STRATEGY

==Harris==
by Edward Cree
--Strategy Guide--

Of course strategy in any game is a highly individualistic thing, and I won't attempt to declare The Best Strategy™; in any case, that changes every time anything affects the game balance.  Instead, I will discuss the effect of each independent game feature.  I will, however, end with a broad outline of the strategy I tend to follow.

-Bomber Stats-

Cost: This one is fairly obvious; essentially all other data are inputs to the value of each aircraft type.  Divide this (conceptually; the value probably can't be computed mathematically) by the cost, and build the aircraft type that maximises the result.  Of course, the value changes over time... and in the case of the Wellington, so does the cost.

Speed: There are a couple of ways in which bomber speed affects losses.  Firstly, faster-moving aircraft spend less time within range of a given flaksite, so fewer shots are aimed at them.  Secondly, faster bombers are more likely to avoid fighter interception, or at least force the fighters to chase them, using up fuel, and reduce the time for which the fighters are among the bombers.  Speed also has an effect on accuracy; a bomber's navigational 'drift' is supposed to represent errors due to differences between the winds present and those accounted for, and so the error tends to be proportional to time, at least when the bombers are relying on dead reckoning (e.g. dark night, in cloud, without navaids).  The game also accounts for sunrise/set times, meaning that time over enemy territory is strictly limited especially in summer, so slower aircraft cannot safely raid distant targets.

Altitude (Ceiling): This currently only affects one thing - the range of ground-based navaids like OBOE.  (GEE's range is also altitude-dependent, until the Germans jam it, at which point its range is the same for all aircraft.)  Thus a higher-flying aircraft can bomb or mark precisely targets that a lower-flying aircraft could not.  This is one of the features which makes the Mosquito so well-suited to Pathfinder operations.  [There are also plans for altitude to affect vulnerability to Flak, and perhaps for fighters' maneuvrability to vary with altitude.  But, in that case, altitude would also have to be affected by 'loadness' (that is, the accounting that is already made of fuel and bomb loads at a given point in time).]

Capacity & Bombloads: Before Area Bombing, aircraft bombloads are entirely fungible; two 14,000lb Stirling loads have the exact same effect as seven 4,000lb Hampden loads.  However, when bombing cities, the different types of bombs combine in various interesting ways.  The ideal combo is to open with HC, incendiaries and TIs, then for the remainder of the raid use a GP/incendiary mix tuned to the target's flammability.  This is because the HC will blow in rooves etc., making the buildings more flammable; the fires and TIs will then help the bombers to find the target, and the GPs will generally do the most damage.  Note that as incendiaries are bulky rather than heavy, you can only carry two-thirds as many by weight as you can HE, and the Air Ministry pays you by weight; however, damage done to cities improves the Air Ministry's confidence in you.  Also, while the heavier fuel load for distant targets limits the weight of bombs you can carry to them, it doesn't limit the bulk - so you may be able to carry a full-weight load of incendiaries.

Serviceability: There are really two factors controlling serviceability; this one, which sets the equilibrium level when not operating, and the Failure stat, which affects how much serviceability is reduced by Ops.  This is because a bomber which returns home with mechanical trouble (or, for that matter, with heavy damage) is counted as unserviceable.  The obvious effect of serviceability is that it controls the maximum size of raid you can launch relative to your total number of bombers.  But there is a secondary result of this: if two types have the same loss rate per sortie, the more serviceable type will endure more rapid losses - because they are making more sorties.  Conversely, types with a low serviceability will see much slower changes in their battle-ready numbers, since the front line are buffered by the large numbers of u/s aircraft.

Defence (Survivability): This may well be the most important stat; more important even than capacity.  For not only does defence control how many sorties an aircraft - and its crew! - will (on average) complete before being shot down, it also has significant effects on morale.  On the other hand, the defence stat is definitely not the only factor influencing loss rates.

Failure (Reliability): While 'early returns' tend to concentrate toward the beginning of a raid, they do continue throughout it, so raids on more distant targets are likely to be further reduced in this way.  Curiously, the first two long-range bombers (the Whitley and Manchester) both have a rather high Failure stat, which long-range raids tend to magnify.  As mentioned above, the Failure stat also has an indirect effect on serviceability.  Lastly, it is possible for mechanical failure to down an aircraft, and the probability of a failure resulting in this is also proportional to the Failure stat - so the probability of this happening in a given sortie goes like the square of the stat.  Manchesters, in particular, are hence over three times more likely to suffer this fate than any other type.

Accuracy: The modelling of navigation is fairly complicated, but its outcome is likely to be a roughly Gaussian error; this means that the proportion of bombs on target is likely to be a nonlinear function of the Accuracy stat.  For instance, if the target is large and sprawling (or, as with the Ruhr towns, surrounded by other targets) and the bombing is sufficiently accurate that everyone hits, increasing the accuracy further will gain nothing.  As several other factors also affect navigational errors, the trade-off between accuracy and other stats changes depending on many conditions.

Range: Not only does range produce an absolute restriction on the targets that can be attacked, it can also make raids within that restriction unprofitable, because at the limit of the range, some of the bomb load must be removed and replaced with fuel.  The fuel load can also affect a bomber's maneuvrability and hence evasive action under fighter attack.  The ability to raid more distant targets gives you tactical flexibility, both to deal with the vicissitudes of Met. and to fool the fighters.  However, a deep raid, particularly in the south, means more time over enemy territory and thus more opportunity for the defenders to inflict losses.

Base location: This can have an effect on the routes chosen and whether the aircraft succeed in weaving around the flak or go right over it, until everyone is fitted into a single bomber stream after March '42.  It also affects the range required to hit a target.  (The player is in control of this, through Squadron assignments; you might choose to put short-ranged aircraft in Nos. 2 and 3 Groups so they have less far to go, or you might prefer to put the long-ranged aircraft there so that they can raid the most distant targets, which they might not be able to from (say) No. 6 Group.)

Service dates: When a bomber enters service, you have to buy it at Low for four months.  But you can game this somewhat, by timing other production plans so that as many other types as possible are at High during those four months.  My particular version of this is that in the run-up to the Manchester, I stop building Hampdens (and so build only Wellingtons), then when the Manchester is novel I put both Hampdens and Wellingtons at High, thus minimising the amount of budget 'wasted' on Manchesters.  As for the obsolete dates, it helps to know them, as you obviously want to stop building those types well beforehand.  You don't get your money's worth out of a bomber that gets taken out of service mere weeks after it's built.

Crew: An aircraft which carries a Bomb-aimer will have a slightly higher accuracy than its stats would suggest (more so if he's skilled); an Engineer or Wireless-op contributes to defence in the same way (by providing damage control).  Adding more Gunners doesn't improve defence (the stat is assumed to incorporate that already), though increasing their skill does help against fighters.  Engineers also reduce the failure rate, and improve altitude, speed and fuel efficiency - a highly-skilled Engineer is a useful chap!  Some heavies early on take two pilots; this is mostly just a quick way to deplete your stock of these valuable aircrew.  Pilot skill mainly affects evasive action under fighter attack, but also fills in for the Engineer (at half-strength) if none is present.

Navaids: The only bomber-navaid restrictions that really have any effect is that Wellingtons can't carry OBOE, and Mosquitos can't carry H₂S.  The former is not likely to matter as you want to put OBOE in either a heavy bomber (I usually put the early OBOE sets in my Stirlings, which have a heavy payload and an otherwise poor accuracy) or a Pathfinder.  As for the H₂S, Mosquitos can generally navigate well enough without it, and they're usually given priority for OBOE sets anyway.  However, before December 1942, Mosquitos can't carry OBOE either, which is rather more significant.

Flags:
	NOARM: The unarmed Mosquito relies on speed and evasion, and this works well for it.  However, Me262s could trouble them a little.
	PFF: Being assigned to the PFF chiefly affects an aircraft's tactics; Pathfinders are always the first over the target, increasing the risk from flak but often reducing that from fighters.  Pathfinders also drop skymarkers, which can do a great deal for the accuracy of the Main Force behind them if the target is obscured by cloud.  PFF also generally get the best crews, as joining the Pathfinders requires 15 completed ops (30 for Mosquitoes).  (This isn't actually a bombertype flag any more; the player is in control of what aircraft get assigned to No. 8 Group.)
	HEAVY: These large, lumbering aircraft are easier for fighters to spot, so might want to avoid bright moonlight.  To achieve their full potential, they need concrete runways.
	INC: The Halifax III's 'Usual' load has more incendiaries than most; as these are bulky rather than heavy, they have less effect on ceiling and maneuvrability (and thus loss rate) than a full load of HE.)
	OVLTANK: By carring overload tanks in the bomb bay, Halifaxes can fly extra-long-range missions (albeit with a reduced bomb load).  A mod in May '42 applies this to the Halifax I.
	SLOWGROW: Production of Lancasters ramps up more slowly, meaning you're likely to have to bulk out your forces with other aircraft.

Mods: A bomber's stats (and other data) are not constant; as time passes, various modifications or "mods" are applied.  This is done simultaneously to all aircraft of a given type already in service, and the corresponding change is made to the production line.  Probably the most significant mods are the 'bulbous rudder' which reduces Halifax losses, and the Broughton shadow factory which makes Wellingtons _much_ cheaper to build.  "Marks" are similar, but apply only to new-build aircraft; outdated marks are often a good choice for training units.  All these changes are specified in the file 'dat/mods'.  When a new mark is about to arrive, you might want to hold off building aircraft of that type until you can get the better ones.

-Bomber Summaries-
Blenheim: Its high accuracy and low defence reflect the fact that it should really be a day bomber.  These and its small bombload make it best suited to shipping strikes.  It's also handy for tactical targets in France; or, at least, its limited range means those are almost the only things it can be used for.  Blenheims will probably feature heavily in the Battle of the Barges.  They're also very cheap, so they may be more profitable than they seem - even losing three Blenheims per ship sunk is enough to break even.  However, risking trained pilots and navigators for such a small reward might not be a good idea.
Whitley: The slow and vulnerable Whitley has the advantage of a fairly heavy payload.  Try to conserve losses until CIV (May '40), by using them mainly for minelaying (or even just keeping them off Ops); after this point, they can be quite useful for bombing Ruhr factories and - especially - steelworks.  They will still suffer fairly heavy losses in such raids, though, which can damage morale.  [If we ever implement per-type morale, the Whitley crews will probably be an unhappy lot.]  Although they're fairly cheap, I don't usually build many more beyond the starting force.  Their crew includes a bomb-aimer, which can be a handy way to train up crewmen in that role.  Whitleys get slightly better in August 1940 when the 'Electric Bomb Release' mod improves their accuracy.  Nonetheless, they'll probably quickly find themselves circuit-bashing at OTUs (the Mk IIIs in particular should probably never be used for anything but this).
Hampden: Being fast, maneuvrable, and fairly accurate, it's good for shipping strikes (Hampdens are the _other_ contributor to the Battle of the Barges), as well as attacking bridges in France and the Low Countries.  However, it also does well at some of the harder-to-hit strategic targets; I quite like to use them for oil and chemical works.
Wellington: The main advantage of the Wellington is its good Defence stat; it's also notable for its reliability and serviceability.  Its accuracy however is mediocre.  During the CIV era (ie. May '40 to February '42) they do well by scattering bombs all over the Ruhr, and they can sometimes turn a profit on Berlin raids.  When the "Broughton shadow factory" 'mod' approaches, you probably want to reduce your Wellington production, and then bring it back up when it happens; it doesn't cost you much to wait for those Wellingtons and you get them a good deal cheaper.  Once Area Bombing begins, Wellingtons do a good job in bulk, but the increasing effectiveness of German nightfighters eventually pushes them out.  Still, for the first three years of the war at least they are likely to constitute the backbone of Bomber Command, and even after that your OTUs will rely on them.  By the time the Mk X comes along, it's really only fit for minelaying and other such low-risk tasks.
Manchester: Largely a disappointment, but its good accuracy and long range can prove useful for raiding Berlin, Augsburg or Munich.  However, it's unlikely to turn a profit regardless of how it's used - its loss rate is just too high (especially losing two Pilots with every bomber), and its tendency to break down and turn for home limits the tonnage it can deliver.  They're improved slightly by Crew Rationalisation in April 1941, gaining an Engineer and no longer needing two Pilots.  It's unclear whether the Engine Derating in September of that year improves matters - the failure rate goes down but the reduced performance may increase losses from enemy action - and the payload is reduced.  The main use of Manchesters, however, is to have squadrons ready to rapidly convert to Lancasters when those come along (since they can make the switch in 14 days, instead of the 28 anyone else would need).
Stirling: Personally I've come to quite like the Stirling; it does a surprisingly good job in Area Bombing.  Its limited accuracy does make it unprofitable on precision targets, though; until February '42 it should mix short-range factories with minelaying, and building up the force in readiness for Area Bombing is a good use of funds until they can all be spent on Lancasters.  It's also a good way to train up a cadre of experienced flight-engineers.  Once nightfighters develop, though, it will suffer increasing losses and eventually have to bow out.  Towards the end of this time it can also be handy for diversions.
Halifax I: These seem to be reasonably competent but suffer from a fairly high loss rate.  Successive modifications should eventually cure this, and the Halifaxes can deliver plenty of incendiaries to German cities, until nightfighters make the skies of the Reich too dangerous for first-gen heavies.  The type is also likely to end up forming the backbone of your HCUs.
Lancaster: The Lanc is the Area Bomber par excellence; until nightfighters develop you can't really fail to make a profit with it.  You end up in a race, trying to grow your forces as much as possible - and do plenty of damage - before the nightfighter force becomes too large; towards 1944 things get rather difficult for the Lanc, but with any other 'heavy' they'd be impossible.  Lancasters also provide the heavy lift for the Pathfinders.  The Mk X upgrade doesn't really change its strategy; it's just generally 'a bit better', which is certainly nice.  Its "Plumduff-Plus" bombload includes a 12,000lb HC, which is quite nice too.  I'm fairly confident it's worth the slightly higher price tag.
Mosquito: Pathfinder Mosquitoes equipped with OBOE and dropping TIs are a very effective way of improving bombing accuracy.  But they are also rather efficient bombers in their own right, and an analogue of the Light Night Striking Force - each carrying a 'cookie' to Berlin in all weathers and heedless of the moon cycle - is a tempting option.  The very good Defence stat (helped by the high speed), and the cheap price, make the Mosquito highly economical.  However, delivering a serious tonnage with just Mosquitoes would take a very large and unwieldy force, requiring a great many experienced pilots and navigators (the PFF requires Mosquito crewmen to have completed at least one tour, though the same doesn't apply to Mosquitos in other Groups).
Halifax III: The Mark III Halifax is a reasonably competent second-generation heavy, at least for delivering incendiaries to German cities.  With the Lancaster already around, it can seem rather unnecessary, but there may be some advantage in having a force of dedicated firebombers, and existing Halifax I/II squadrons can convert to it more easily than to Lancs.  More data are needed before it can be definitively declared profitable or not.  [And, of course, its effectiveness on Overlord targets is still one big unknown, as they haven't been implemented yet.]

-Extra-bomber Summaries-
These are the alternate-history aircraft which you can enable when starting a new game.  I haven't done enough playtesting with them to definitively pronounce on their qualities, but here are some educated guesses.
Southampton: Poor defence and quite expensive, but the payload is good and so is the speed.  Fulfils a similar rôle to that of the Stirling.
Elswick: Dross.  Essentially retains all the flaws of the Whitley
Windsor: Probably a bit too expensive, but might outperform the Lanc for some purposes.
Selkirk: Accuracy a bit poor, range limited, but payload great.  If you're still raiding the Ruhr when it comes along, useful for that, but probably unsuited for deeper-penetration raids.

-Target Systems-
Steel and Armament: In almost all circumstances the armament industry will be limited by steel supply rather than capacity, so bombing armament factories doesn't have much effect (other than you getting paid).  Going after the steel industry is an effective way to weaken the flak, and conveniently the Rhine/Ruhr region (where most of it is concentrated) is quite nearby, so you can carry full loads and not have too much trouble with fighters.  On the other hand, it is full of flak, so you have a bit of a chicken-and-egg problem.  I usually prioritise steelworks in the CIV era (1940-41), then make a concerted effort to flatten Ruhr cities in early '43 (which is after all when the historical Battle of the Ruhr took place).
Oil: Early on, this doesn't matter too much, because nightfighters are largely ineffective.  But once they have radar, reducing the Luftwaffe's petrol supplies is an important factor in protecting your bombers.  By late '43 you really don't want them to be operating at full tilt.  It may be a good idea to take out refineries in winter '41-'42 while they're still available as targets, rather than having to wait until you can destroy the entire containing city.
Aircraft and Ball Bearings: Much like the armament industry, these don't usually matter too much, as fighter activity is limited more by fuel than by available aircraft.  However, they do affect how quickly the Germans can build up a force of a new type after developing it.  I haven't really found out yet what kind of effect Me262s have, for instance, but it might be one you want to prevent.
Electronics: Damaging this industry would be really useful - cut down on the number of radar-equipped nightfighters chasing your bombers - but unfortunately the targets are distant and difficult to hit, so it's tricky to achieve.
Rail and Minelaying: Both of these are a useful supplement to attacks on other target systems, because they reduce the rate at which those other targets can be repaired.  (On the other hand, if you send a few bombers every night to the same target, they will prevent it being repaired at all.)  Minelaying is also low-risk, and a handy way of getting high hit-rates to pump up Air Ministry confidence in the early years.
U-boats and Shipping: These are just ways to get more money.  (The in-universe explanation: sinking ships means you can argue the Navy's budget should be given to you instead, while fighting the U-boat menace means more convoys make it across the Atlantic bringing more stuff for all the armed forces.)  Prior to the opening of the strategic offensive in May 1940, shipping strikes can be your main source of income.  The Battle of the Barges is also a profitable opportunity.
Tactical targets: Again, just money at the moment (because simulating the effects of your bombing on the land war is not something I have any idea how to do).  Bridges are high-variance, but distinctly lucrative if you succeed (one bridge will pay for five Hampdens, or ten Blenheims).  And you want to take out Soesterberg Airfield just to get rid of its flak, as it's right in the way of your routes to the Ruhr.
Leaflets: As a way of harming Germany or boosting your budget these are pretty much futile; their main uses are raising Confidence (through high hit-rates) and training aircrew (milling them through low-risk ops to get them onto Instructor duties faster).

-Airfield Building-
The rate at which Stations can be upgraded with paved runways has been calibrated to check the growth of heavy squadrons, and encourage the player to continue building Wellingtons at least into 1943.  Of course you _can_ always operate heavy bombers from grass strips, but this is only likely to be economical for short-range targets.
On the other hand, forces based at unpaved Stations will suffer lower sortie rates during the winter, particularly around January and February, when these bases frequently turn into churned-up seas of freezing mud.  And of course the winter is when you want to be launching your deep-penetration raids, taking advantage of the longer nights.
I generally prefer to spread the concrete between multiple Groups, with their different heavy types, rather than concentrating on building up one area.