It's very hard to give a general answer without knowing what methods are being considered.
"Best for what" is often a good question... As an example, hash tables can be very fast to read best-case, but used badly they can be no better than a linear search. They're also space-inefficient, strewing information thinly by design... You wouldn't want to use them to store huge amounts of data. Trees are slower to read than a hash table's best, but a balanced tree's worst case time is smaller than a poorly hashed table -- but adding to or changing a tree can be complicated and slow since it may need balancing. Not all data is really suitable for either anyway...
I'll say whay has already been said another way - alogrithm choice affects resources in a really bad way when you pick the wrong method for the job. Resources = time, memory, i/o.
There's probably value to inlining a simple equation like that. You could make it an inline function though, and get the benefits of inlining while still encapsulating it. Sort of like a macro, but with the same side-effects as a normal function call.