To satisfy my own curiosity I have taken a stab at estimating the fatalities that directly resulted from guerrilla warfare in Missouri in 1862 using Bruce Nichols' "Guerrilla Warfare in Civil War Missouri, 1862." This is just a first attempt, partly with the idea of getting a feel for how to categorize and arrange the data. I used this particular book because of its orderly chronological and county-by-county approach. I resisted the urge to supplement with other sources partly because of time and partly because I was hoping to keep the basis cleaner. (I wouldn't be surprised if Bruce has accumulated significant new material for the 1862 period while working on subsequent volumes.)
My current approach uses all of the actions in the work, rather than trying to use only those that were primarily guerrilla, primarily recruiter, or primarily Confederate/MSG participation. So "regular" engagements are included. This seems necessary since as Bruce notes, Missouri warfare in 1862 was all essentially behind the lines for the Southern forces; and in many actions there was a mix of regulars, guerrillas, recruits, and recruiters. It also includes the bloody border raids into Kansas and some limited pursuit into Arkansas as well. The KS numbers bump up the civilian tally noticeably, even in 1862.
I attempted to consistently classify the fatalities by type of action as well as the role of said persons. Roles came down to: civilians, slaves (1!), Union soldiers/militia/teamsters, Southern soldiers/recruits/recruiters, and guerrillas/Jayhawkers/bushwhackers. Of course many times even these general lines are blurry or the forces and fatalities are mixed. I also kept track of captured/paroled men in the actions as well, but only broke this into Union vs. Southern. Some deaths defy simple classification, often because details are lacking. Similarly, numbers are very approximate when "several", a "couple", or other vague numbers were reported. I used the lowest estimate where possible. This skews some numbers to the low side, slightly offsetting some of the known bias of overstating the other guy's casualties.
I'm sure that if I go back through this again, my classifications will change for some early events...and I'll find errors and inconsistencies in my first pass. I can already think of a few I should reconsider or add.
Okay some numbers for 1862.
Civilians killed: 153 (at least one was a female)
Slaves killed: 1
Union soldiers/militia/teamsters/scouts claimed killed: 412
Guerrillas/bushwhackers/Jayhawkers claimed killed: 491
Confederate/MSG soldiers/recruits/recruiters killed: 962
TOTAL: 2,019
This is larger than I anticipated (and works out to 9.6 killed per page of main text.) Likely the murders of civilians are the least documented and were likely greater since they were not subject to systematic headcounts and reporting as were regular forces. With the nature of the majority of the source material, Union losses from action are probably under-reported and Southern losses over-reported, at least on a relative basis.
The comparison of captured prisoners is also interesting. These values are less likely to be as inflated for the Union sources for the men they captured, since they were usually small numbers of men actually in custody rather than vague estimates of how many might have been killed (and subject to not having all the bodies on the field.) The Southerners mostly paroled their captives out of necessity. As a result there is likely serious undercounting.
I came up with 791 Northern combatants in enemy hands, and 1,901 Southern combatants in enemy hands during the period.
I haven't decided whether or not I'm going to attempt this for the remaining volumes, or if I'm going to go back and clean up my first pass at 1862 to improve my methodology.
Again, this is just my first estimate and it likely has a number of errors of my own. While I used Bruce's book to try to get a handle on this, interpretation errors are mine, not his.