Good CORs Bust!

[ereblog]

Ding COR32!

Part of why I am leveling Corsair is to prove to everyone how to play the job. I had a pretty good idea beforehand, but now that I’m actually playing it and having to make the hard decisions, I’m getting a lot more insight.

This is my first COR-based mathy entry, about how busting on rolls is actually a good thing.

Comments

[brimah]

Remember when I had to teach you how to play COR? I remember that.

[ereblog]

I remember it as a lot of discussions, mostly about tier 2 merits. But I don't have any particular problem taking various technical pointers from you. You leveled the job a bit before I did.

[brimah]

Each pointer will cost you a cookie though. And it was more of a joke :p

[anonymous]

First off, love the blog.

You are right, corsairs that play overly safe only do the parties a disservice in the long run. Not only do they settle for sub par buffs, they are skipping the entire fun of the job which (at least for me) is variety the anticipation of whether or not I'm going to land that heavenly 11 right before a big fight. Sometimes, when I find myself getting overly nerdy, I find it useful to plot the roll values overlaying the expectation values for a double up. It can give some useful insight into when a double up is generally a good idea. Of course this mostly goes out the window once merits are taken into account, since snake eye can mitigate the risk of doubling up on bad rolls down to almost nothing and fold gives more room to take chances without having to worry about busting twice in a row. - Pyrefox

[anonymous]

Nice entry, but I had hoped to see more equations and such in here. Especially when you wrote "This is my first COR-based mathy entry".

I mean, your explanation on getting off the unlucky numbers was quite descriptive, but I think numbers would have made that clearer. Like "proving" that doubling-up would give 8.17% more attack on average than if you didn't do so.

Or, on the other hand, showing that the difference between doubling-up on a 6 or not is all but really relevant in the case of Choas Roll: Doubling-up has an EV of 22.5% without a DRK, while failing to do so gives 22%. Just so people don't feel like they're teh uber n00bz, just because they don't double-up Chaos Roll on a 6. ;-)

[anonymous]

Sure, in that case there is only a .5% increase for the EV, but if you don't risk it then you know you are guaranteed to never get a roll of 41% so you have to just do what you feel is right for the circumstance. In a party, I would definitely go for it; buffing for a HNM fight, I woudln't unless there was time to fold and try again. Again this is mostly a moot point once snake eye is taken into consideration. Even simply saving it for getting off of an unlucky bumps the EV from 22.5% to almost 25% since you effectively eliminate the 13%(VIII) value and replace it with another 29% (IX). Using it to push the roll up higher on IX and X as well even further widens the gap. Considering, that when fully merited, snake eye has only a 5 minute recast, you will be able to do this on every roll cycle if you so choose giving you a much greater deal of control over the rolls you produce. Playing smart with chance will give a greater benefit in the long run. My point is, if I had wanted to play a job where my buffs were reliable but

[ereblog]

I generally try to cut down on the math because I fear it doesn't interest people. But since I'm getting feedback to the contrary, I'll consider taking things a little more in-depth next time.

[hex_16]

As a poker player, and someone who's studied the math side of that game, including Expected Value (EV), I had thought to make a cheatsheet for the wiki regarding whether or not to double-up on various roll values (similar cheatsheets are actually available for blackjack at casinos). I think the reason I haven't really done it is because I haven't put in the time leveling Corsair.

Thank you for posting your insights. I'll have to keep them in mind (and give you proper credit) if I ever go forward with that project.