Skeptics mods, can you take a look at my answer here?

I'm being accused of not answering the question and not being objective, even though the data is fully cited, based off NOAA data and excellent research on past storm damage

I believe there's bias against my answer because it goes against the accepted dogma (climate change = more storms), so I'd like to request an objective review

You can't judge a hurricane just by windspeed

nhc.noaa.gov/aboutsshws.php

> The Saffir-Simpson Hurricane Wind Scale estimates potential property damage. While all hurricanes produce life-threatening winds, hurricanes rated Category 3 and higher are known as major hurricanes*.

> Major hurricanes can cause devastating to catastrophic wind damage and significant loss of life simply due to the strength of their winds. Hurricanes of all categories can produce deadly storm surge, rain-induced floods, and tornadoes. These hazards require people to take protective action, including evacuating from areas vulnerable to storm surge.

@JoeW sure but I also added a graph showing hurricane damages since 1901

if you don't trust the hurricanes classifications, you can look at the damages chart

And your graph shows major hurricane numbers going up

if a hurricane or storm didn't cause damage, then nobody cares about it. Tons of hurricanes just pass over the ocean and don't damage anything on land.

@JoeW no, I did a linear regression to prove its not statistically significant

the decrease in the total number of hurricanes is not statistically significant either

Just pointing out that the windspeed isn't the only thing to consider when looking at how bad they are

Again, I am just saying what your graph reads to me

That the number of major hurricanes is increasing

@JoeW right and that's why linear regression is used by scientists to prove or disprove trends as statistically meaningful

for temperature data linear regression shows a VERY clear trend

@JoeW and again I'm explaining why this is misleading

You don't need to keep explaining the answer to me, I am just stating what I see when I read it

OK, so the answer is fine?

@JonathanReez Again, I am not here to debate your answer just explain what I see in it

Yes, I get that, so should I add a section explaining what linear regression is?

since it seems like people are not familiar with statistical significance in science

That value is still going up over time

OK but you do understand that if (say) there was no climate change, we'd still expect random variation in the number of hurricanes?

if we could chart them all the way back to 5000 years ago you'd randomly see lots of fluctuation

for temperature no such ambiguity exists, we NEVER had temps as high in the past 5000 years

xkcd.com/1732 - as shown famously by XKCD

similarly total number of hurricanes is going down, but I'm not claiming that's meaningful

I understand that the graph starts at under 5 and grows to around 7 and we only get a few major hurricanes each year so that isn't just a minor increase

there's also models showing LESS hurricanes projected into the future due to increased temps, but the question didn't ask about those

and despite that 1851-1900 was quite stormy

again, I'm 100% convinced antropogenic climate change is real, and will cause problems for many countries, especially places like Florida which are very low situated. But that doesn't mean ALL climate phenomena will become worse

some will actually become better

Florida will be underwater in 100 years, but might actually see less hurricanes

You can argue what your answer says all you want, but that won't change what I or others when reading it.

OK, that's fine, I'll let people ignore linear regression if they want

And I am saying that is what shows me that it is increasing

yep, I understand that, you're free to ignore Stats 101

that's fine and nothing I can do about it

Republicans in the US also ignore Stats 101 all the time to deny climate change, that's fine too, nothing I can do about it

That second graph sure looks like the number of hurricanes are decreasing and the number of major hurricanes are increasing based on the liner regression lines decreasing/increasing

@JoeW based off your StackOverflow profile you're a software engineer. You can write some Python code to simulate a random process in your spare time and you'll see why eyeballing data is not a good solution

you will see that zooming in on various intervals of a random graph can produce wildly misleading conclusions, even though the underlying function has no real trendline

I can also recommend a good Stats 101 book if you're interested

@JonathanReez Again your regression line is higher at the end of the graph then it is at the start of the graph

you need to look at the R^2 and p-values

Again, that regression line is higher at the end of the graph then it is at the start of the graph

You are talking about a difference of a couple more hurricanes each year slowly over time

No, the point of the exercise is to compute the R^2 and p-value

for global temperatures the p-value is <0.01 and R^2 is above 0.8

You can't just tell people to learn stats and expect them to treat your answer differently

@JoeW correct, which is why we can't make any statements about the impact of climate change on hurricanes as of this time. We can do some weather modeling for the future, but that was outside of the scope of the question

@JoeW I only care about the answer being factually accurate

I can't force people to learn Stats 101

Again, I am just pointing out what I see with the answer, telling me to learn stats isn't going to change how people view your answer

@JoeW that's fine, I'm okay with people not understanding what p-values mean

I'll update it later to add a section on the p-values for global temperatures, might help a bit

Then why are you complaining about how people are reacting to your answer and accusing it of being political?

@JoeW my ask was to see if something in my answer is not factual or wrong

what you're saying is "your answer is fine, but people don't understand what R^2 and p-values are"

I deleted an older version of my answer which falsely blamed the increase in damages on shifts in population structure without good proof, I was able to find a paper that took that into account for damages and produce a much better answer that doesn't rely on my own reasoning

No what I am saying is your answer isn't fine because you say it isn't while your data shows that they are increasing

my answer computes the R^2 and p-values

what its missing is a better explanation of what it means, as I assumed everyone gets that on Skeptics.SE

I'll work on it, adding a graph for global temps, computing the R^2/p-value for that, then explaining why its important

so yes, your feedback is helpful because it tells me I need to provide more Stats 101 context

otherwise my advice on Stats 101 to you was just a recommendation, understanding Stats 101 is very helpful for a Skeptic and for analyzing claims

@JoeW ok thank you for the feedback, updated the answer

I will look at it again later