Hacker News new | past | comments | ask | show | jobs | submit login

Really? You're looking at the graph, showing a clear and frankly ominous anomaly, especially in the context of the clear wording around "the oceans should be warmest in March, not August", and this is your response? Well done.



Right?

You'd hope that the kind of audience that this site has would be able to read and understand graphs and statistics. But seems like nope.

Imagine looking at e.g. some service RPC logs and seeing P95 latencies off the charts during certain peak hours of the day but then just taking the average of the whole day and being like "only 10ms higher than the day before so it's fine"

I mean, I failed high school math but have since had to learn enough stats in my career to understand the consequences of ignoring variance...


Anomaly (your word) != trend . It only becomes a trend when it happens more than once, by definition.




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: