I boast a relatively low defensive ranking in W1. One could argue that I have less experience than the average player defending my villages. To counter this, I can honestly say that I've watched teamates defend and when I attack, I see how players defend their villages. I've seen the best players defend against unbelievable odds, and have learned in my opinion how to defend effectively and efficiently.
First, what is the objective of defending? For me, the objective of defending my villages is to slow down an opponent until I can resume the offensive. In the long run, you can NOT defend and win this game. The best outcome you can hope for is a stalemate, and who wants THAT as the best thing that can happen? By defending effectively and efficiently, you maximize your potential to strike back and make gains offesnsively. Threats do not disapear by defending correctly, they only abate when you show agression and your opponent backs off or you make offensive gains toward them.
Like offense, there are three levels of defense in tribalwars. Village level defense, player level defense and tribe level defense. Unlike offense however, the area of focus is at the player level defense. I will examine in detail each of these levels, focusing on the player level defense.
Village level defense
As a relatively advanced player, you should know that there are three primary defensive units in this game. Spears, Swords and Heavy Cavalry (hcav). Just a brief recap just so that everyone is on the same page. Spears defend primarily against cavalry, swords and hcav defend primarily against axemen. I advocate the even defense strategy, that is having an equal number of cavalry and axemen defense. There are strong arguements for a lopsided defense and it IS highly situationally dependant.
10k/10k Spears/Swords Split
This is by far the most efficient setup in terms of farm space. There are two downsides however. First, it takes a very long time to build. To be exact, with a lvl 25 barracks it will take 45 days to build, in this game, that is a long time. Secondly, moving troops around is slow. When villages are clustered together this isn't a big deal, but the further apart villages are, the more problematic this becomes.
8k/2k Spears/hcav Optimized Defense
This is a highly popular model of defense. Essentially, you sacrifice 20% of your troop effectiveness in exchange for a faster build time and faster movement. In contrast to the 10k/10k split, this configuration only takes 17 days to build. This is a THIRD of the time required to build the first configuration.
What should you use?
As a default, I strongly encourage players to use the 8k/2k spears/hcav as a default. Besides the reasons I gave above, it is simply easier to rebuild your defenses after an attack with it. After an attack, being able to re-build your defense 3 times faster is a huge advantage. It is virtually impossible to realistically re-build a 10k/10k defense when fighting good players, you just don't get that much time to recover. There is however a place for the 10k/10k defense. For long term support, or other instances where you just need a village stacked, the 20% fighting effectiveness proves useful, just don't lose those troops or you will have a village completely out of commission.
I strongly caution against deviating from these two standards. I have seen people mix spear/swords/hcav. While I can sort of understand the mentality that a hybrid village can fulfill both situations, in reality, you are much better off choosing one method per village and sticking with it. There are very few instances where the hybrid army will be better than one of the more pure forms of defense. Hybrid villages also make thinking through your defense much more complicated, something that you don't want to do when you have 50-2000 attacks incoming.
Player level Defense
There are two primary ways of defending your villages effectively and efficiently. Both are extremely effective, but can prove disasterous if use improperly. Which system you use depends on your answer to this question: Can I hold off my opponent through defensive strength alone? This does NOT mean, "Can I destroy my oppenent's offense?" Offensive troops build much faster than defensive troops, so the concept of troop preservation should be paramount. If the answer is yes to the above question, then your method should be Stack. If your answer is no, then your method should be Dodge/Snipe.
This is a pretty simple tactic to explain and use. Simply put, you put as many troops as you can in a village before an attack lands. The more troops you can put into a village, the more you minimize your troop losses. A W1 player named Undo stated that it wasn't worth defending a village with less than 30k/30k worth of spears/swords or the spear/hcav equivolent. While this is a decent guideline, depending on what world you are on/what your situation is, this number could vary from 10k/10k to 200k/200k (I've personally seen both of those extremes in the past 3 months). Essentially, you are trying to kill enough of your opponent's offense so that you can re-build your defense quicker than they can re-build their offense. This tactic works well when you know an attack is coming, or it is launched from a long distance.
The only way that you lose a village is by your opponent nobling it. Lets face it, resources in a village are cheap, walls are cheap. Losing an army or a village is not cheap, both in terms of resources and time. Preserving those two, an army/village are the important things. Many times when you get attacked from close range, or when you are caught low on defense, it is impossible to effectively stack your villages.
The first step is to clean your village. Minimize your losses. Remove your troops before an incoming attack. Send your resources to one of your other villages. What do you lose by doing this? You lose your wall and you lose any remaining resources in the village, small price if this allows you to stop the attack. Very few players disguise their nobles effectively. First, nobles have a distinct time signature because no unit travels at the same speed as the noble. Secondly, most people use noble trains, four nobles sent in quick succession from a single village. By identifying incoming nobles, you can pick out the attacks that you MUST defend against. Beyond this, remember that you can take up to 3 noble hits before you lose the village.
You can't escort a noble with more than one army, and rarely do people escort with more than a handful of troops. By preserving your troops from full armies, you give yourself a chance to take out noble trains. This is obviously a temporary fix since your opponent can simply resend their armies and nobles later, but it buys you time, time that you can use to find defensive troops (yours or teamates) to stack your village with.
A slight adaptation of the noble snipe is quite possibly one of the most effective ways to stop an attack. Instead of trying to stop your opponent from taking your village, you help him. The single biggest bottleneck offensively are the number of nobles available. Killing a noble costs your opponent 1 bundle. Allowing your opponent to take your village and taking the village back costs your opponent the full cost of that noble (anywhere from 1 to several hundred bundles). Usually, what you will do is clear out your village and just before their noble train arrives hit the village with your own nobles, lowering the loyalty as close to zero as possible without taking it (usually 3 nobles). Then, timing a noble to land just after your opponent's noble train. By doing this, you will use one noble to re-take the village, but hopefully this will cost your opponent up to four FULL nobles. For a large player, this could be a couple thousand bundles worth of resources wasted.
There are obvious disadvantages to this strategy. First off, you waste a noble, in other words its costly. Second, you lose any troops that you have from that village, so it is usually best to do this with a village that has already lost its troops (like an offensive village that you have just attacked with.)
Ideally, when defending, you should be using a combination of all of these tactics to hurt your opponent enough so that they will stop attacking. The key here is defend in full, or don't defend at all. Do NOT spread your troops out thin.
By forcing your opponent to take army and noble losses, you force them to delay any further attacks on you by several days or several weeks, depending on how effective your strategies are. I can't stress enough how playing effective defense is not enough. Once again, all you are doing is delaying your opponent. The only way to really secure yourself against them is to go on the offensive against them.
Tribe Level Defense
I've heard it said many times, "I'm a defensive player". In this game, there is no such thing as a successful defensive player outside of extremely strong, highly organized and structured tribes. The only tribe that has implemented what I'm going to cite is T:V (to my knowledge) and was pioneered by me.
It is highly inefficient to dodge/snipe between accounts in this game, it is really only a player level tactic, not a tribe level tactic. Thus, the primary function of the tribe should be to stack vulnerable villages that are essential to holding. This is again, the all or none philosophy. When you defend a village, defend it strong. I never support someone unless I can ensure that at least 20k/20k worth of defense will be in that village. If there isn't, either I don't send support or I send the full 20k/20k myself.
Who should be supporting? Anyone who is not under attack or is unlikely to be attacked. Caution is good in this game, but there is a limit. Long ranged attacks can easily be sniped, so what is the point in keeping excessive ammounts of defense in untouchable villages. Typically, at any given moment, I have 90% of my defense out defending other people that I know are likely to be attacked today. Of course there is always a reserve number of troops, just in case something unexpected happens, but in my experience, you can predict which villages are likely to come under serious attack, and which ones aren't. The key for a tribe is cordinating the movement of troops from their strong points to their borders with enemy tribes.
To me, a teamate or an ally losing a village is as big a problem as ME losing a village. First off, it hurts the tribe's interested, but also, more selfishly, teamates losing villages puts you in a more dangerous position. I have zero qualms losing troops in teamate's villages instead of my own. The fact that my teamates know this also means that if I need support anywhere, they will do what they can to help me quickly, no matter how much defense I need. It is simply good business to protect the interests of the tribe as a whole.