On Feb 8 the German FAZ reported that the actual death toll of the conflict in Ukraine is ten times higher than the official figures suggest. According to their sources (‘Sicherheitskreise’ – military / intelligence) 50’000 fighters and civilians have died since April 2014. The UN estimates 5’358 casualties, Poroshenko recently said that 5’638 died.
Some remarks on this:
1. There is always a gap between different estimations. There exist different methods to count killed fighters and civilians. On the other hand information on killed soldiers and dead civilians is a strategic asset since information is part of warfare and conflict management… but this variance is bizarre.
2. Maybe the figures of the FAZ report are blown up, maybe the report was planted in order to soften Western opposition to the latest German / French emergency-talks in Moscow and Minsk. That was my first thought after I read about it. Then I did some calculations…
3. Proportionality  between civil casualties and population size:
For the Afghan conflict in the 80s it was estimated that 0.011% of the civil population was killed per month . In Bosnia the ‘genocide-rate’ was estimated experimentally 0.2% per week . Let’s assume that in Ukraine civilian casualties (fighters excluded) are five or ten times lower than in Afghanistan (0.002% – 0.001% per month).
Around 5.2 millions are immediately affected by the conflict. I do another more conservative estimate with 4.3 millions, the population of Donetsk Oblast in 2013. Those parameters from Afghanistan and Bosnia are estimations, they are not necessarily consensus among historians. No need to mention that Afghanistan is not ideal to get a statistical baseline for an estimation on Ukraine. Data on Georgia or Chechnya would be better for that purpose.
4. Proportionality between casualties and refugees:
In Syria the amount of refugees is estimated around twenty times higher than the amount of dead (roughly, data from wikipedia). Maybe the war in Ukraine is five to ten times less deadly than the war in Syria.
5. If this guessing makes any sense, the figures of the FAZ report seem too pessimistic. I think the estimations I did are conservative (optimistic). If this is the case, the UN figures are at the very low end of the range. More crucial than the numbers is the definition of ‘civilian casualties’. Does it include only direct victims of the armed conflict (shelling of urban areas, bus stops etc.) or do dead seniors, sick ones, wounded fighters, children who died due to lack of infrastructure count as well?
6. The high ‘genocide-rate’ of Bosnia (0.2% of pop per week) was a result of encircling and shelling cities, temporary local superiorities, volatility, taking organs not prisoners…
 Please note: I did this estimations on Sunday afternoon with the stuff I had at hand. It’s a priori and not precise, rather statistical voodoo than science… (A proper analysis on this (including battle deaths) would take at least three or four weeks. My email is in the about section in case someone likes to fund such a project.)
 Allan/Stahel, 1983, Tribal Guerrilla Warfare Against a Colonial Power: Analyzing the War in Afghanistan, 602ff.
 Jerman et al., 1999, Simulating Future Wars, 126. In: König et al. (eds), Konflikte und Kriege – Simulationstechnik und Spieltheorie.