Quote:
Originally Posted by Appocomaster
I've tidied up the thread a bit to remove unnecessary conversation and tried to keep the slightly more civil comments.
My only comments to make about the 90% rule:
1) Do scanners scale linearly with alliance size? My initial feeling was no, but I guess alliance of 20 having 2 hardcore scanners vs alliance of 60 having 6 hardcore scanners... It may be that it's right, but it feels like once you have 2 you probably won't end up with many more than 2-3 smaller scanner / covert op planets. I'm sure those with more history can comment on this better, though.
2) I was going to comment on the maths itself but I misread a couple of times. The only thing I would say is that the formula doesn't really cover the <=10 condition very well unless you take ceil(#planets * 0.9) - which is a fair enough approach but usually not the one Planetarion takes in this sort of situation. If we amend the formula to include this, then I would agree with the formula being viable.
It's certainly an interesting approach, but I would say it actually makes it a much more interesting decision about whether to have more than 40 people or not, as adding people lower than your average actually harms the alliance score. With this being the case, I can see towards the end of the round people will kick people from their alliance so they get down to the 40 or so people with the best scores - I assume scores are linear enough that dropping below 40 will cause a score decrease.
With this being the case, I'd probably choose n=50 and keep the actual limit at 60 if I were to go with this. Seem reasonable?
|
Thank you for cleaning up the "irrelevant" posts.
I have played a scanner a few times and it really comes down to what timezone they are and when they can be online. If you get 2 good North American/Aussie scanners you are fine. Or a European that is a night owl and awake at that time. Generally when I HC I try to make sure at least 1 scanner is around at TP and 2 more during the prime incoming time period with maybe another 2 that take their time becoming a scanner. An alliance can function on 2 scanners but you probably want 3/4 to ensure you have coverage or if someone cant be around for a day.
What I like about the formula is that the potential exists for more alliances to go for the win. It might not be perfect for a tag that is less than 20, but in those cases the alliance in question are not really going for ally tag win but probably aiming for top planet(s) or best average score/value/roid ranking or some other goal.
In the current round there are a lot of alliance that have the chance to win and its going to come down to the last tick. This is pretty rare and usually its a 2 legged race. Having more endings like the current round I think is more feasible with mathematician suggested formula (with some tweaks).
Alliances can kick member to go down to 40 in order to increase their chance of winning. That is an option and could lead to abuse, however we currently have that scenario where alliances hide out of tag planets to add last minute to slingshot into the win. I dont think it will make much of a difference. Would be interesting to take the current data and use the formula and see how much of a difference kicking down to 40 would make.
I understand where you are coming from with the 50 limit over the 40, but I think going up to 50 reduces the potential alliances able to go for the win. Right now norse with 35 has a very good chance of winning the round. With a 50 counting limit they would be out of the race (I think).