Flying is not a risk-free activity, and military flying is even more dangerous. Since the first fatal army aircraft accident in September 1908, aviators have made great strides that have lessened, but not eliminated, both risks and accidents. The late 1940s and early 1950s were a critical time for the airmen, as the transition from prop power to jet power was neither easy nor cheap.
The premier U.S. fighter of this period was the F-86 Sabre. It is a classic: beautiful in looks, impressive in performance, and well regarded, if not loved, by its pilots and aviation enthusiasts. It was a great success in service, especially in combat, and the USAF employed more F-86s than any other jet fighter in its history. It did, however, encounter problems, accidents, and casualties.
The transformation of the Air Force from propeller-powered aircraft to jets was a challenge. Besides greater performance, jet-powered aircraft had a number of advantages. Although tricycle landing gear was novel to most pilots and posed the possibility of dragging the tail on landing, it had an overall positive affect because it eliminated ground loops and gave the pilot better visibility during ground operations.1 Torque-less jet power simplified flying. Jet engines were less complex than prop engines and propellers; for example, jets had only a throttle while propeller aircraft had throttle, mixture, and prop controls.
Initially, jet-powered fighters proved more dangerous than piston-powered fighters. The major accident rates of the prop-powered P-51 (later F-51) from World War II to August 1953 exceeded that of the jet-powered P-80 (later F-80) in only one year.2 Performance of the P-80 was higher, as were takeoff, approach, and landing speeds.3 Higher speeds not only gave the pilots less time to react but also caused the altimeter to lag, giving the pilot erroneous altitude information.4 Jet engines were new to Air Force pilots and ground crews. One of the most dangerous aspects of jet power was that it took jet engines some time to deliver increased power (to “spool up”), unlike prop engines that delivered power almost at once.5 Another peril was that jamming the throttle too rapidly forward could cause a jet engine to “flame out” (stall) and lose all power. F-86 pilots were warned to “accelerate very slowly (ten to twenty seconds from ‘IDLE’ to full open) at high altitudes in order to minimize [the] possibility of acceleration flame-out.”6 A unique problem of jet engines was their tendency to suck up debris on the ground and damage their internal parts.7
In addition to a higher accident rate, jet power brought other changes: new equipment. Because “crash helmets,” later tactfully called “flight helmets,” were slow to enter the system, pilots were forced to do without or use expedients such as football-type headgear borrowed from the Army’s tank drivers.8 The use of “g” suits was less of a problem, as the Army Air Forces (AAF) had used them during World War II to increase pilot tolerance to high-acceleration maneuvers (high gravity [“g”] forces).
Escape from a crippled jet fighter in flight was more difficult than from a prop-powered fighter because of the increased speeds, “g” forces, and altitudes. Not only was there a problem getting out of the cockpit because of increased “g” loads on the pilot and air loads on the canopy, but there was also difficulty clearing the tail. The Germans led the way in improvements with a patent for an ejection seat in 1939 and experiments with seats propelled by compressed air, springs, and gunpowder. The first emergency seat ejection was made from an He 280 in January 1943. By war’s end the Luftwaffe had fitted a number of both prop and non-prop aircraft types with ejection seats, and reportedly more than sixty pilots used the device.9 The Allies took the German invention and developed it. The first American emergency use of the ejection seat was made from an F2H-1 in 1949.10
Although the seats worked well, the pilots distrusted them at first. Some understandably feared sitting atop a seat powered by a 37 mm cannon charge, and they flew with the device deactivated. As with standard bailouts, aircraft attitude, altitude, and speed played a major role in determining results. Ejection below two thousand feet was found to be “extremely hazardous.” A study in May 1951 noted that seven of nine USAF pilots killed in forty-two ejections had ejected too low. It found that the Sabre had the lowest fatality rate of the three USAF jet fighters (F-80, F-84, and F-86).11 A 1956 study of Canadian experience reinforced these conclusions, reporting that the F-86 had the lowest fatality rate (7 percent) of the three jet types they flew.12 In over six hundred major USAF F-86 accidents prior to August 1953, pilots bailed out eighty-five times, thirteen (15 percent) with fatal results.13 This was better than the Air Force average during 1949–53, when 23 percent of USAF ejections proved lethal.
USAF experience demonstrated that it was safer to eject than to stick with the aircraft and crash land. One study found that while 11 percent of all jet fighter bailouts were fatal, 20 percent of the crash landings proved lethal. (However, pilots might have little choice if, for example, the seat malfunctioned or if the aircraft were too low to eject.) Although the Sabre had a better ejection record than its rivals, it had a worse record in crash landing: 11 percent of F-80 and F-84 crash landings were lethal, compared with 21 percent of F-86.14
The F-86’s superior ejection performance is somewhat surprising. Initially, the Sabre’s ejection seat was not recommended for exit in all circumstances. The 1948 flight manual instructed the pilot in low-speed bailouts (speeds not specified) to jettison the canopy and then roll the aircraft over, release the safety belt, and push clear of the aircraft. It went on to note that: “As seat ejection subjects the pilot to extremely high forces, this method of exit [ejection seat] should be used only when normal bail-out is impossible (at extremely high speeds or when [the] airplane is uncontrollable).”15 No wonder pilots distrusted the system and some deactivated the ejection seat! In addition, in the early F-86A models the canopy jettison handle was mounted on the lower forward cockpit panel between the pilot’s legs forcing the pilot to reach forward and extend his arm to activate the canopy jettison, not an easy task if the aircraft was violently maneuvering or experiencing heavy “g” loads. Furthermore, the ejection seat was only armed when the canopy was jettisoned. This system was changed in the F-86A-5 so that the pilot could jettison the canopy by raising the right seat handle. In mid-1951, North American proposed a system that would permit the pilot to eject through a closed canopy and fitted twenty aircraft with the system, but the USAF disapproved the change. It was not until 1953 that the seat could be fired through the canopy. (This was a last resort maneuver. Of twenty-six ejections through the canopy, 42 percent proved fatal compared with the USAF ejection fatality rate of 17 percent.)16 Another potential problem was that the F-86 ejection canopy had the tendency to “dish down,” and therefore the pilot’s manual warned the pilot to “keep head and body as low as possible.”17
A 1951 study revealed that, relative to the F-80 and F-84, the Sabre had less difficulty with the canopy. The F-86 suffered only two canopy failures in flight, compared with fifteen in the F-80 and eleven in the F-84. However, in late March 1953, Fifth Air Force F-86Fs experienced four inadvertent canopy departures.18
Capt. Richard Barr (94FS) was the first man to abandon an F-86 when he was blown out of the aircraft following a midair collision in June 1949. Two months later, 2nd Lt. Robert Farley (71FS) made the first ejection from a Sabre. With no aileron boost, one drop tank that would not jettison, and in an inverted spiral, he punched out at about 500 mph at one thousand feet. The shock of the parachute opening tore off his shoes, watch, and dog tags. Nevertheless, like Captain Barr, he survived the ordeal, albeit with major injuries.19 He was certainly lucky to have survived such a high-speed, low-altitude bailout, one that would certainly have proved fatal without an ejection seat.
Training was crucial to flying safety. During World War II, the AAF produced approximately 200,000 pilots, 40 percent of whom flew single engine aircraft.20 The availability of so many pilots, along with the drastic draw-down in the military after the end of the war, led the AAF to stop all pilot training until October 1946. Since the classes that followed were not as large as those during the war, the majority of USAF fighter pilots on active duty in the early 1950s had trained prior to 1946. In the period from July 1949 to June 1950, 62 percent of the USAF’s 1,100 jet fighter pilots had earned their wings prior to 1946. At first glance, this did not seem to be a problem: these men had earned their wings when young, and, with their flying experience, they should have been able to easily transition to jet aircraft. However, the introduction of the jet modified the conventional wisdom that experienced pilots (measured in total flying time) were safer than inexperienced ones. Although pilots with 1,000 to 2,000 total flying hours were safer in jet fighters than those with less total flying time, pilots flying their first 100 jet hours—even those who had flown many hours in prop aircraft—had the highest rate of jet accidents. Put another way, pilots with less than 300 hours total flying time had 1.5 times more accidents than the entire group while pilots with less than 100 jet hours had a rate almost 3 times that of the entire group. A later study covering January 1951 through June 1953 found that pilots with less than 150 jet flying hours accounted for 45 percent of the jet accidents.
During the period July 1949 until June 1950, the 290 pilots who earned their wings in 1949 comprised 25 percent of the fighter pilots, flew 27 percent of the jet time, and yet accounted for 37 percent of the jet accidents. Their major accident rate was more than double the USAF average.21 The class of 49C that graduated in September 1949 was a particular problem. Between 1 December 1949 and the end of April 1950, six of sixteen fatal F-80 accidents and three of eight fatal F-84 accidents, (although neither of two F-86 fatal accidents) involved graduates of 49C.22 Five of the fatal F-80 accidents were in the Far East, prompting the commander of Fifth Air Force, Gen. Earle Partridge, to restrict pilots from that class from operational flying until they were reevaluated. Partridge’s command gave these pilots a training program that included twenty hours flying in the T-6 and T-33 before they were returned to operations.23
An early 1950 report indicated that the accident problem arose from three causes: a lack of accelerated testing and development of solutions before the fighter aircraft were produced in quantity, inadequate training, and a lack of pilot indoctrination in the new aircraft.24 In support of the third conclusion, the report cited the case of the 33rd Fighter Group. It converted from flying F-84s to flying F-86s and within a five-day period had lost one pilot and one aircraft and damaged two aircraft.25 Perhaps the 1st Fighter Group, the first unit to transition to the F-86, would have been a better example. It received F-86s in February 1949 and suffered its first major accident in April. During its first quarter of operations, from April through June 1949, it suffered five major accidents, including two fatalities and three total wrecks. Five more major accidents, including two write-offs, occurred in its second quarter, and two major accidents followed during the last quarter of the year.26
The report blamed neither Air Training Command nor the tactical units for the inadequate training. It diplomatically explained, “the phrase ‘inadequate training’ must be considered to be due to several factors, chief among which are the urgency of international affairs, the reduced budget of the Air Force, and the state of the arts.”27 A later report was more blunt, noting the “inept flying on the part of pilots recently graduated from the Air Force flying schools.”28 The report indicated that new pilots were not getting enough fighter flying time; the F-51 was an unsuitable aircraft for jet fighter training; instrument and night flying training were inadequate; and the curriculum was lacking certain important elements, such as the characteristics and limitations of jet aircraft, gunnery, rocketry, and bombing. In brief, “graduates therefore are not proficient fighter pilots.”29 This report went on to criticize the lack of standardized USAF flight instruments, specifically the attitude gyro and airspeed indicator. These instruments were located in different positions on the instrument panel of the various jet fighters, and the trainees were using three different types of attitude indicators. In addition, some jet fighters and trainers were using airspeed indicators graduated in miles per hour while others were using knots per hour.30 Clearly this was a poor situation for flight training and inexperienced pilots.
Early on, the airmen realized that special training was required to transition to jet-powered aircraft and began planning such a program in January 1946. The program used P-80s, which were easy to service. Nevertheless, maintenance proved a problem because of a shortage of maintenance personnel experienced in servicing jet aircraft.31
Another major problem with USAF training was the absence of a two-seat jet trainer. New pilots flew only prop-powered aircraft in training, and from day one, the jet fighter pilot trainee had to fly solo with an instructor flying in an accompanying aircraft. This arrangement worked well when there were no problems but could prove hazardous, and sometimes fatal, if there were difficulties. In fact, sixteen pilots were involved in a major accident during their first F-86 flight. (One of these accidents was fatal and another three resulted in destroyed Sabres.)32 In mid-1947, Lockheed lengthened the fuselage of an F-80 by three feet to accommodate a second pilot in tandem. The two-seat trainer first flew in March 1948, initially designated as the F-80C; it was redesignated T-33A in May 1949 and went into service at Williams Air Force Base in June 1949. The T-bird greatly enhanced pilot training and was used to train many pilots throughout the world.33
The Korean War forced changes within the Air Force, which was already in turmoil as it transitioned from prop to jet aircraft. The war’s demand for large numbers of pilots was met by increasing the graduation rate of the training schools by lowering standards, reassigning pilots from desk jobs, and recalling reserve pilots to active duty. The result was an influx of what an Air Training Command historian called “notoriously unqualified” individuals going through jet training. Many of the recalled reservists had been out of the cockpit since the end of World War II, and a considerable number of those in the regular air force had been in administrative jobs for years.34 A retraining program in the F-80 began at Nellis Air Force Base in mid-July 1950 but was discontinued in January 1951. There is no explanation for this action, although by the end of June 1951, the USAF had two jet fighter schools, labeled “fighter-bomber/escort,” based at Luke and Nellis Air Force Bases, where students flew F-51s, F-80s, and F-84s. It was not until the end of 1951 that the USAF established a training program for the F-86 day fighter; and by 1953 there was a program for both the F-86 day fighters and fighter-bombers at Nellis.35
For all of the shortcomings and criticisms of the USAF pilot training program, one fact should be emphasized: it produced pilots far better trained than their foes. For when these men were called upon to perform in combat, they did so in a spectacular fashion. During the Korean War, Sabre pilots ran up high scores against MiG-15s, despite an at best equivalent aircraft, superior numbers of Communist fighters, and the location of the combat. The major American advantage in the air-to-air battle was the skill, training, aggressiveness, and experience of the Sabre pilots.
Like all new aircraft, the F-86 had early problems; fortunately, none were serious or long lasting. Chronologically, the first involved the landing gear. This is surprising because that system was neither new nor exotic. North American test pilot George Welch had difficulties with the gear on the fighter’s first flight in October 1947, and four of the Sabre’s first five major accidents involved the nose gear. As late as November 1950, the Air Force complained of an excessive number of accidents involving the nose gear. The problems with the system included an undersized hydraulic cylinder, weak nose gear trunnion, faulty cylinder rod, and gear vulnerability during ground-towing operations (a towing wedge was required, but sometimes was not used).36 If the pilot retracted the landing gear at too high a speed, it could slam back into the aircraft and damage the piston rod. Almost 10 percent of the F-86 major accidents through July 1953 were connected with the landing gear, two-thirds of these with the nose gear. The Air Force worked out these problems in the succeeding Sabre models. Whereas 12 percent of the F-86A major accidents involved the landing gear, this figure fell to 5 percent for the F-86E and 7 percent for the F-86F.37
The F-86 also had troubles with its instruments. An early 1950 survey noted problems with the Sabre’s airspeed indicator (reading 15 kts low at the aircraft’s high speed limitation). Non-standardization of attitude indicators created a more serious problem. In early 1951, the 33rd Fighter Interceptor Wing was using four different types of attitude indicators, and, as late as August 1951, at least three different types. The danger was that two of the types gave opposite presentations, a situation that was confusing at best, and certainly hazardous at night or in emergency or weather conditions.38
From the earliest days of flying, one of the terrors was getting into a spin. For its part, the F-86 was noted to be an “honest aircraft,” that is, it gave adequate warning of an approaching stall and potential spin.39 Early on, North American pilots performed over one hundred spins in the Sabre. George Welch, the company’s test pilot, wrote that neutralizing the controls would cause the rotation of a spin to stop within three quarters of a turn, permitting the pilot to pull out of a spin within five thousand feet.40 This performance allowed the company to claim that it was difficult to keep the fighter in a spin, perhaps a commercial overstatement. Initially, the Air Force was somewhat more cautious. The 1948 pilot’s manual prohibited spins. A few months later the revised manual gave spin recovery instructions, noting that recovery was normal and could be completed within one-quarter to one-half turn with the loss of approximately one thousand feet of altitude. An early 1954 report stated that recovery could be made in one-half to one and one-half turns with proper technique and in two to three turns with hands off.41
Despite the great admiration the pilots had (and have) for the F-86, certainly the early Sabres (“A” models) did have some difficulties. Lt. Col. Harrison Thyng, then Commander of the 33rd Fighter Interceptor Group and an ace fighter pilot with five kills in World War II and another five in Korea, noted in June 1950 that “the F-86 is a highly specialized jet aircraft. It cannot be handled and flown like a typical Air Force fighter plane, or even like the F-80 and F-84 jet.”42
The F-86A-5 had difficulties with both its J47 engine and heavy stick forces in high-speed pullouts. As a result of the latter, one accident board found it unsuitable for air-to-ground gunnery. The commander of Nellis Air Force Base, where the Air Force conducted F-86 training, recommended that these aircraft be restricted from dive-bombing and high-angle strafing.43 Almost two and a half years later, another commander at Nellis curtailed student flying of F-86As in both air-to-air and air-to-ground firing. He noted that this model was known for its “violent snap roll tendency” in a high-speed stall.44
Another serious problem concerned the control stick. In January 1953, accident investigators reported that the separation of the stick grip from the control column caused a fatal accident in which the pilot flew into the ground at Nellis. A review of the incident records and an inspection of F-86s at the base revealed three other incidents involving separated stick grips and another four cases of loosened grips. The USAF noted that since 1 January 1952, there had been a dozen undetermined fatal accidents that might have been caused by the same problem.45 (The one-year lapse between the first incident and the fatal accident in January 1953 indicates that the Air Force’s system of quickly circulating hazard information was flawed.)
Engine failure was the greatest cause of F-86 accidents, accounting for 15 percent of the major accidents.46 Some of these failures could not be further pinpointed because both rapid throttle movement (pilot induced) and system failure (materiel or maintenance failure) could cause flameouts. A case in point is the F-86 experience with the emergency fuel system. In addition to the standard fuel system, the Sabre was fitted with an emergency fuel regulator that could pump fuel to the engine in the event the main system regulator or switch failed. It could either be switched “on” to replace the normal regulator, or engaged in the “standby” position (on takeoff), in which case it would automatically take over fuel regulation if it sensed the fuel flow was not acting in accordance with the throttle position. Standard procedure directed the pilot to turn off the emergency fuel regulator after takeoff and use it only in the event of the failure of the main regulator. If the emergency fuel regulator was left “on” or inadvertently switched “on,” it could, along with rapid throttle movement, call for too much fuel and cause an engine flameout.47
This safety device proved troublesome from the start because the first emergency fuel regulators were “sensitive to rapid throttle movement.” In the year beginning March 1949, the 1st Fighter Wing suffered seventeen flameouts, fifteen with the older of two types of emergency fuel regulators. (One pilot survived two flameouts in one day; another, two flameouts in one month.) In early 1950, an accident board recommended that F-86s equipped with this device be restricted to flying below twenty-five thousand feet, and that fall, one engineering officer urged that these Sabres be used only as lead aircraft and flown only by experienced pilots. The USAF replaced the early emergency fuel regulator with another, yet this did not completely solve the problem.48
In September 1952 an accident board supported the USAF study that the device should be removed from the North American fighter. Two weeks later another accident board made the same recommendation and further counseled that the system should only be used when the primary system had failed. The engineering officer wrote, “This system might possibly cause more accidents than it prevents.”49 An accident board in February 1953 echoed these views. It went on to note that the Navy’s version of the Sabre (FJ-2) had no emergency fuel system. Perhaps more persuasive was the statement by the commander of a training unit using the F-86: “Pilots forgetting to turn the stand-by switch off after take-off proved to be so destructive . . . [that] this Group was forced to abolish its use for take-off.” Col. Clay Tice went on to state, “The emergency fuel regulator has been responsible for the destruction of far more aircraft at this base than it has saved. It is inconceivable that it has not been redesigned.”50 The next summer the authorities reported that malfunctioning emergency fuel regulators had caused five major accidents at Nellis. They wanted a new switch (with only a manual “on” and “off” position) that would satisfy their current standard operating procedure that the device was to be used only in the event of the failure of the main regulator. Tice again criticized the system: “The emergency fuel system in the F-86 aircraft has been unsatisfactory since the first models of the F-86 were delivered to the USAF. The system is still unchanged, although it has been repeatedly U.R.ed by using organizations.”51 Air Materiel Command recommended that the device not be used in a standby mode and only be engaged if the main fuel system failed.
Despite these criticisms, the system had its supporters, including the Fifth Air Force staff and Maj. Gen. Glenn Barcus, vice commander of Air Training Command, who did not agree to remove either the device or its automatic feature.52 In the end, the system was retained without the “standby” feature, consisting only of “on,” “off,” and “test” positions.53
A further fuel control problem involved the main fuel pump. A 1954 report stated that 32 of 144 F-86F accidents (22 percent) were determined to be or suspected to be related to the fuel control system. It noted that the F-86F’s engine (J47-GE-27) was “very susceptible to compressor stall.” The report blamed the pilots for not following proscribed procedures.54
USAF accident statistics lump all F-86 accidents together, not distinguishing between the various models. Three Sabre models served in the Korean War: the “A,” “E,” and “F.”55 By the end of the Korean War, the Air Force had received 554 “A”s, 396 “E”s, and 1,139 “F”s. By that time, two-thirds of the “A”s, one-third of the “E”s, and under one-tenth of the “F”s delivered were involved in major accidents.56
An analysis of 637 major F-86 accidents through July 1953 reveals little difference in accidents of the various Sabre models. One significant difference involved ejections. Up through this time period, ninety-one pilots ejected from the Sabre. The “A” had the lowest percentage of parachute/ejection seat use in major accidents, half that of the “F.” Perhaps more remarkably, only 5 percent of the pilots using parachutes in the F-86A were killed, compared to almost five times that rate in the “E” and five and a half times that in the “F.” This may actually understate the differences because there were additional incidents involved with the latter two models where the pilot may have attempted egress.57 A lower percentage of F-86A pilots were injured in major accidents than in the later model Sabres.
There were few differences between the models in regards to the causes of accidents. Problems with the landing gear markedly declined in the models that followed the F-86A. Engines were the most frequent and significant problem for each variant and accounted for thirteen fatal accidents.58
Prior to August 1953, there were twenty-one F-86 accidents involving two or more aircraft in which one or more fighters were destroyed. These multiple accidents involved forty-seven Sabres, wrecked forty-two fighters, and killed twenty-one pilots. As expected, the bulk of these were midair collisions.
Paradoxically, the three accidents involving three or more F-86s were not midair collisions. On 18 October 1950, the 335th Fighter Squadron (4th Fighter Group) lost three F-86s and two pilots when the leader let down through the clouds and flew into the Potomac River. Weather was the primary cause of another multiple Sabre accident on 31 January 1953. Seven F-86s (432FS) took off from Traux Field, Wisconsin, on a practice mission and ran into unexpected rapidly deteriorating weather. As a result, four Sabres were destroyed and two pilots killed. The third incident was somewhat different, involving a flight of four F-86Fs (67FBS) taking off on a combat mission in Korea. The leader crashed on takeoff, most likely due to an improper elevator trim setting. The number four man in the formation fell behind the other two Sabres as they began a return to base and disappeared. He was killed in an unexplained crash. The original number two man attempted to land but, clearly shaken by the events, aborted two landing attempts and on the third skidded off the rain-slick runway. He survived without injuries from the crash that destroyed his aircraft. The result was three fighters destroyed and two pilots killed.59 This incident is set apart from other multiple accidents in that the three aircraft, even though they took off together, were destroyed by different causes.
Since 1921, the first year in which army airmen kept comprehensive records, the flying accident rate has declined. This has not been a straight-line trend; the rate shot upward in 1941, when America began the buildup for World War II, and again in 1946, a result of the rapid demobilization following that conflict, before resuming its downward trend.60 Even during the hectic expansion of the Korean War, the USAF’s worldwide accident rates fell. In fact, in 1953 the flying accident rate was the lowest yet registered in terms of major accidents. Moreover, the fatal accident rate was only better in one year, 1939; and the rate of wrecked aircraft was only better in two other years, fiscal 1950 and calendar 1952, than those rates in 1953.61
These figures lump together all Air Force aircraft. As might be expected, there was a great difference in the accident rates for aircraft types. During World War II, the major accident rate for fighters was about three times the AAF average, over seven times that of transports, and almost five times that of four-engine bombers. Although all accident rates fell after the war, the gap between categories remained. There were also differences between specific aircraft within categories. During World War II, for example, the P-39’s accident rate was over twice that of the P-51. The accident rates of jet fighters displayed similar characteristics: a higher incidence than other types of USAF aircraft, and a spread, albeit somewhat smaller, between specific fighter types.
The standard measure for flying safety is based on the number of incidents per one hundred thousand flying hours. On its face, this seems flawed. Fighter missions are shorter in duration than those of bombers or transports and are frequently at the edge of the performance envelope—clearly more dangerous than the straight and level operations of transports. There are other ways to compare the flying record of these aircraft.
An alternative to using flying hours would be to use landings as a measure. Between two-thirds and three-quarters of all major accidents occurred during taxi, takeoff, approach, go around, or landing, maneuvers that take approximately the same amount of time for all the aircraft. The remaining one-quarter to one-third of the accidents occurred in the air, where the flight time of the different types and models of aircraft vary.62 Using landings as a basis, the spread of accident measures between the entire USAF and jet fighters widens. This measure also changes the relative position of the four fighters. Using landings as a measure, the F-86 replaced the F-51 as the safest fighter. (The F-51 had a considerably greater endurance than the jet fighters.)63
Another alternative method would be to compare the various fighters at comparable stages of their life cycle. This would account for the learning curve. If we stagger the chronology by comparing the accident rate at the same cumulative flying-hour intervals, the F-86 once again comes off best.64 The learning curve can be seen because the major accident rates for all three fighters declined sharply between the one hundred thousand and five hundred thousand flying-hour mark, with the F-86 experiencing the largest drop.65
The superior flying safety record of the F-86 is not easy to explain. The F-86 was a more advanced aircraft than its peers, capable of a higher top speed and, more importantly from a flying safety point of view, higher takeoff, approach, and landing speeds than the other fighters. Its higher performance put additional strains on man and machine. The F-86 was designed shortly after the F-80 and at essentially the same time as the F-84. One possible explanation for the Sabre’s superiority focuses on the way the three fighters were employed. The F-80 and F-84 were primarily used as fighter-bombers and thus carried heavier loads than did the F-86, which was initially fielded as an air defense fighter (interceptor) and primarily used in the Korean War as an air superiority fighter until 1953, when a fighter-bomber version went into action. Bombing and strafing runs took these aircraft close to the ground, where a small miscalculation or mechanical problem could easily cause an accident. The same incident at altitude, however, where most air-to-air training was conducted and where air-to-air battles began, gave some margin for recovery. The first operational jet fighter, the F-80 was extensively used in pilot training, both for those learning to fly and also for those who would fly the Shooting Star in combat. Through 1953, 38 percent of the fatalities in F-80s occurred in flight training units, compared with 25 percent of the F-86 fatalities.66 It is possible that better pilots were attracted to or assigned to fly the F-86, or that pilots flying the Sabre were more careful. Or it could simply be that the F-86 was just a better aircraft.
Just about everyone familiar with the F-86 agrees it was a fine aircraft, and all use positive, if not superlative, words when speaking of it. It was the top aircraft of its day and was a better performing and safer fighter than either of its two USAF teammates, the F-80 or F-84. Having said this, it is true the Sabre had problems. It is also a fact that its safety record does not compare with Air Force fighters that followed it. Even the much maligned F-100 and F-104 had superior safety records. Bringing the story up to date, comparing the Sabre’s safety record with the F-15 (the Eagle), the safest fighter in USAF history, is astonishing. The Sabre had eighteen times the Eagle’s major accident rate, eleven times its wrecked accident rate, and eleven times its fatal accident rate.67 However, in its day, the Sabre was a standout in terms of flying performance, combat record, and safety.
Clearly the USAF has come a long way in terms of fighter performance and flying safety since the days when the F-86 ruled the skies. The progress since this period has been absolutely remarkable. Compared to a major accident rate of 24 per one hundred thousand flying hours in 1953, the lowest rate to that point, the major USAF accident rate has been less than 2 per one hundred thousand flying hours beginning in fiscal year 1983. Between fiscal year 1975 and fiscal year 2000, the rate averaged 1.92 major accidents per one hundred thousand flying hours.
A number of factors account for this progress: certainly, new and better equipment; probably improved training; and, perhaps most of all, a different attitude about flying safety. During the years prior to 1960, accidents were almost accepted—that is, they were considered a cost of flying. Although we lack documentary proof to support this assertion, there is anecdotal evidence. One former F-86 pilot recalled that checklists were not required during the 1950s, and that “most real fighter pilots did not carry a checklist with them.” He went on to note that, “a lot of our accidents were [a result of] trying to prove our manhood.”68 During the Korean War, a World War II ace and an experienced pilot (over eighteen hundred flying hours), lied about having prior experience in the Sabre and asked for no help other than that of the crew chief to start the engine when he made his first flight. He crashed after logging eleven hours in the Sabre, according to another pilot, because he did not know how to use the fuel in the drop tanks.69 There have been significant changes in the Air Force over the past half century, and nowhere are they more obvious than in the area of flying safety.