Kata

I liked this kata I have to say, good job!

Rounding is not the correct way to handle floating point representation inaccuracy. Comparing with a margin for error is, but that's impossible if you want some arbitrary alive / dead cut-off.

I know rounding hitpoints is common ( Hell, rounding numbers in general is common ). But the way it's implemented now ( rounding down ) means you are not testing for 0 or less hitpoints for death, but for 1 or more for survival. ( See my solution. Note I don't round. )

If you feel an irresistable urge to round, round up. That way, you will at least test for 0 or less for death. You will also occasionally get protests from people who do calculations in a different order, arrive at some value very close to 0, round wrong ( well, the other way than you ), and end up at a different answer than you. This is always a problem when dealing with floating point values; it has always been and it will always be. Floating point values inherently have some inaccuracy, and doing calculations with them amplifies that inaccuracy.

Please don't log anything in testing. Use headers or let solvers do it themselves. This is confusing.

Please use the most recent Node version ( which also means, use chai ). There is no good reason to use obsolescent versions.

For the title to be correct English, please consider choosing from "Am I a suvivor?", "Have I survived?", "Did I survive?", or "Will I survive?"