Maps of Chickamauga The Battle Then and Now 
Order of Battle Methodology How did I come up with the numbers on the Order of Battle pages, and what makes them “better” than those previously published by others? That is the heart of this page. First, breathe a deep breath and take the following to heart: We will never know exactly just how many men the two armies took into battle. There will always be that annoying gap in the records that prevents us from completing the makeup of a particular brigade or division, and thus the army as a whole. So, no system is perfect. There, I said it. So what do we do when we want an accurate picture of how many soldiers were present? We do the best we can. For example, while in Livermore’s classic Numbers and Losses in the Civil War in America he uses the numbers from the September 10^{th} returns to get the strength of the Union Army of the Cumberland, I have instead gone through the reports of each individual unit in the Official Records and taken the strength reported for the evening of the 18^{th} or morning of the 19^{th}. The same hold true for the Confederate Army of the Tennessee. Instead of taking the numbers for the last complete returns on August 20^{th}, I have gone through each individual unit and compiled the numbers reported for September 18^{th} and 19^{th}. I think that makes for a more accurate picture. The truly confusing and daunting task is matching up the myriad different methods of reporting the strength of Civil War units. Each army used similar, but slightly different methods to count and tally the number of soldiers present on the battlefield. They grouped them into several different categories including; everybody present in the entire army, everyone present and able to fight (but not necessarily armed), and just those actually in line and shooting (but may or may not count the officers). That’s just to name three categories. The trick is to find the numbers in the different categories, and if you can’t find them, make the best guess you can. That doesn’t mean I made up the numbers, far from it. There are proven mathematical formulas that have been researched that can give a relatively accurate picture of those missing numbers.
Aggregate Present (AP) this number represents everyone who is present with the army at the time in question. It counts everybody including teamsters, quartermasters, musicians, cooks, clerks, aides, couriers, and of course, the soldiers on the line themselves. [I have not included this number in the current Order of Battle pages on this website, but will add it at some point in the future]
Present for Duty (PFD) this number includes everybody that actually marched into combat. It normally includes a few noncombatants such as regimental musicians and stretcher bearers. While they may not be shouldering a musket, they are just as essential to the function of the unit as those standing on the line.
Present for Duty, Equipped (PFDE) this number counts just the soldiers on the firing line. It is predominantly a term used in the Union army. It normally includes both officers and enlisted men.
Effective Strength (ES) like Present for Duty, Equipped, it counts the soldiers on the line actually firing a weapon. However, it is normally a Confederate term, and usually does not include officers.
The number that does the best at comparing the relative strengths of the two opposing armies is their Present for Duty returns. For the Union army this is fairly simple as it is one of the main categories listed on most returns. Still, it is not perfect. Sometimes the PFDE returns are larger than the PFD numbers! I have no explanation for that other than inaccurate book keeping. Perhaps later research will solve that mystery. When it is not listed, as it is not on most of the smaller unit returns for Sept. 19^{th}, I have applied a formula based on each Army Corps. For example, the TwentyFirst Corps consistently had a PFDE strength that was 97% of its PFD numbers. The other Corps (when the Twentieth Corps PFDE strength wasn’t larger than its PFD strength) typically had that number at about 99.6%. The process was similar for the Confederate Army. According to Livermore and Newton , you can generally take the ES of a unit and divide by 93% to add the officers. Then, according to research by Newton, you can multiply by 1.065 to add the essential noncombatants necessary to achieve a PFD tally.
Infantry & Artillery (ES/.93) x 1.065 = PFD Cavalry (ES/.85) x 1.065 = PFD
These formulas work better the larger the unit is. When you get down to regimental strength there is much more room for variation. However, faced with a lack of numbers in many cases, it is usually better than nothing, and can help fill in critical gaps. So know you know where I got the numbers. However, what about when there are numbers for the same unit from multiple sources? Typically, I have built the Order of Battle from the bottom up. If a brigade has regimental returns from September 10^{th}, but an individual regiment has given numbers from Sept. 19^{th}, I have gone with the number for the 19^{th} for that one unit. If I have the strength of 3 out of 4 regiments in a brigade, and a record of the brigade strength, then simple subtraction can give the probable strength of the unknown unit. In extreme cases where no record for a unit exists, but I need a number to fill out the strength for its parent brigade or division, I have entered a number based on the average strength of similar numbers. For example, if I need to know the strength of a battery, I have taken the average strength of known batteries in the army and used that number. In all cases, I have differentiated by color which numbers are taken directly from primary and secondary sources, and which numbers are based on formulas.
Black numbers in black were taken directly from primary and secondary sources. If you simply want to know what is available from the records, you can use these numbers and disregard the others. Red numbers in red are derived from the formulas explained above. They fill critical gaps, but are not in the records themselves. Still, if you recall, many of the numbers and battle strengths historians have used for decades are based on just these types of formulas. Blue numbers in blue are a best guess. Examples include subtracting the strength of three regiments in a brigade from its known total strength to get the strength of an unknown fourth. It can also include guessing at the number of staff members a unit may have based on the relative strength of other staffs for similar units.
Finally, I am very open to constructive criticism on my methodology. If you feel I have made an error somewhere, please let me know. The only way to improve this site and the research behind it is through feedback and more research. I appreciate any constructive feedback. It’s the only way to grow. 

All images, text, and video © Historic Imagination Optimized for viewing at 1024 x 768 or better 