The problem with OldRoutine is that it was intended to be a short term design that was made so EB had something that worked after the 1.3 update since a large part of the api/bot was rewritten. I attempted to solve a design problem with the last routine where each update with skills, you had to modify the CR to get support.
At the rate this game adds skills and changes the way mechanics works, it wasn't feasible to keep doing that, since everyone wants their skills to be used a different way depending on what their build is. BasicRoutine (what OldRoutine is), I tried out the idea of expressing CR logic using skill slots, so no matter what skill you want to use, it'll always be supported. However, it's a barely implemented design, because you need to code specific behaviors for each skill use, so you can then bind it to a skill slot.
For example, look at Cold Snap. It's a skill that best represents a 'nuke', a high power mini-aoe attack on a lengthy cooldown (which can be bypassed with a power charge). To model using that skills in a generic way, you'd have a "target" from what you'd wan't to assign from a very large list of possibilities. Things that would be like best target rarity is rare+, or there are 3+ mobs in a certain range, or any. Then the condition for each of those, so something like not on cooldown to cast it while its not on cooldown, power charge available to cast it when its on cooldown but a power charge is present, or never to handle more complex cases of ensuring something else gets cast.
Now if you do that at a basic level for all skills and simple mechanics, you have the start of being able to support more complex logic in an easy to configure way, but there's more to it than that. Ideally you need to support custom conditions and provide a way for users to easily import/export them so the core skill logic never changes unless there are bug fixes or improvements.
As part of the new skills, there's starting to be some direct synergy between them more so than in the past. If you have contagion on a mob, and it dies while having essence drain on it, the effects jump to the next mob, which can make for some powerful disease like effects seen in other games once you buff up the skill and duration. This is a direct case for needing logic to check a buff on a mob and cast a skill on it to accomplish a synergy effect, as opposed wanting to just cast it because it's not on cooldown or the buff isn't applied. That is, you don't want to be using these skills as your primary attack skills, you'd have no dps, but for support skills, On the other hand, casting such skill combos takes time, time that can be spent using other skills, so you don't want to always just cast them for the effect if the benefit won't be worth it, if that makes sense.
So that's why Contagion casting was added. I was testing it to consider logic required for handling such things better for a future CR. The current CR has quite a performance penalty simply because it does too much. If I were to keep adding in logic and skills for a lot of things as has been requested, performance from the CR would continue to go down more and more. As a result, the point of the current CR is to have something basic that works for a large number of skills and to serve as a big example for people to come up with their own customized, build optimized routines. The downside of course, is that all the new skills from like the past year aren't really added.