We have the following data for the Israeli Defense Forces from their website here: IDF Fallen and Wounded in War | ATC (www.idf.il).
Killed:
Killed since the beginning of the war (7 October 2023): 639
Killed: 290 dead (fighting in the Gaza Strip from 27 October 2023 to 29 May 2024 among those “whose names were permitted to be published”)
Now, 639 – 290 = 349 killed on 7 October or shortly thereafter.
Fatalities from operational accidents: 44 (this in 22 from
two-sided shooting, 5 from “shooting anomalies” and 17 from “accidents”). Data from fatalities from operational accidents is correct as of 15 May 2024.
Now, I do not know if operational accidents are included in the war dead. I am assuming they are not, so 290 + 44 = 334 or 639 + 44 = 683
Wounded:
Wounded since the beginning of the war (this means from 7 October):
3,643
- 2,124 “easy”
- 955 “medum”
- 564 “hard”
“Casualties” (do they mean wounded? – I assume so) from the beginning of the maneuver (this means from 23 October):
1,831
- 874 “easy”
- 591 “medum”
- 366 “hard”
Injuries:
Injuries for operational accidents in the Gaza Strip
714
- Accidents: 453
- Shooting anomalies: 36
- Two-sided shooting: 57
- Road accidents: 49
- Other: 119
Okay, time for some simple math:
A. Wounded-to-killed ratios:
Overall force Wounded-to-killed ratios (not counting operational accidents): 3,643/639 = 5.33-to-1
Gaza Strip operations wounded-to-killed ratios: 1,831/290 = 6.31-to-1
7 October wounded-to-killed ratios: (3,643 – 1,831)/639-290) = 5.19-to-1
B. Accidental killed versus injures
Gaza Strip operations: 714/44 = 16.20-to-1
This is not a surprising figure, but not one that I have calculated before.
From “two-side shootings” and shooting anomalies: (57 + 30)/(22+5) = 3.20-to-1
From “two-sided shootings” (57/22) = 2.59-to-1
This are not surprising figures, being from I assume mostly direct gunfire.
C. How about friendly fire?
Percent killed by friendly fire in Gaza Strip: 22/(290 + 22) * 100 = 7.05 %
Note, the percent of expected friendly fire casualties has never been firmly established. Traditionally the figure from WWII was 1 or 2%. Many people considered these estimates low. It was clearly higher than that in Vietnam (1965-1973), but no one has assembled any systematic data. It was much higher than that in the Gulf War (1991).
Some past references:
Wounded-To-Killed Ratios – The Dupuy Institute
Also note on page 187 of War by Numbers there is a discussion of weapons effects in the 1982 Israeli-Lebanon War. The lethality figures of bullets was 0.31 and for “small arms” was 0.28. This comes out to wounded-to-killed ratios respectively of 3.23- and 3.57-to-1.
C.A.L: “Note, the percent of expected friendly fire casualties has never been firmly established. Traditionally the figure from WWII was 1 or 2%. Many people considered these estimates low. It was clearly higher than that in Vietnam (1965-1973), but no one has assembled any systematic data. It was much higher than that in the Gulf War (1991)…”
-I’ve never seen anyone point out the obvious: If the enemy inflicts fewer casualties, the percentage of friendly fire losses will be greater.
Yea, that certainly was the case with the Gulf War. It did lead to a couple of people in the 1990s looking into friendly fire cases. The most complete paper I saw on the subject was by Charles Hawkins and Eugene Visco. They have both since passed away, but I used to have a copy in my office. Not sure what file cabinet or box it is in now. I am sure it is somewhere.
OK found one of them: https://www.usni.org/magazines/proceedings/1994/june/friendly-fire-facts-myths-and-misperceptions
Also see:
https://www.airandspaceforces.com/article/1291friendly/
To quote: Mr. Hawkins, formerly a platoon leader and rifle company commander in the 101st Airborne Division, went back through casualty reports from one battalion during a four-month period of the Vietnam War in 1970. He found that more than thirteen percent of battlefield losses were due to friendly fire.
Maybe I should do a blog post on friendly fire (which isn’t all that friendly).
U.S.N.I article “Friendly Fire: Facts, Myths and Misperceptions”:
“…Third, there is no evidence to suggest that battle experience is associated with decreases in fratricide rates…”
-They may not have been a study to show a correlation between inexperience and blue-on-blue (although it would be right up your alley), but it falls into the category of common sense, and contradicted in the next paragraph:
“First-order causes of friendly fire are understood—target misidentification, coordination problems, inexperience, and discipline problems…” Besides the literal use of the word “inexperience”, target misidentification and coordination problems can be the result of inexperience.
As for American Civil War anecdotes, the 2nd Wisconsin (of the future Iron Brigade) took losses at 1st Bull Run due to their gray uniforms, and the 29th Illinois took losses in the dark the first night at Ft. Donelson.
For 10th Mountain Division anecdotes, a young Bob Dole wounded his platoon sergeant with a grenade that ricocheted off a tree in the dark.
Yea, they had not gotten very far in their research. They were looking at getting funding from the U.S. Army to do a proper study of the subject but never tagged any funding. So, it died there (simple equation: no money = no research, I am amazed at the number of people in DOD who do not grasp this).
I have never done any exploration on friendly fire, in part because I knew that Hawkins and Visco were looking into it. If they couldn’t find any funding for it, then I probably could not either. At the time (early 1990s) my focus was on the Kursk Data Base.
P.S. It is clear that unit experience probably does reduce the number of friendly fire cases. It would be probably easy to deduce from the data (not that anyone has done any systematic collection of data) that green or inexperienced units produce a higher number of friendly fire incidents. Not sure what data could be collected for analysis. It is nature of friendly fire incidents that they are often scattered and anecdotal.
Probably they were looking at assembling 30-40 case studies and then looking at the factors behind each incident (evaluate each case study) and see if there is a pattern. This is probably 2-4 man-years of effort.
Chris, if you do some friendly fire analysis, it would be useful to compare/contrast friendly fire percentage of total casualties/wounds/deaths/incidents with percentage for journalists or aid workers or resident civilians or . . . in order to test hypotheses concerning whether IDF is being more careful or less careful when it comes to its own soldiers versus other categories of people (not that factoring out situational effects such as civilians being in the position of serving as human shields would be an easy thing to do). Accidents do happen, but is there the bias that is sometimes attributed to the IDF? Is there recklessness, in some cases while not in other cases?
Right now my focus is to complete The Siege of Mariupol, then I am probably going to start on The Battle for the Donbas and possibly one other book project (Either The Battle of Tolstoye Woods or The Combat Effectiveness of Eight Divisions in Italy). That will kind of speak for the rest of 2024 and certainly a significant part of 2025.
I am probably not going to look into friendly fire.