How often will a node be sieged? (Warning: Includes math)
Quick answer: Not as often as you expect.
We know sieges will happen monthly. We know that there will be 5 castles and 103 node locations. We know that servers are planned for either 8,000 or 10,000 concurrent users.
Let's start with concurrent users on a healthy server. Sieges are planned to handle 250 attackers and 250 defenders (500 players) and might grow to 500 vs 500, but let's start with the 250 v 250. IF we say 10,000 users are on the server and 80% participate in the sieges (though both numbers may well be lower), then 8,000 players will be sieging. Is it reasonable to assume that the castle sieges will be the biggest draw? Then five castle sieges drawing up to 500 players each could absorb 2,500 of the sieging players, leaving 5,500 for nodes, which is still a lot. On a higher population server with developed cities, can we then assume that the next siege targets will probably be the five metropolis or largest cities? Those five sieges could draw 500 players each, which is another 2,500 players leaving maybe 2,000 players to fight other node sieges.
If the node is fully developed, there are 103 nodes. Less the five metro/large city nodes being sieged above, that is 98 more nodes split between 2000 sieging players...about 20 players per node to attack and defend! Do you agree we are unlikely to see sieges of 10 vs 10 often? So, of those 98 other nodes, how many will be sieged? Choose your own number, I'll go with one in ten.
Where does this leave us? All five castles sieged, great fun. The five largest nodes sieged, great fun. Ten or a dozen other nodes sieged, great fun. 90 of the 103 nodes not sieged in any given month, which means each node (with the exception of the largest ones) will likely be sieged only once or twice a year. Good diplomacy, competent mayors, healthy guilds, and other factors could decrease that more. Consider further that there is no guarantee a siege will succeed, shall we say that half will, half won't? Therefore, if you have an apartment or freehold in a moderate sized and well governed node, you can hope to go a couple of years without ever losing your housing.
Yes, those calculations required quite a few assumptions on my part. If siege participation is less than 80% or if server population is less than 10,000 concurrent users, then the chances of a node being sieged are even lower. But if sieges end up tending to be smaller than 250 vs 250 players, then the odds of a node being sieged is higher, but the number of attackers lower which might make the static defenses comparatively more effective, decreasing the odds of a losing siege. If the sieges end up being 500 vs 500 (how fun!), the castle sieges will probably pull more players to them leaving less for node sieges, making node sieges even rarer.
******************
The above example was based on the assumption of a high population, fully developed server with all nodes siegable. That will not always be the case. What if we have a server which averages 5,000 concurrent users, a low population server. If we still get 80% siege participation, that is 4,000 participants in the sieges. Again, is it fair to assume that the five castles will be the biggest draw, on average, for the battles? If each castle has a 250 vs 250 siege, that would be 5 x 500 or 2,500 players, leaving just 1,500 to do all the node sieges.
But on a server like that, there will be less developed nodes to siege. How many will there be? My speculation becomes less sure here. Maybe there are four largest nodes, perhaps cities rather than metropolis. Each of those four cities has three vassal towns and each town has three vassal villages, which would be 12 towns and 36 villages. If we assume that the four largest nodes are the prime ones to attack, since that is where the rich loot resides, then those four sieges cannot each be 250 vs 250 (requiring 2,000 players) since only 1,500 players remain. So, the castle nodes fights will likely be less than 250 vs 250 as will the node siege fights. But the point here is that it could turn out to be rare that many of the 12 towns or 36 villages will be sieged in any given month! Again, it could be one in ten, or less, each month. We end up with the same situation on lower population servers than we did on full servers, only castles and large nodes may expect frequent sieges, the small towns and villages may see a siege every six months or less. They will win perhaps half of those, so again a healthy smaller node will seldom suffer defeat.
The larger nodes will be sieged and defeated more often, allowing the smaller nodes to grow and replace them, which is exactly what the game design intends.
Yes, I have made quite a few assumptions and I welcome improvements and other methodologies.
