Accuracy also counts. As shown in the papers I posted recently the B-24s clumsy handing characteristics resulted in much fewer bombs on target. In 1944 B-17s put 15.0% of bombs within 500' of aiming point whereas the B-24 could only manage 10.2% a huge difference. This was in good visibility. I would assume the poor handing of the Halifax as compared to the Lancaster also resulted in a disparity.
The USSBS Bombing Accuracy report is not as harsh on the B-24, it uses within 1,000 feet as the measurement. The majority of the targets the 8th Air Force attacked had a large footprint. The USAAF raid on Huls on 22 June 1943, after the raid the bomb craters were found over an area of 12 square miles, 20% of the bombs dropped hit the 541 acre (0.84 square miles) site. If the site were a square then each side fence would be 4,650 feet long. The USSBS excludes missions where less than 5% of bombs landed within 1,000 feet of the aiming point and boxes that had a circular error of more than 3,000 feet. Given the 8th Air Force reporting the B-24 had a higher percentage of gross errors that would help explain the USSBS difference with the 8th Air Force. If their was a problem with the B-24 versus the B-17 it appears to be about hindering correctly identifying the target.
USSBS Average attacking altitude B-17 21,542 feet, B-24 19,800 feet, average attack size B-17 6.64 aircraft, B-24 6.23, average box size B-17 14 aircraft, B-24 11. All of which should make the B-24 more accurate, but for bombs within 1,000 feet of the aiming point, 8th Air Force B-17 40.77%, B-24 37.8%, 15th Air Force B-17 32.4%, B-24 30.4%, based on data from 1,348 8th and 800 15th Air Force Missions The half way point for bomb tonnage dropped was second half of September 1944 for the 8th Air Force and end August 1944 for the 15th Air force. Part of the 1944 difference between the 8th Air Force B-17 and B-24 results is as of end 1943 there were 18 B-17 and 7 B-24 groups operational as of D-Day that was 21 B-17 and 18 B-24, when the 21st B-17 group went operational on 6 May there were 11 B-24 groups operational. The B-17 units had on average more experience until the second half of 1944.
The USSBS says the effects of bombing order were, % of boxes placing at least 10% of their bombs within 1,000 feet of the aiming point, box 1 82%, box 2 60%, box 3 48%, box 4 47% box 5 or more 30%. Richard Davis figures have from October 1943 to the end of the war attacking Germany the 8th dropped 41.2% of bombs visually. The 15th Air Force started non visual bombing in April 1944, from then to the end of the war it managed 89.4% visual bombing.
One query is the number of boxes of B-24 attacking a target, the same number of bombers in smaller boxes would mean more subject to the problems of smoke and dust obscuring the target.
If you go to the 8th Air Force September to December 1944 accuracy report it has 38.3% of B-17 and 28% of B-24 bombs dropped in good visibility landing within 1,000 feet of the aiming point. Below 12,000 feet it was 31.7% B-17, 19.8% B-24, over 24,000 feet 12.6% to 11.8% as the bomb sight limits were being reached. The 8th noted 37% of visual attacks were done at 21,000 feet or higher in Q1/1944, going to 97% in Q4/1944. Similarly raids using visual bombing that had over 9 boxes average 28% for 1944, but 55% in the final quarter. The idea of deliberately setting the sight incorrectly so when it sees a nearby landmark the bombs released will hit the target was considered, not possible with the Norden, it was possible with the mark XIV.
The Visual Bombing Summary for 1944 notes 19.8% of B-17 attacks in good visibility resulted in gross errors, for B-24 29.5%, that would imply the major difference in accuracy between the two bomber types is in target identification, rather than better bombing arrangement in the B-17. The sweet altitude for the B-17 was 3.3% gross errors at 15 to 17,900 feet, compared with 10.3% below 12,000 feet and 26.2% at 24,000 feet or higher, B-24 sweet altitude was below 12,000 feet at 14.9%, next was 18 to 20,900 feet at 25.8%, every other altitude it was 30.8% or more. The B-24 had a higher percentage of attacks when the visibility was poor.
The 1 January to 15 July 1943 8th Air Force accuracy report rejects gross errors, aiming with radial errors of 3,000 feet or more, however the all missions table, with between 6.3 and 13% of bombs within 1,000 feet of the aiming point looks like it includes gross errors as the successful group attacks table has 18.3% to 28.7% of bombs within 1,000 feet of the aiming point. This report showed bombing by group was the most accurate, plus the fall off in accuracy with bombing order, from 23.4% for the first to 4.2% for the 5th or later formation (within 2,000 feet went from 49.9% to 19.8%). Gross errors comprised 10% of first formation attacks, 72% of 5th or later attacks. Exclude gross errors and it was 26.2% within 1,000 feet for the first attack, 15% for the 5th or later. In other words it was the target becoming obscured that was a big problem. Go from 1 minute between attacks to 5 minutes and the accuracy goes up, mostly by eliminating gross errors, the 5th and over attacking formation go from 4.2% to 19.8% of bombs within 1,000 feet of the aiming point. As of mid October 1943 24% of group attacks were gross errors.
There is a definite difference between the 8th Air Force B-17 and B-24 accuracy using visual bombing, yet it could mostly be the B-24 bombed on average with less time between formations.
I recall a study of Crossbow bombing results which showed the Mosquito to be by far the most effective bomber in that particular circumstance.
Mosquito VI fighter bombers versus the light, medium and heavy level bombers.
