I need to back up a minute here.
Though my Bachelor's is in History, my Master's is in Business and I am, by trade, a statistician. I spend the day working my way through performance statistics for my organization or those reported by other organizations to try to paint a clearly understandable picture of trends. So, for me, looking at performance statistics from WWII is an entertaining application of the everyday. In my particular case, I tend to look at the data as presented by the service in question, usually from official post-war compilations, and sometimes from the actual reports from which those compilations are drawn. I find the presentation of the data without the filter of intervening scholarship to be of historical interest in and of itself. Further, I specialize in USN statistics.
There is a basic concept that must be applied when looking at statistics and making statements about statistics, in order to draw a conclusion (at that is the purpose of the study of statistics, to draw conclusions or inferences from available data) from the mounds of available information, you must apply some pretty rigorous rules with regard to the reliability of the data. In the case of air-to-air credits it is best follow a pretty straight forward convention, that is, credits should be verifiable, witnessed by another or captured on film.
In USN service, much of the early discrepancies in the war between claims, probables, and actual victories, in my opinion and from discussions over the years with folks on the scene, resulted from a lack of an institutionalized air intelligence assessment program. Early intelligence officers went so far as to warn pilots to expect, since it was "well known" that the Japanese really did not have any decent fighter planes, to see Me-109's as adversaries. Small wonder that even Lieutenant Commander Paul Ramsey, CO of VF-2 at Coral Sea, reported even shooting down an Me-109 in that battle. Lieutenant Noel Gayler also reported seeing them, mentioning his observation in an interview conducted at BuAir on 17 Jun 42 while discussion the Lexington Air Group attack on the Shokaku for which he flew strike escort:
"After about two minutes in this clear space, we were jumped by fighters from the Jap carriers. I should say there were probably four or five Jap fighters. At first they were all air–cooled type Zeros or some modification – it was the first I'd seen of them. Then, after a minute or two of fighting, more fighters appeared on the scene that were liquid–cooled jobs that looked very similar to the ME–109F. I can't say definitely what they were, but they were planes similar to them."
A month later, at Midway, Ensign Albert Earnest of VT-8 reported being attacked by Japanese fighters that appeared to be Me-109s. He once told me, with a smile, that he had been briefed by the intelligence types to expect to see 109's so that was what he saw.
It would appear that this sort of identification error in the early days stemmed from an exposure to intelligence briefings prior to deployment. Yorktown's VF-42 pilots never received any briefing as to what to expect to see the Japanese flying and never reported seeing 109's, even though they were in the thick of the early fighting up through Midway.
The fighter action over the Japanese carriers at Midway is an example of the confusion one encounters in air-to-air combat. The action surrounding Lieutenant Commander John "Jimmy" Thach's six VF-3 F4F's must have been fast and furious for the surviving Japanese CAP pilots later reported they had encountered some eighteen Grumman type fighters. The Hiryu CAP reported downing nine Grummans, the Akagi CAP reported shooting down two and damaging one. Overall though, the Japanese CAP lost some 14 aircraft out of 41 engaged, 11 shot down and 3 ditched (two from battle damage). Certainly, though the only US fighters engaged over the Japanese fleet, not all the Japanese CAP losses could be credited to Thach's VF-3 escort. One of the escort pilots, Machinist Tom Cheek for one, saw one Zero go down from defensive fire from VT-3. At least one Zero is known to be lost from friendly fire. The Japanese CAP losses correlate with the VF-3 claims enough to conclude that the claims and credits for Thach's pilots were pretty much on the mark, six Zeros shot down and 2 damaged against one F4F shot down.
There is little doubt that a goodly percentage of probables claimed in air-to-air combat resulted from the same aircraft being shot at by several pilots over a brief period of time, yet in separate, discrete actions. Discrepancies between claims and actuals are not hard to understand in an aerial melee. Shooting and trying to keep from being shot ... a pilot fires on an opponent and he falls away into a cloud . . . "got that one . . . no time to watch him go in!" Meanwhile, a section fighting 3000 feet below latches on to the same opponent as he recovers and pops out of the cloud and the leader hoses him down . . . he falls away smoking (could that be exhaust?) again and that's one more claim. Someone else gives him a snap burst or two about 30 seconds later and there's another claim . . . "I'm sure I nailed that SOB, I saw smoke!" . . . (more exhaust? crappy fuel? oil leak?) . . . meanwhile, the poor sap's guns have been jammed all this time and he's just trying to get the hell out of there. Finally, he manages to clear the area and is headed for home when he's bounced by a roving section that makes a run at him as he is landing. He goes in with a great blast of flame and smoke and he is well and truly shot down. So, who gets the credit?
In the Navy and Marine Corps the advent of the Air Combat Intelligence Officer program in late 1942 seems to have been substantially controlled, over time, over-claiming by shifting the questionable (questionable meaning no witnesses, no film) claims to the probable category. The Navy was able to keep some pretty meticulous records, especially in the last 20 months of the war and more especially in the last 8 months. The close questioning, and later examination of gun camera films for verification tended to eliminate, in my opinion, a substantial portion of the error . . . if nothing else moving a number of shoot down claims to the damaged or probable category as opposed to confirmed. So, rather than some enthusiastic squadron yeoman recording the experiences of a keyed up pilot returning from a mission there were some rather steely-eyed and trained to be skeptical folks sitting on the other side of the table. The USN was very, very hot on accurate reporting and the ACIOs were the tip of the spear to achieve the highest level of accuracy possible. I know ACIOs from the 1944-1945 period who served in CV based squadrons who wouldn't award a claim if it wasn't on film . . . no picture, no credit. The same applied for credit for the destruction of ground targets, especially defuelled aircraft. By the end of the war it was not uncommon, in the absence of some spectacular explosion or obvious fire, for ACIOs to be the guys who decided whether a gunnery run on a parked airplane resulted in its destruction or just damage. Gun cameras eventually became the great leveler of claims for the USN.
The ACIOs (a disproportionate number of whom were lawyers and stockbrokers) were trained to question and evaluate with cool detachment. Pilot accounts were made in one-on-one sessions with a non-aviator, skeptical by nature and training, interrogator. It was the ACIOs who decided who got credit for what. The presence of these party poopers tended to keep things in line. With a knowledge of the program, and long association with a few of the ACIOs from the period, and discussions with them on this exact subject, I'd say, especially from 1943 on, USN/USMC credits were probably much closer to the reality than anyone else's.
The point, here, is that the USN, and I believe if you were to look into it, the USAAF were by far more interested in reporting accurate numbers than making an individual, squadron, or command look good. This is an intelligence/order of battle deal where it is important to be able to draw a picture of an enemy force, usually based on an initial intelligence estimate and compare that to what you believe your forces are actually doing to them. It does not pay to overestimate the damage you believe you are inflicting. The most effective way to do this is to establish the rules and stick to them . . . no witnesses, no film, no credit. The USN went to great lengths to enforce such rules and make sure everyone understood.
All that being said, as I read some of these posts, some are complaining that someone was not credited with actual victories for probables. Well, I'd ask where do you want to draw the line . . . just this one guy? 1LT William D. Sells, AC, USA (AO00337325) was killed when his plane crashed on landing due to battle damage on 12 April 1943. He was credited with two victories, one on 16 December 1942 and the other on 28 March 1943. Any other probables for which he may have submitted claims were and are just that, probables, and probables means maybe, not verifiable. But because someone finds a report that addresses these probables we should give him credit now? And what is the basis for that credit? Has some verification come to light or is this just wishful thinking.
Is the 5730 number some wish to credit P-38 drivers in the Pacific a combination of actual credited victories and their probables? Frankly that's what it sounds like. If so, well, what about everyone else? By that standard my father goes from a six victory ace to a seven . . . not that he'd go along with it, and I can tell you he wouldn't. So do we credit all the probables by fighter type as well? You realize that that would mean everyone else's scores would go up proportionally, right? Where would that leave the P-38 in the overall ranking? Why don't we just add the destroyed on the ground as well? Where do you stop? Enthusiasm for the subject is fine. Having a favorite aircraft is fine. But all that needs to be tempered with a healthy appreciation of reality. I've been searching with diligence; I can't find a single source that credits P-38s in the Pacific with 5730 credited victories.
Regards,
Rich