BoB - Kevin v Swift

[Rob Brennan UK]

that is one blind looking britain ! quite terryfying from a british perspective .. Well played.

[K.Pooley]

ORIGINAL: Rob Brennan UK

that is one blind looking britain ! quite terryfying from a british perspective .. Well played.

Thank you kind sir, you are obviously a man of great intelligence and discernment!

[K.Pooley]

The effect of concentrating on hitting 11 Group is apparent here. As with the SECAFs and RADAR tables it also shows that PR is nicely up to date.

[K.Pooley]

Morale

As the table shows, morale amongst the crews in my test gruppen is remaining remarkably stable. Now I obviously don't think that I've discovered a new Universal Constant, and that there is law of nature that says the morale of LW bomber crews will tend towards 31.7 under conditions of heavy combat, but it will be interesting to see if it stays so stable in the future turns - assuming the game lasts much longer [;)]

[K.Pooley]

Bomber Strength
This comes to 1,608 bomber-equivalents since stukas are single-use only, even though it's more complex than that, due to individual types different bombload/range profiles. It is however useful to have a rough idea of what's available when I start allocating units to specific attacks.

I don't know why my '88s should be out of action in such a large proportion, especially compared to the '111s, but the stukas' figures at least are easier to explain. For the past week or more they have been almost exclusively engaged in bombing RADAR and AFs on or near the coast, and given the lack of radar they have been much more likely to reach their target and return with little or no grief from the interceptors than was previously the case.

[K.Pooley]

Drawing on the experience of the last few days, and with a pretty clear recce picture I am now in a position to plan the next days ops in detail.

With so much data allocating resources for attacks can be difficult. For some jobs, to help avoid the rather arbitrary decisions that can result, I make up my own rules of thumb. An example is my system to decide which RADAR sites to hit, based on %age damage and age of photo. The simpler system of less than75% damage and outside certain geographical areas is another. Another advantage is that it ensures that a particular strategy is implemented consistently from one turn to the next, either to enable a strategy to be evaluated fairly, or because the approach is felt to be a winning one.

The graph took seconds to produce, because the data is already in the table, and the y-axis just needs to be changed to a %age to produce a nicely symmetrical graph. If I wanted to assign a greater importance to either damage or aircraft numbers then I would draw lines at either more or less that 45 degrees, but I want them equally balanced in this case. Drawn as shown the top left-hand corner of the graph is the area of highest priority, while the lowest is in the bottom right-hand corner.

Of course I wouldn't usually go to the lengths of drawing the diagonal lines and colouring in the different zones, and would do it by eye, that is purely to show how the technique can work.




EDIT: To remove non-standard characters.

[K.Pooley]

On another thread someone was asking what sort of bombing results they should expect, and I suggested that they set up some bombing runs and experiment. So how do I decide what altitude to bomb from? Obviously I try lots of similar runs and look to see what happens, just like in RL. I don't want to be tangling with any LAA while bombing the SECAFs this turn, since some of them have several dozen light guns.

So, how do I got about deciding the best height versus the known threat. The chart below shows estimated AA for 11 Group. Looking at SECAFs with HAA cover they can have 1, 3, 4, 7, 8, or 10 guns. Therefore I thought it might be useful to look at 4 x 3" and 8 x 3" hvy, and altitudes of 7,500' and 10,500'.

[invernomuto]

AAR subscribed. Very useful infos. Thanks a lot!

[K.Pooley]

Glad you found it!

[K.Pooley]

Flak Trials
The graph below shows the effects of flak during the course of 140 single-gruppe missions, divided between all four possible combinations of the two altitudes and two weights of flak. It's fairly straightforward once you have studied it a bit. Each mission consisted of a gruppe of Ju 88s approaching a defended target in one of the configurations. Of the 140 missions roughly a quarter were of each of the four types. All I did was record the number of aircraft damaged both before and after hitting the target, then used an Excel spreadsheet to calculate the average percentage of aeroplanes damaged in each attack.

Unfortunately the maths consist of a bit more than adding all the numbers up and dividing by the number of raids, and take into account such things as sample size and standard deviation, but if anyone is really interested in the details e-mail me.

What the graph shows is the 'range' in which the the average lies in each case. So, if you want to know the amount of damage a gruppe approaching a target at 10,500' is likely to suffer against eight 3" Hvy AA guns you look up the 4th category, and see that the average is given as being in the range of 6.7 - 13.1% That is to say that 95% of the time a gruppe of 30 aircraft will suffer between 2 and 4 aircraft damaged.

[K.Pooley]

Odds & Ends
This set of trials does provide some INTERESTING FACTS, some of which are likely to be less statistically valid than the others (or less interesting come to that)

55-65% of the damage to the gruppen was inflicted on the approach to target.
9% of bomb runs resulted in an 'Instant Kill' in addition to any other damage being scored by the flak. I did not count crashes on the way home or on landing.
75% of these Instant Kills were scored on the approach, although with a sample of just 13 this is almost certainly not statistically valid, but it's always good to get in the habit of being on the lookout for patterns.
0% of the raids suffered >1 Instant Kill.

Past experience, of actual games as well as similar trials, suggests that as the intensity of flak increases, due to more guns, less height or maybe even gunners being more alert, so does the chance of scoring multiple Kills.

NOTE that strictly speaking I should talk in terms of 'damaging hits suffered to an aircraft', and not 'aircraft hit'. That is to say that the trials show the number of damaging hits scored on the aircraft of a gruppe; I have no way of knowing if any particular aircraft took more than one hit, of even how the game system handles multiple hits.

So, after all of that analysis and pretty charts I suppose I ought to get on with actually fighting the war before I get too carried away.

[invernomuto]

Thanks again for the infos and graphs. Very useful for newbie.
Do you use excel or similar to keep track of your game?

[K.Pooley]

They're all from Excel, and I usually use it to keep track of a few things at any one time, to keep track of the key things. In this case a lot of what I've used has been done especially for the AAR. Glad you find it useful.

[K.Pooley]

Timetable
I want to vary my normal routine, of several waves attacking at fairly predictable intervals. I don't want to be so predictable, especially as the game is at such a critical point (I know, I've been saying that for the last couple of weeks).

0600 - All 20 RADAR strikes.
Since I already have a perfect set of recce photos I'm not doing any PR against the RADAR sites at this stage. I also have current photos for all Sector AFs and all but one 11 Group SECAFs, so will be dispensing with all pre-strike recce today. The first wave will be

0610 - Following on almost straightaway will be fighter sweeps against 11 Group SECAFs. Hopefully these will blend a little in with the previous wave of attacks. And just a short time later,

0640 - Bombing raids against the same AFs targeted by the fighter sweeps.

As to the actual targets of these two waves, I have gone for the two highest priority categories as in the graph above, and as ringed on the map below, for a total of eight targets. The raids' altitudes will vary depending on, amongst other things, the amount of flak as previously mentioned.

[K.Pooley]

This RL example is how I imagine (hope) some of my targets are looking about now:

'The devastation at Hawkinge was appalling, hangars, workshops and domestic buildings flattened and others burning. From the water tower sprang a series of fountains caused by bomb splinters and over the station lay a slowly drifting clouding of dust and smoke. Hawkinge was out of commission…'
[Biggin on the Bump, page 51, by Bill Ogley, pub. Froglets Ltd]

[K.Pooley]

Despite having 20 RADAR targets to hit, I managed to use Stukas, Do's and fighter-bombers for most of them, with only some of the east coast targets calling for the use of main-force bombers.

[K.Pooley]

Turn 26

[K.Pooley]

As for the RADAR picture, this just goes to show that you can't win 'em all:





Although it's not bad out of twenty targets!

[K.Pooley]

So, what happened?

[K.Pooley]

It's obvious that by shifting the bulk of my attention from the Sector AFs for just 24 hours has had serious consequences. I thought I had allocated enough resources to keep them 'simmering', but obviously not. As you can see below I have gone back to recce - bomb radar - bomb Sector AFs - recce.

I have tried to strike a balance between reaching the less damaged (and usually more distant) AFs to increase AS directly, hammering individual (usually closer) ones in an attempt to cripple their comms and so degrade response times, and conserving the strength of my own forces.

The turn includes some large, well-escorted, raids, some unescorted and very vulnerable raids, and Raid No.54 on North Weald, which is comprised of 20 '110s of ErprGr 210, plus their 209-strong escort of '109s. If Swift intercepts it (and why wouldn't he) he will at the least get a shock. The staggered timings for this day's missions are intended to make a unified response by FC more difficult, and maybe just throw him off balance a little.