Estimating Numbers (Part I)
Trice, et al.,
Your assessment pretty much agrees with what I have too. Just FWIW, I'm reposting something I had posted on another forum; might find it interesting (sorry for the length of the post) :
Excerpt from:
"The Secret War for the Union"
Edwin C. Fishel
Houghton Mifflin
© 1996
"A consensus of historical belief holds Allan Pinkerton's severe overestimating of Confederate numbers largely responsible for the extreme cautiousness that brought about General McClellan's failures - failure that delayed the successful prosecution of the war by a year or more. According to this view, Pinkerton, though a highly successful detective, was incompetent at the business of military intelligence, and McClellan, brilliant organizer and administrator though he was, was thoroughly deceived by Pinkerton's inflated estimates.
Neither half of this widely shared judgment is correct. As already seen Pinkerton was successful in spy-chasing (counterintelligence). In positive intelligence his early performance was weak, but before the end of his service he was keeping track of the composition of the opposing army with surprising accuracy. As for McClellan, he could not have been deceived by Pinkerton's overestimates, for he was party to them - the dominant party, in fact.
The first of the famous exaggerations of Confederate strength was the 100,000 figure McClellan gave in his August 8 letter than General Scott branded a false alarm. Pinkerton probably did not contribute to that estimate, for on the day it was written he had just begun setting up shop, with only three of his detectives on hand. And for some weeks or months he was preoccupied with tracking the activities of spies and would-be spies in the secessionist population of Washington.
The spies that McClellan said were among the sources of his 100,000 estimate could have been members of the short-lived bureau headed by William C. Parsons; but that group was disbanding at this time, and its positive-intelligence activity is not known to have extended beyond the vicinity of Washington and Alexandria. Thus it is more likely that the spies McClellan referred to were self-appointed amateurs. However, the likeliest single source was not a spy but a Confederate deserter, a Kentuckian named Edward B. McMurdy. McMurdy claimed that he had been impressed into the Confederate service at New Orleans, had been in the Bull Run battle, and had made his way to Washington, via Harpers Ferry, to seek an army commission. According to his later account, McClellan took him to see General Scott, whom they found with Lincoln and Secretary Seward.
Those officials, McMurdy said, "thought my views of the resources of the Rebels to be extravagant." After the interview McMurdy was a guest in McClellan's house, where the general "told me I had saved Washington City and in all probability the very existence of the Government." It may be that until McMurdy appeared McClellan's view of Confederate numbers had not reached so alarming a level.
McClellan kept the estimates flowing. In letters to his wife he gave the enemy "3 or 4 times my force"; then 150,000 to his own 55,000; after new troop arrivals at Washington; the enemy still had, he said, double his own strength. On September 8, in a letter to Secretary Cameron, he lowered the enemy's estimated total to 130,000, against which he could oppose only 70,000. Five days later, again writing to Cameron, he revised the 130,000 figure to 170,000, and his own total to 81,000. Although the Confederates' Manassas army during these weeks received a new brigade from Georgia and half a dozen regiments from other states, it still numbered only about 35,000. And though other substantial accretions were to follow, McClellan's estimate of enemy strength grew at an even faster rate.
McClellan claimed that in issuing these estimates he was giving figures below the "real strength" of the enemy, while taking "the reverse course as to our own [strength]." His superiors would have wondered why he did not give the figures he believed "real," instead of some other figures. Predicting an early advance by the enemy, he urged that the entire regular army except for artillerists, and all volunteer troops from other theaters of war, be placed in his own army. So extreme a proposal indicates that he really believed, as McMurdy claimed, that "the very existence of the Government" was threatened.
With those communications McClellan's preoccupation with troop counts and presumed enemy advantages is in full view. The 100,000 estimate he gave Scott, after only two weeks in command, committed him to six-figure estimates for a Confederate army that in four years of war would never reach that level. When Scott pointed out that McClellan had thus placed himself on record, the old general was implying that that commitment might embarrass him in the future. But McClellan proved to be impervious to such embarrassment.
The proposal to strip other theaters for reinforcements, relying on the Federals' regular artillery to maintain a defensive stance there, indicates that McClellan was crediting the Confederates with having so many men in uniform that they could place a huge force in northern Virginia without seriously depleting their strength elsewhere. Although the South had scarcely a third of the white population of the North, it had mobilized a far greater proportion of its men of military age. This fact repeatedly came to McClellan's attention when Pinkerton reported on his interrogations of Northerners and foreigners who were in the South at the outbreak of the war, had been forced into Rebel regiments, and deserted to the Federals.
How McClellan was deriving the estimates that appear in these letters is, except for the August 8 letter to Scott, unknown and unknowable. No strength reports traceable to Pinkerton appear in the records of these weeks; if bits and pieces of information about the Confederate armies were diverting his attention from the search for enemy spies, he may have delivered them to McClellan orally. Apparently McClellan already had begun the practice, quite evident in later months, of reporting intelligence from sources independent of Pinkerton. He did not identify these; he also avoided naming Pinkerton, and, so far as records reveal, he withheld Pinkerton's reports from Lincoln and his military superiors. Such was the obscurity with which he surrounded his estimating of enemy numbers that we are led to assume that some of these figures were inventions, made by guessing, on the basis of what the Confederates were believed to have had last week, how many they probably had this week.
When Pinkerton finally accumulated enough enemy information to produce a strength estimate, it seriously overrated Confederate numbers, but not nearly enough to support the 170,000 that was McClellan's latest estimate. Pinkerton's report, written October 4, gave the opposing army 98,400, more than double its true strength, which was somewhere between 40,000 and 45,000 at that time. Pinkerton made no reference to the fact that this estimate flew in the face of McClellan's much higher figures. He may have simply been avoiding an argument with his chief, but more probably he was unaware of what McClellan had been telling his superiors about enemy numbers." [End quote]
There's more of this (this is only the first couple paragraphs of the chapter). Two things stuck out to me: (1) the text seems to be implying that McClellan's inflated numbers might have resulted from him being unwilling to detach himself from a statement of troop strength he had gone 'on record' for (i.e., if he was wrong about that, he could easily be wrong about any other number/modification he made subsequent to that); and (2) McClellan's method of predicting troop strength is unknown and unknowable (Fishel - and perhaps, by extension, any other historian wasn't able to find anything on what sort of 'logic' McClellan might have used). Additionally, Fishel suggests that this business of exaggerating troop strength was largely McClellan's initiative and less so Pinkerton's.
A couple of observations for now (based on some of the subsequent material):
(1) Even though Fishel intimates that Pinkerton and McClellan were somewhat in collusion on this (more so on McClellan's part) - Pinkerton's number (98,400 provided in early October '61) was far short of McClellan's 150,000 (and even as high as 170,000 provided to Cameron in September '61). Notwithstanding that Pinkerton's number was also grossly overstated (at 98K)…Pinkerton had applied a liberality based on 'assuming the worst' (which is not necessarily an unprecedented - nor unwarranted - practice). Fishel states that McClellan didn't want to include Pinkerton's numbers in his final reports (to Lincoln) because they failed to match McClellan's: "But a stronger reason for withholding, at least at this stage, was that Pinkerton's generous estimate still fell far short of the figures McClellan had been claiming. Pinkerton's 98,400 figure on October 4 should at least have reined McClellan in from his own hard-charging estimating, but it did not [ibid]." Fishel doesn't come right out and say it directly, but one gets the impression that McClellan doesn't want to present any kind of significant discrepancy in intelligence reports; since McClellan had started with a six-figure number - and went 'on record' with it, McClellan was unwilling (for whatever reason) to lower it.
(2) Fishel states that McClellan would explain that his 150,000 figure (and other variations on that number) represented the entire force (in Virginia) of the Confederates contrasted to what McClellan referred to as his maneuverable force (subtracting everyone on guard duty, etc.) of 76,000. Fishel doesn't understand why McClellan didn't subtract the same category populations from the Confederate force (i.e., those sick, those guarding, etc., not available for combat).
(3) Fishel provides the following to explain where McClellan might have gotten these numbers:
"All his sources agreed on the 150,000 figure, he [McClellan] declared. In this case we finally have a clue as to how he acquired his six-figure estimates. Frank Ellis, at this stage not yet considered to be working for the Confederates, had just come in from Richmond, reporting 80,000 at Manassas-Centerville and 20,000 on the lower Potomac. Ellis gave as his source Jefferson Davis and the three cabinet members with whom he said he had dined. (McClellan chose to ignore the certainty that the Confederate leaders would not have entrusted Ellis with the truth about General Johnston's numbers). Another recent report, arriving at third hand from a Unionist citizen of Fairfax County, gave Johnston 100,000 at Manassas-Centerville and 30,000 at Leesburg. By taking the higher (but less reliable) of the two Manassas-Centerville figures, the 20,000 from downriver, and the grotesquely exaggerated 30,000 for Leesburg - thus straining to get to the highest possible total - McClellan could have produced that round 150,000. Another possibility of course is that the figure was simply a shot in the dark.
Part of the draft of McClellan's paper is in the handwriting of Edwin M. Stanton, the future secretary of war, who was at that time practicing law in Washington and serving McClellan as a confidant. Stanton's professional experience would have told him that McClellan's claim of unanimity among a variety of sources was unlikely to be true; he also may have had his doubts about the validity of the 150,000 total. But that figure satisfied his lawyerlike proclivity for putting the client's case in the strongest possible terms." [ibid]
(4) Fishel goes on to show various (possible) contributing factors such as logically poor arithmetic (reducing a quantity by 1/15 rather than 15%), poor statistical skills for the time period, and other such factors.
(5) Dovetailing off of #4, to my knowledge, there was no 'intelligence agency/bureau' in existence in the Army or government at large at any time prior to the Civil War. To this end, there's no Jomini or Clausewitz equivalent on intelligence and intelligence gathering (protocols, procedures, training, lessons-learned)….nothing. From what I can gather Pinkerton's detective skill was respectable; but intelligence gathering is something of a different animal.
CC