Tuesday, June 21, 2011

Value Over Replacement

VORP is an advanced statistic used in baseball to determine how useful a player is. The basic idea is you create a mythical 'replacement player' and compare how your specific player performed compared to that mythical player. The idea is if that player got abducted by aliens or whatever and you were forced to take your basic shmoe batter how much worse would you be? You can then use this value to determine things like how big a contract you should offer him when his current one expires. If he's barely better than replacement level (or even worse) then you're not going to give him a very big contract. If he's a lot better then you should make sure he doesn't leave for another team!

I use a similar theory in the game Galaxy Legion to justify not defending my planets. There I actually can derive precisely the value of a replacement planet since I know specifically which planet I will take if I lose one of my current planets. The difference in production between the two planets is my VORP. From there you need to then determine how much of that VORP you'd have to give up in order to build the appropriate defenses and also work out the odds of those extra defenses mattering.

As an example let's look at a planet I recently had stolen from me: Cern'. This was a large metallic planet with very rich research, extremely abundant mining, and extremely sparse artifacts. According to my spreadsheet it was worth about 122 production per hour. My best available research planet is worth about 107 production per hour. So the value over replacement for this planet is 15 points per hour. It's about 14% better than replacement. In short this planet was certainly worth having but if I had spent even so much as 6 of the 51 size on defensive buildings it would have been worse. So even if I know someone has scanned the planet, and is willing to attack it, and willing to defend it themselves, it wasn't worth putting more than 3 defensive structures on it. At the time that would have been about 600 extra defense (to go with the passive 400 on the planet) which would have meant someone with 2700 or more attack would have had the maximum invasion chance against me anyway. (My invasion attack without using anything special is near 3000, so that much defense wouldn't deter someone of my strength at all.) In order to actually defend the planet from myself I would need to make it worse than replacement. Which is exactly what the guy who took it from me did, as it is now worse for him than planets I have uncolonized. I was hardly hurt by his invasion and he's actually hurting himself long term by defending the planet from me. He may actually be better off undefending it, letting me take it back if I want it, and picking up something else. But I don't really expect logic from him, so whatever.

I haven't ever defended a planet I didn't conquer from someone and expect immediate retaliation from and I've only lost 4 planets ever from the 46 or so I've colonized. I believe that means your run of the mill, average planet needs no defenses at all, so you can look at any uncolonized planet you've scanned as a potential replacement using full production buildings on it. (After all, if someone actually wanted to take it they could just go colonize it for free right now! Defenses aren't going to deter them. No one else who has scanned it wants it!) So the replacement level for a planet is actually pretty high, I think. It's full undefended value.

In conclusion, before building defensive structures on a planet you should take a look at how much production you're actually giving up to build them. You need to consider just how much of a deterrent they actually are (getting up to 1000, for example, is a lot like nothing at all to me) and what you could take to replace it should you lose it. Unless the planet is pretty special or you have an unhealthy attachment to it beyond the numerical production it provides you should just leave it open. Maybe someone will take it and you'll lose a little bit of production. Probably they won't and you'll profit forever. The alternative, where you defend everything and no one would have attacked anyway, is much worse. You're just hurting your production for no reason.

No comments: