Analysis of Naval Gunnery Combat

[Sardaukar]

I had a chance to check the WPO game that my friend did buy, so obviously I was curious if I should too. In that I was seeing similar things. Capital ships were just doing nothing..or firing only their secondary armament. And those were British capital ships..and they didn't even fire when they surprised IJN several times. Admittedly those were night engagements...but still...

[Speedysteve]

From "The Rising Sun: The decline and fall of the Japanese Empire," by John Toland

FWIW I am currently reading this book and find it a very useful addition to a Pacific War collection if not already obtained.

[String]

ORIGINAL: Sardaukar

I had a chance to check the WPO game that my friend did buy, so obviously I was curious if I should too. In that I was seeing similar things. Capital ships were just doing nothing..or firing only their secondary armament. And those were British capital ships..and they didn't even fire when they surprised IJN several times. Admittedly those were night engagements...but still...

I guess you were playing the "Pacific Jutland" scenario. In WPO the starting night experience for all ships on all sides is around 20-30. Very low. Therefore, the first night engagements tend to be very very inconclusive. However, after a few battles the ships do quite well in night combat.

I do not know if the naval combat routines were tweaked in WPO, but I haven't noticed such problems there. My BB's almost always use their main battery and do quite a bit of damage. So do enemy BB's.

Ofcourse, there has been a night battle or two with no main calibre hits scored, but those were between ships with 30ish night exp.

I have only witnessed one single day combat battle where a battleship failed to score a hit with her main calibre guns while being undamaged. That was when a single destroyer was engaged by a japanese pre-dread and a cruiser escort. The range never closed to under 23000 yards and I presume the destroyer escaped utilizing her superior speed.

[Sardaukar]

Yep, was Pacific Jutland (since I was just trying the game). Got bit annoyed when my Brit BBs just did suck in gunfire and torpedoes even when they surprised IJN 3 out of 4 fights...[:'(]. But I can accept it as being night battles mostly 2000-7000 yards... Just wanted to see how the surface combat was in WPO, so tried to get as many of those as fast as possible...[:D]. I'm still wondering how the hell my BB crews coudn't see burning IJN ships in that distance, tho...[:D]

[Feinder]

I don't know crap about any of this, but this thread is an interesting read! D_mn Tom, you have -WAY- too much time to be watching this stuff! (* just kidding, thanks! *).

Not to dillute your thread, but if you're ever really, extremely bored, you might want to test the "declining penetration value vs. deck armor hits in day-time combat".

a. At greater than 12,000 yds you have have a greater chance to score a deck hit (at least in WitP, sounds reasonable in real life = plunging fire).

b. Day-time combats usually open at about 18,000 yds, and several rounds are often exchanged as the range closes.

c. During these exchanges, you are quite likely to score a deck armor hit (a good thing I believe).

d. However, shells hitting the deck armor, RARELY, EXTREMELY RARELY, ever penetrate, regardless of the size of incoming shell (I've seen 8" shells from CAs not able to penetrate the tinfoil deck armor of CLs).

e. I understand that the devs decreased penetration values as range decreases (I don't see why it would really matter, maximum velocity x mass is the same at 8000 yds or 18000 yds, if your hitting max velocity). Still, I'll take their word for it. But I've noticed that penetration drops WAY off at 12000 yds (as above example). If it's going to drop off (again, I'll take their word for it), I think it's dropping off TOO much. Usually, a ship was at least vulnerable to plunging fire from guns of comperable size. But I just don't ever see penetrations in the replay from deck hits. I can't speak for my opponents, but I know my own ships only take paint damage from "non-penetrating deck armor hits" (and I'm thinking that 8" shell -should- have penetrated the deck armor).

Dunno. Has anybody seen this?

Frankly, I love to watch a good naval battle in Witp (esp one that I "win"). The .txt file doesn't tell you the story, you MUST watch the battle to see if those 6x hits were 3.5" pop-guns or 15" ruin-your-days (altho Tom's research seems to indicate that that they are rarely the ruin-you-days, and I think he's right!).

Whatever.

Interesting thread.

-F-

[el cid again]

Jutland, an analysis of the fighting, has a hit rate of 2-3% for the run to the South and to the North, Iron Duke firing at 6,000 yards was doing 10%.

You do realize that, in offering this data as a criticism, in a technical sense you CONFIRMED my data: the 2-3% for medium range is indeed possible with skilled crews in good conditions - and a single exceptional datum point is just that - not something you can build a typical simulation algorithm on.

[el cid again]

From "The Rising Sun: The decline and fall of the Japanese Empire," by John Toland, Page 474 USS Washington fired 75 shells, and hit 9 times which is a 12% rate.

Perhaps you should consult a mathmetician. Single ship datum points are not what you need to build a rule of thumb or algorithm that is useful in simulation. If you add in all the occasions on which there were no hits at all, you will be depressed at the statistics. If you rule all of those cases out - then you must also rule all the other exceptional datum points out at the other end of the range. You can either include EVERY datum point - including all the zero cases - or you can focus on those within a certain distance of the mean case - and ignore BOTh extremes. But you cannot just look at the extream case on ONE end and say "that is the norm." Turn it around: how would you feel if I cited a case where a single ship fired abc rounds and scored no hits, concluding 0% is the norm?

[el cid again]

"Life and Times of Old Shiny" has her getting the drop on a German DD at point blank range and doing 'over a donzen hits in two minutes" Sheffield can fire a maximum of 120 to 132 shells in 2 minutes which is also a 10% hit rate, IF she fired the maximum number of shells possible. This was at very close range, under 1000 yards, but she also had a very short time to react due to low visibility. Sheffield did much better than the German DD, which failed to fire at all.

We don't have the statistics, but a German raider managed to sink Sydney - at point blank - with inferior weapons. She didn't sink at once - but limped off burning and never was heard from again. The raider was also sunk in the same action - but some of the crew survived. At point blank range it is normal to score very well. But it is wrong to believe that average ships in average conditions will routinely score more than 10% hits even under such circumstances. This is not to say a superb ship under exceptional conditions could not sometimes achieve that - a different statement entirely. But you should not be expecting this, and you would be much more justified to cry "naval combat is broken" if you saw it regularly - than if you see lots of misses. Lots of misses ARE normal - even when you cannot explain why.

There is a submarine simulation game which got modified to the point it is "too realistic" - it is no longer easy to get a fire control solution. Players hate it - but that does not mean it is a bad simulation - it just means players like to see things go boom.

[el cid again]

But inside these combat rounds we do know EXACTLY how many rounds of ammunition were expended. The game tells us.

I do not think it does. For example, a "mineshot" is 8 mines. A torpedo shot is typically 4 or 6, but sometimes 2, and sometimes other values. I mean in the game - forget real life - if we did real life a Japanese sub would ALWAYS shoot ONE torpedo at a TANKER! When a rapid fire gun is involved, it is shooting bursts, not rounds. Do we even know what the tacical impulse of time is? You must take the ROF times that many minutes to calculate the theoretical maximum number of shells that might have been fired. Do you know how many guns could not bear on ship XY in impulse C? No you do not. Do you know how much below the maximum rate of fire it was shooting? No you do not. The game tells you "shots" - but do you know how many shells that represents? I doubt it.

[el cid again]

For all I know or care the combat model assumes only one gun per ship fires.

Look at the data. The database knows how many guns are forward, how many aft, how many port, how many starbord, how many center (both port and starboard but neither forward nor aft). It also knows how many of which are damaged. It picks the number of guns based on the situation - the bearing of the enemy - if to port and you are down a turret on the port side it matters. If your T is crossed only your forward (or aft) guns bear. And so on. ROF probably varies. At long range the ship does not start at maximum ROF - it is ranging. AFTER it finds the range it probably goes to something near maximum - depending on the rating of its captain. Maybe not all the way. And how many shells is a "shot" varies - it is a low number for heavy guns but hundreds or even thousands for machine guns - over a space of x or xy minutes. We do not know. Analysis based on so many unknowns is bound to lead to false conclusions.

[Sneer]

naval gunnery was pointed to be invalid during beta tests of WPO - but unfortunately it lies too deep inside witp engine
it isn't a big problem in Witp / but huge in WPO

[el cid again]

Basically your 2-6% hit rate figure is not correct, and there are numerous sources that say so.

No it is not. It is zero point 2 to zero point 6% for optical ranging at very long ranges (just inside the visual horizon). It reaches only 2 or 3 % at reasonable ranges. It reaches only around 6% at point blank range.
This is the base for naval combat with optical fire control methods in the period in question. If you talk about major coast defense batteries, it goes up. If you were to talk about something done long after the war with modern radar and computers, it can go up too - USS New Jersey in 1968 typically fired single shot salvos - the need for turret salvos or broadsides was a thing of the past. And note these values are GOOD shooting - a poor crew will nearly always miss. In bad sea conditions a skilled crew will nearly always miss.

[el cid again]

This is simply not true. All of the possible explainations would result in a lower ammo expenditure, which is a critical point in my analysis. Ships that can't hit the target don't fire 132 rounds. Not ever, you will never find a real life example.

Sure they do. There are more amazing things than that. Two nights before the Japanese landed on Lingayen Gulf, the night sky was filled with artillery shells - due to a false sighting of apparently nothing (wave surf perhaps? a local boat ?). This went on and on for quite a period, and impressed locals, and caused some reporters to think the invasion must have come. But there were no hits at all. But when the real landing came, it appears those same batteries managed nary a shot. A generation later guns had a bad reputation, and many ships didn't believe in them. [I, on the other hand, said they could shoot down guided missiles. This has since happened in real battle.] On my first ship we always won the E for gunnery, and we were trained to a standard of "if you don't hit with the first round, you didn't solve the fire control problem correctly. To INSURE you solved it correctly, we will ONLY fire ONE round per tube - and it better hit - in practice." But while our gunners rarely missed, it was normal for all the ships firing in a day at San Clemente range to score fewer times COMBINED TOTAL than ONE of our gun mounts did! It is quite easy for humans to mess up things like gunnery, particularly if they are on a moving ship, shooting at a moving target.

[el cid again]

You cannot find high ammo expenditures and hit rates at 1% or below in the 20th century, you have to go to the 19th century for that.

Au contraire, mon ami, you cannot find hit rates ABOVE 1% at captial ship ranges for any case where there is a statistical number of shots in the 20 th century - stipulating this is shooting from a naval vessel at a high speed naval target. For hit rates above 1% you must reduce the range, or get lucky, as in the case of Bismarck vs Hood. But note that that case does not count - since Bismarck didn't expend a statistical number of rounds to do in Hood. [ A statistical number is NEVER less than 30 FYI]

[el cid again]

The more I crunch it the worse it looks. Loot at the Japanese hit rates at 4000 yards! Any Japanese officeer that bad would not be running guns on a warship.

Consider Japanese gunnery in the action off Samar. Of course, these ships were under air attack (some of it was real, some of it simulated, to mess up their gunnery). The weather was not ideal. The crews were not well trained any more - lack of fuel and ammunition contributed - and losses contributed - most ships had not been practicing at all. It was pretty dismal that day. The operational goal was achieved - the US carriers were off on a wild goose chase - the US fast battleships were off on a wild goose chase - and the heavies got into the landing forces and their supporting CVE groups. But it didn't pay off very well. Had this been the crews and ships similar in quality to Savo island, we would have been hurt a great deal more.

[Feinder]

Now you're just being cranky El Cid.

Step away from the keyboard...


[;)]

-F-

[el cid again]

Hi Cid. How old are you? What "battle" was this? Desoto Patrol instigated Gulf of Tonkin "Incident" or what? Curious.

You will be hard pressed to hear about it. We don't brag about our strategic defeats! But you WILL find discriptions of the "incursion into Cambodia" that say we could not stop the supply over the sea from Malaya - without telling you how we know that. In the beginning, we did NOT know that. We tried. We failed.

Every night hundreds of junks would make a crossing from Northern Malaya to Cambodia - 200 to 500 of them. We felt that "with radar fire control" we could just go and clean house - and stop them. We did not know they were armed - and had we known we would not have greatly feared army wheeled guns on the decks of pitching junks! Boy were we wrong. Unarmored warships do not do well even under rifle fire (see a CBO study on that, and the reason we now put kevlar armor on ships).
Machine guns can ruin a ship as a fighting unit in a minute or two. Actual 122, 130 or 152 mm shells can to horrible things. We had so many near misses that it was regarded as near miraculous we didn't get creamed, and we dared not close the range as we had planned. [As far as I know this op was not repeated. If it was, it failed too - since we never did cut the supply line until the Army incursion more than three years later.] This is a lot like "the Glorius First of June" or whatever the British call it- the French fleet got hurt badly that day - but in France they teach it as a strategic victory - because the OBJECTIVE was achieved - the sugar cane convoy got through. We did sink junks - but they didn't play naval battle from the Western book and run (getting more sunk while running). They just closed the coast, merged with it and we lost targets, and delivered the goods, using heavy fire to force us to stand off out of maximum effectiveness range. It is amazing to see how a division of ships totally sure of their technical superiority can utterly fail to do what on paper seems a simple taks. But that is the nature of real war and real battles. We cannot blame this one on politicians.

[Tom Hunter]

El Cid

I'm supporting my statements with refrences and anaylsis. Your welcome to any opionion you like, about the gunnery combat model or or the hit rates of big guns, but I would prefer that you either produce some stats from the game or from some written source. Of course you don't have to do this, but it would contribute a lot more if you did.

More sourced material and analysis coming soon.

[Tom Hunter]

Here is an intersting table showing what the US Navy thought it could do in the late 20s


The first number is the accuracy if the spotter is in the fighting top of the BB, the second is using an arial spotter. These are daylight tables.

Table 1: Accuracy of Battleship Gunfire[25]



Range

(Yards)
Percentage of Hits

Top Spot
Plane Spot

12,000
12.3
---

14,000
8.9
---

16,000
6.2
---

18,000
4.2
---

20,000
2.6
4.3

22,000
1.5
3.4

24,000
0.7
2.7

26,000
0.1
2.2

28,000
---
1.8

30,000
---
1.5

This was published in 1923 so we are two generations of fire control farther on, with a substantial increases in accuracy.

[Tom Hunter]

Now a look at the DDs

The British had 6 DDs plus the Le Triomphant. I don't have ammo figures for the Free Frnch DD so I am just looking at the Brits.

During the suprise round HMS Voyager took 3 hits from Kuroshio, Vampire was fired on by Hatsukaze, Isis by Hayashio and Le Triomphant by Hayashio none of which scored hits.

Voyager fired back on Kuroshio in round two and hit her once. She fired in round 3 and missed, and did not fire at all in the second engagement. She fired 22 shells and got a respectable .045 % hit rate. It takes Voyager 46 seconds to fire this many shells.

Vampire did not fire at all in rounds 2 and 3 and missed Hatsukaze in the second engagement.

Vendetta fired on Natsushio in rounds 2 and 3 hitting once at 4,000 yards and once at 8,000 she fired 22 shell getting a .09% hit rate again this is a 46 second barrage for this ship.

Isis opens up on Asagumo in round 2 of the first engagement hitting 1 time at 4,000 yards. She then misses at 8,000 yards in round two and does not fire in the second engagement. This is a .02% hit rate fired over 90 seconds.

Jupiter engages Natsushio in round 2 of the first engagement hitting 3 times, she then switches fire to Nagara in round 3 hitting twice, and in the second engagement she puts 2 shells into Hagura. In total she expends 154 rounds of ammunition, it takes her 3 minutes 41 seconds to shoot this many times. Her hit % is a respectable .045.

Norman does not fire in the first engagement, in the second engagement she opens up on Hayashio firing 44 shells and hitting twice for a hit rate of .045. It takes her 61 seconds to fire these 44 shells.

Of all the ships we have looked at in these battles the DDs are closest to having the historical hit rates, in fact they may be about right. They are certainly not off by an order of magnitude. However it is likely that they are not firing enough shots, but that an argument for another day.

One other thing that is interesting is how much the battle was fought be just 6 of the 14 Royal Navy ships present. Warspite, Exter, Leander, Achilles, Newcastle and Jupiter fired 69% of the shells, though Le Troimph also fired regularly and pounded the tar out of a Japanese DD. That means the remaining 8 ships fired about 3% of the total output each. Also the amount of time it takes to fire the shells is wildly divergent. Warspite fired full broadsides for 8 minutes, or for longer if they were not full, Vendetta fired for 46 seconds but did substantial damage, Jupiter, Achilles and Exeter fired 3 minutes or more, Leader over 2 minutes and Newcastle for 90 seconds. I don’t have any evidence to prove that something is fishy here but I may get some when we delve into historical battles.

So far we have shown that BBs are completely FUBAR, failing to use their main battery when targets as big as heavy cruisers are present, something which never happened historically even down to DD sized targets and has zero justification. Cruisers do not get historical hit rates, and DDs for seem to be operating more or less correctly in terms of their hit rate but may not be firing enough.

I also have data for the second big fight between the Prince of Wales TF and this group of Japanese, and between an experimental TF I built with one BB and a lot of PT boats. More coming, and more problems to be exposed, plus a good reason to group BBs and PTs.