Occam’s razor is often misstated as “the simplest answer is the correct one,” but it should more accurately be “the simplest answer is the best starting point to investigate.”
The idea is that the more different variables or assumptions have to add up to get to a solution, the more difficult it is to investigate, and the less likely it is to occur in general. “Entities should not be multiplied beyond necessity.” is the classical way to state it.
So the classic example is: you hear hoofbeats outside, is it a horse or a zebra? Well unless you live in the African savannah, it’s very unlikely to be a zebra. We’d need more assumptions to get there – a zebra was imported to a local zoo, it escaped captivity, and now it’s running amok. Whereas a horse requires just one assumption – a horse is nearby. That doesn’t mean that it cannot be a zebra, it just means that you should start at “it’s probably a horse” and investigate from there.
I had a fun moment the other day, when I went to my kitchen and saw a jar of pickles left out on the counter. I knew it wasn’t me, which left two possibilities that my brain somehow jumped to:
- A burglar broke in, stole several other items, and also ate a pickle. He left the jar out to taunt me.
- My wife had a pickle and then forgot to put away the jar.
I could have totally checked my locks, made sure my valuables were still in the right place, etc. Instead I just yelled “Hey, did you leave this pickle jar out?” and got the simpler answer right away. Starting with the simpler solution (fewer assumptions than my burglar story) got me to the right answer efficiently.