 Lee, 30 M3 Stuarts, 3 MK-II, 12 Mk-IIIs, 22 SU-122, 7 SU-76. Sum 1,204 tanks and 29 Assault guns written off.
    Damaged: 7 KV, 419 T-34, 177 T-70, 4 T-60, 20 Mk-IV, 10 M3 Lee, 8 M3 Stuart, 9 Mk-II, 11 Mk-III. 655 tanks damaged.


    Losses of armor 15-23.7.43:
    3 KV, 237 T-34, 87 T-70, 1 T-60, 8 Mk-IV, 22 M3 Lee, 9 M3 Stuart, 18 SU-122, 10 SU-76. For a total of 367 tanks and 28 SP guns written off.
    3 KV, 115 T-34, 43 T-70, 2 T-60, 5 Mk-IV, 8 M3 Lee, 3 M3 Stuart, 11 SU-122, 4 SU-76. Sum 179 tanks and 15 SP guns damaged.

    For a total of 2,477 damaged and written-off vehicles.

    The write off figure was usually 15% vs a approx. 50%, which would yield 1,200 for the Soviets and 230 for the Germans, ratio: 5.22, much closer to personnel casualties.
    If combat is the ultimate measure of effectiveness, then casualty infliction is a byproduct of the quantity and quality of the force, the training x the gear it possesses and the circumstances present of that day. Salvaging, fighting, commanding are all part of this. The goal is attrition and to inflict the maximum of irrecoverables on the enemy.
    The Germans had shortcomings when massive territorial losses occured (this was also very true for Normandy) or spare parts may not have been available. New units were sent with replacements, while on the other hand the Soviets relied on an influx of AFVs, for two reasons: 1.) They focused more of their smaller resources on tanks 2.)They had higher losses and lower quality equipment. Repair figures can reflect standards as well, but they can also mean that certain units vehicles simply survived longer. This is also very true from an economic perspective. The most relevant thing to consider here is the exposure (duration) to combat and context.
    Another thing can be deduced from the figures: It was far more difficult for a Soviet unit to destroy a german tank, while it was significantly easier to annihilate a Soviet AFV for a German unit, understandable but a very important fact has been ignored frequently: Soviet projectiles were of poor quality.

    Also, I think there is a typo, you meant 1,536 German tanks and not Soviet tanks destroyed.

    • Judging by the statistics available on Normandy tank crew losses, AFV personnel had a high chance of surviving such engagements from woundings to usually 1 dead (or 0.98), contrary to the poor folks in the foxholes.

      However, the linked article, or the introduction for that matter is nonsensical: “[…] Subsequently, the pendulum swung back, […]
      As if destroying the tank forces would result in an instant defeat of the entire armed forces. People on the internet always misinterpret the nature of WW2 or warfare in general. They have that “image” of WW2 imprinted in their heads, of dashing, dueling tanks. Must be movie influence. It is the same with Ninjas and masks, or Vikings, cool tattoos and horned helmets.

      • Stiltzkin said: “They have that “image” of WW2 imprinted in their heads, of dashing, dueling tanks. Must be movie influence. It is the same with Ninjas and masks, or Vikings, cool tattoos and horned helmets.”


  2. For the purpose of evaluating performance of army combat units, a number of variables can be used. The selection should be based on relevance. If reliable data for the most relevant variables cannot be obtained, then there may be proxy variables that can serve reasonably well.

    If we focus on the choice between what I will henceforth call alternative 1, destroyed vs destroyed, and alternative 2, all tanks put out of action for any reason vs all tanks put out of action, our choice should be based on which these alternatives that is most relevant.
    Why are losses relevant for combat performance evaluation? The main reasons would appear to be i) that reduction of enemy combat power is a goal in itself, ii) that it is a means to attain other goals, iii) that it tends to correlate with other variables and can thus be a useful proxy variable.
    To me, it seems that in general, the first of these is the most important. The reason is that few battles, if any, during WWII were of much importance in the greater context. What mattered was long term effects on the enemy formations, not temporary effects. This suggests that actually destroying (or capturing) a tank is of far greater importance than merely damage it and allowing the enemy to bring it back to action after repairing it. Also, even if we look at reasons ii) and iii), it is far more valuable to destroy than damage a tank.

    There are other reasons why we would prefer alternative 1. Many tanks (but the percentage varies considerably with the circumstances) are out of action due to other factors than enemy action. Should they be included? I think not.

    As it is, I find it hard to see any important advantage of alternative 2.

    One could of course argue that if one side reported a substantial number of tanks as damaged, when in fact they were destroyed, this would distort comparisons. Of course, there are occasions when this happens. For example, closer inspection in workshops may reveal that it is not worth repairing a tank, or a tank may be cannibalized for spare parts. However, in such cases the tank will be recorded as lost, but on another date. In short periods of time, this may pose problems, but hardly in the long run.
    Anyhow, I have yet to see a convincing argument that this is actually something that occurs often enough to be a significant problem. But even if it would be, it does not follow that using alternative 2 is a better choice.

    Another objection to alternative 1 is that the Soviet forces did not repair as many tanks as the Germans. But from this observation does not follow that alternative 2 is better than alternative 1. The most fundamental is that one important reason we consider own losses as an important factor when assessing combat performance is that spending resources means reducing chances to have an impact in the future. If the Soviets did not repair tanks that were repairable, they wasted valuable resources and that should result in an unflattering assessment of their performance.
    In the absence of better data, I would guess that there are two main reasons for the fact that the ration destroyed/ damaged differ between the Germans and the Soviets at Kursk. The Germans usually controlled the battlefield and could recover damaged tanks more easily. Also, the damage on Soviet tanks probably was more serious. In the absence of detailed data on the damaged tanks, these are of course just hypotheses, but in this case, there are no other hypotheses that appear more reasonable, as I see it.
    As to the Soviet recovery and repair system being less well developed. This can be due to various factors, such as an economy not focused on spare part production, a lack of trained mechanics and other personnel to man field workshops, etc. But why would lack of trained personnel not be reflected in calculations of combat efficiency? Lack of skilled fire directors, commanders, tank crews, staff officers, engineers, NOC:s etc would result in poor combat performance as well. Why not let lack of skilled personnel have that effect in this case as well. I would regard maintaining the combat value of a unit something that deserves very high priority and vital to combat efficiency.

    So in the end, I find that alternative 1 remains the better choice.

    • Already familiar with these figures. Kavalerchik, who is also an author of JOSMS aricles and considered more transparent – still I have a few reservations with his figures. Krivosheev’s figures are operational losses, on the other hand, Ilenkov delivers an upper boundry of all war related (military personnel + attache) fatalities which is comparable to Overmans methodology. The ratio between irrecoverables to the impact on demography by losses of military personnel is around 70% of the total value, this is more or less true for all belligerents. The Germans suffered around 3-3,5 million irrecoverables to 5 million est. deaths (as the upper boundary), comparable to the Soviet figures of 9 million and 13,5 million respectively (or take the US e.g. 290,000 battle deaths to 400,000 war related deaths, if I recall correctly). 3,500,000/5,000,000 = 0,7 ; 9,000,000/13,500,000 ~ 0,67 ; 290,000/400,000 = 0,73

      This would indicate that about half of the total deaths were military personnel and not as usually claimed, predominantly and overwhelmingly civilians.

Leave a Reply

Your email address will not be published. Required fields are marked *