Upvote:3
Estimates do change over time. There is a typical evolution to this modern-day conundrum:
Phase 1: The War is Still Ongoing
Almost always, opposing losses are over-stated and own losses are under-stated. Over-stating opposing losses is practically guaranteed, driven by battlefield confusion and a desire to prop up the home-front morale. Likewise, own losses may be not yet compiled and also under-stated (or not stated) due to "operational security," for both reasons of hiding weakness from the enemy and also propping up home-front morale.
Sometimes one or more sides can get carried away in over-stating their opponent's losses - it becomes a pandora's box that can't be easily shut after being opened, due to the political momentum.
Numbers published during this phase are also "anchors," in that any future adjustments or research has to relate to that initial anchor, no matter how far off that initial claim was.
Ukraine is currently in this stage. The numbers that have come out are not reliable and are incomplete, and neither side wants to be honest at this point because of operational security and home-front morale.
Phase 2: The War is Over, Rational Study Begins
Now historians take over. Initial claims get scrutiny as records are released. If a combatant had good record-keeping (and those records survived), these are researched and published. Statisticians may also give it a go with some math models, which can be refined with more inputs and correlation to known records. It can also be a matter of counting graves.
Some Examples
One easy-to-study example is the Axis surrender in Tunisia in May of 1943, and specifically how many Axis prisoners were taken. Liddell Hart [1] compiled these numbers from different sources:
Liddell Hart tended to believe the logistics reports the most. The admin staff is likely the most aware and honest about the number of mouths they are feeding and have the least incentive to over-state or under-state those numbers. Liddell Hart goes on:
Here it is worth remark that much larger discrepancies still, between the last known German ration strength and the Allied claims about the number of prisoners captured, were manifest in the final stages of the war.
This kind of data is likely only to be found by researchers after the war, who are seeking these records and that those records can still be found.
Sometimes records that come to light later on can spark a controversy as well, if they disagree significantly with the existing narratives.
Another example is the controversial book, John Mosier's The Myth of the Great War [2]. The thesis of that book was to deduce what was really happening on the First World War battlefields, when records from the German side are used to correct the greatly inflated British claims of German losses.
[1] History of the Second World War. B.H. Liddell Hart, c 1970.
[2] The Myth of the Great War. John Mosier, c 2001.