Regarding Iowa's performance in Operation Hailstone, an overreliance on calculations and practice gunnery exercises which are not backed up by actual combat performance is an issue. The percentages of expected hits which you referenced (14%?) are, as far as I'm aware, well short of historical gunnery performance by any capital ship during WW2. Everyone shoots better in practice than they do in actual combat when the prospect of physical harm is present, and I believe that this is no less true of warships than it is of infantry. The common denominator is human beings, and warships in action are collections of individual human beings attempting to function as one. I don't question the science or the data that you've presented. I'm by no means an expert in the field of naval gunnery, much less a professional. The cause of Iowa's dismal performance against Katori, I believe, is due to human application. In what specific area that occurred, I'm unable to say. What I do know is that Iowa should have secured a number of hits on Katori and ravaged the ship under the conditions of that engagement, and yet her captain (the last person who would be inclined to acknowledge it) made it clear in his after- action report that Iowa obtained no main battery hits in a short to medium range engagement which took place under ideal conditions. Regarding shell splashes obscuring possible hits on Katori, I believe that the air strike coordinator observed Iowa's fall of shot, and that he did record a possible hit on the second salvo- which Iowa's captain did not claim in his after- action report.
I think it’s important to point out that action reports are really a compilation of observations and reports from the various department heads. These reports are rarely personalized observations, and, in that regard, the captain of a battleship would rarely be in the best position to observe the effects of gunfire. The performance of the batteries would be promulgated from the Gunnery Officer to the captain in his own report which would then be incorporated within the official action reports.
There is, of course, a significant difference between combat and exercise and I am certainly not arguing otherwise. However, I question the validity of that observation here. By your own statements previously,
Katori was dead in the water with half the ship engulfed in fire and smoke. She was not observed to have been returning fire and the amount of return fire in general from the handful of ships that were intercepted was rather light. While not entirely stress-free, it was hardly a taxing engagement, and I remain quite skeptical that it played a significant role. You've essentially described conditions more favorable than many peacetime exercises yet still claim what would essentially be a catastrophic degradation of accuracy.
I have previously illustrated why hits were difficult to discern which is why it is important to discuss other possibilities. This is precisely why mathematical analysis matters - it quantifies just how improbable zero hits would be under these conditions. The gunnery data we have accumulated was used to calibrate a couple of math models. One was from Princeton University’s Statistic Research Group as well as a Monte Carlo simulation. Both models have been validated against the actual fleet data.
These models are predictive in nature and can accurately recreate observed events when provided with specific data inputs on MPI and dispersion. For example, one must plug in values of Mean MPI errors, Standard Deviation of MPI errors and TMD. When this is done then they give very accurate results. While combat conditions rarely provide such precise measurements, however, radar spotting data offers a reliable alternative.
Specifically, radar spotting proved capable of determining MPI to within ±60 yards, providing accuracy broadly comparable to photographic or mechanical rake methods. I have used this data as a substitute for photographic rake data, which is how fall of shot is normally measured, to plug into those models and have found that the results are accurate to within 0.5-0.75% of the more precise rake data. Ex. for a given event, the models using rake data predicts 6.5% hits whereas the models just using the radar spotting data might show 7%.
Now why is this relevant? Because we have radar spotting data here. I do not have TMD data but making a fairly conservative estimate on TMD gives one a good starting point for analysis. When I did that, I found that every single simulation run I would make I always produced hits. I then decided to start increasing MPI and TMD values and simulating them again.
What I eventually found was that it required me to quadruple the MPI errors and TMD values to get situations where no hits were predicted and even then, it was relatively rare at around 15% of the simulation runs.
One may ask would such MPI and TMD errors be obvious to the
Iowa crew and yes it would have been blatantly obvious. A TMD that is 4 times average would be 2.52% of range which would translate to an 8-gun pattern of 9.8% of range. To put this in perspective, the Mark 8 radar scope, when in PRECISION SWEEP, displays about 1000 yards on either side of the target and in this example, the splash pips would have been distributed across 3/4ths of the scope in range.
In addition to that, the patterns would have been shifting wildly salvo to salvo, (on average 400 yards or ¼ of the scope) and the amount of straddles would have been roughly halved. So in any case such wildly inaccurate gunnery would have been obvious and would certainly have been noted in the action reports. The absence of such notes is telling.
I believe that there's no question regarding his statement but that Captain McCrea was referring to the effect of his ship's secondary battery only.
No this is not at all what the Action Report says. I have included the entire gunnery narrative in this message however this is the only details of the secondary battery:
At 1532, the port 5" battery opened on this same AOBA class cruiser and fired a total of 124 rounds. Full radar control was used as the 5" control personnel could not see the target because of own main battery smoke. Because of this smoke and the splashes from own 16" and cruiser's 8" shells, the MPI of these 5" projectiles could not be determined accurately, but was considered to be well positioned.
The quote I post previously is clearly in reference to the main battery fire as you can see it is under point 3, reference “e” which is the main battery narrative against
Katori. The secondary battery narrative is discussed under point 4.
All of this evidence - from radar data to the action report to mathematical improbability - points in the same direction. Dismissing this convergent evidence in favor of an unsupported assertion about what is essentially human factors requires more than just speculation. It would be much more productive to quantify these human factors rather than rely on a broad statement that combat reduces accuracy.
