Background

On a blog I follow (at this link), the question about tank losses in CRUSADER was raised. It’s one of those that seems easy, until you dig into it. a bit more. Since I have done a bit of the digging, here’s my view, by the armies involved. The below is from memory, and serves to illustrate the problem, not to provide an answer.

Define ‘Loss’

A few things need to be considered. First, what is considered a ‘loss’ differs at the tactical and operational scale. Tactically, a loss is a tank that is no longer able to participate in battle. This includes damaged but repairable and technical breakdowns, as well as destroyed and captured intact tanks. At the operational level, the first two categories are only losses if the damaged/broken down tanks cannot make it to the workshops, or if the workshops with them in are lost to the enemy. Both of the latter cases often, but not always, happened to the Germans, and probably the Italians. So over an operation lasting weeks, a tank could be lost more than once, if it was damaged, brought to the workshop, repaired, and returned to fight another day, and be lost again. Operationally, the easiest way to look at this is to pick a start date, check the tank inventory, add any known arrivals during the period of the operation, pick an end date, check the inventory, and do the maths. It’s more difficult in reality but still straightforward, if you have all the information.

The use of the numbers is of course completely different. To the commanders on the field, tactical losses mattered, and the reason for them wasn’t necessarily that important. A tank that’s gone is gone. This affects the ability to conduct operations, in some cases severely. For example, within four days of starting the counteroffensive in January, the Germans lost almost half their tank force, even though their written off tank numbers are miniscule (that’s a recurring theme across all theatres they fought in, by the way). 

For historians on the other hand, the operational losses are what matters, since they allow the researcher to evaluate the battle performance in relative terms. It is also a great topic of debate to make the Germans look better than they were in terms of battlefield performance, as in ‘yes they lost, but look at how much it took to take them down…’. More seriously though, operational loss numbers were used to inform high-level planning, so in the case of CRUSADER, the very high British losses drove considerations of the required numbers to be able to attack again.

Indian Soldiers examine an abandoned Italian M13/40 tank of Ariete Division in Benghazi, end of December 1941. IWM 

Estimating Losses

So with all this said, here is my view of tank losses in CRUSADER.

1) The Germans

This is the most straightforward of the bunch. We have the starting numbers, we have daily tank states through to 30 January (so all that matters), and we know how many tanks arrived as replacements. So anyone who has done a fair amount of research can feel confident not just about how many tanks the Germans lost, but also what their daily tank strength was. There were few repair returns during the battle, and most tanks that went to the collection points or workshops were simply lost when these were overrun in due course. The Germans received about 100 replacement tanks in December and January, and fielded about 100 tanks in mid-January.

My estimate therefore is that the Germans lost all of their 255 runners that they had at the start of the battle, and the total loss figure could be a bit higher once we account for returns from workshops. In addition the Germans lost another 45 tanks sunk in the Gulf of Taranto. The German official number is 220 tanks lost, or 85%, excluding those sunk, but I would put the total at about 300 including those sunk.

2) The Italians

Here matters get more difficult. The easy question is: how many light tanks did the Italians lose (they fielded a good number of CV light ‘tanks’ (really glorified MG carriers, some equipped with flamethrowers). The answer is: all of them (about 180 or so, I think). The Mediums (all M13/40 during CRUSADER) are where we have conflicting information. What we know is:

Where things get hazy is how many mediums were held with a rear unit in Agedabia, but it was probably low teens, up to 20. Now… the official Italian history claims that 63 were lost. I don’t believe that for one second. In my view, almost all of the Italian mediums with Ariete were lost. The reason for this is that after Ariete reaches Agedabia, it has only about 20 runners left, according to its war diary, but at this stage it would have been reinforced by the training tanks, and possibly the 24 tanks arriving at the end of November. Returns from workshops are unknown. By mid-January Ariete is fielding 80 mediums, and it had received about 80 reinforcements.

So if someone asked me, I would peg Italian medium losses at over 130 tanks, and consider that a low-ball estimate. In addition the Italians lost another 52 tanks sunk in the Gulf of Taranto. So about 200 tanks lost seems not unreasonable.

This would make combined Axis losses of about 500 tanks in total.

3) The British

Here things get far more difficult. We know the starting state, we know the end state, and we know how many were held in initial reserve. What we don’t know is how many were added on top of the initial reserve from convoy arrivals in Egypt, and how many of those tanks available in February 1942 are due to returned repairs and or convoy arrivals. We also don’t have consistent numbers on a daily basis.

An additional problem is that the same tank may have been lost more than once. For example, 35 M3 tanks are lost when the HQ of 4 Armoured Brigade is overrun. At least 8 of these are recaptured by the New Zealanders a few days later. What happened to these then is anyone’s guess. There are numbers for tanks lost as of 9 January, which come to about 800, and to which another about 150 need to be added for the losses of 2 Armoured Brigade in January. This is just about 100 tanks short of the starting state on 17 November, which was 1,038 of all medium types. On 8 February, the tank returns reported that 1,123 tanks were either repairable or awaiting evacuation, destroyed, or fit/unfit in Ordnance Workshops in the Delta, which means that total losses would have been about 100 higher than the starting date.

Total

So, my rough estimation is that compared to a total of about 1,450 tanks at the start of the battle, almost 1,500 tanks or thereabouts were lost, ignoring light tanks, MG carriers and armoured cars. Of the losses, about 2/3rd were lost by the British, and the remainder by the Axis, who also lost about 100 tanks at sea.

All forces lost almost the totality of their tank numbers from the start of the battle, if not more. These tank losses, in particular combined with the comparatively low personnel losses, make CRUSADER a fairly extraordinary operation, and one of the larger tank battles of the war.