The number of tornadoes per year have been increasing in US since 1990s

Investment Analysis Clubs Macro Economic Trends and Risks
1

https://data.usatoday.com/tornado-archive/united-states/

A history of twisters: Tornadoes in the United States since 1950

This interactive map pinpoints where a cyclone touched down and traces its path of destruction.

The National Oceanic and Atmospheric Administration has been tracking tornadoes for decades. This interactive map, which contains data from January 1950 to May 31, 2026, pinpoints where a cyclone touched down and traces its path of destruction. For more recent tornadoes, clicking deeper provides more details, damage estimates and whether someone was injured or killed in the storm.

3 Likes
2

From NOAA:

With increased National Doppler radar coverage, increasing population, and greater attention to tornado reporting, there has been an increase in the number of tornado reports over the past several decades. This can create a misleading appearance of an increasing trend in tornado frequency…

The bar charts below indicate there has been little trend in the frequency of the stronger tornadoes over the past 55 years.

Screenshot 2026-08-30 at 7.04.38 AM
Screenshot 2026-08-30 at 7.04.38 AM694Ă—348 37.9 KB

DB2

3 Likes
3

There is more autism now than before vaccines, therefore vaccines create autism.

There are more asteroids now than before telescopes, therefore telescopes create asteroids

Polio spikes during the summer. Ice cream sales spike during the summer. Therefore ice cream creates polio.

Yeah, you have to be careful about research, but not too careful.

7 Likes
4

But your chart does not show the last 5 years from 2020 to 2026.

4 Likes
5

Here are the recent numbers…

2019   36
2020   24
2021   24
2022   24
2023   31
2024   47
2025   43
2026   11

DB2

1 Like
6

2025 was 44

2026 is only 5 months of data

7

You are correct.

Also correct, although the tornado season does peak back in May.

DB2

8

Not always. I am sure you can pick out several in June, July, August and so on.

2 Likes
9

Here’s the monthly distribution:

Screenshot 2026-08-31 at 2.24.39 PM
Screenshot 2026-08-31 at 2.24.39 PM1670Ă—534 47.4 KB

DB2

1 Like
10

Wrong year - generic distributions are changing.

11

That is the monthly distribution based on the last 74K tornadoes.

DB2

1 Like
12

We will see at the end of 2026 the real tornado count for all of 2026. My prediction is that the count will be much higher than 11.

1 Like
13

This is a classic example of how data can be misleading without the right context. On the surface, the raw number of tornado reports has gone up, which seems alarming. But as NOAA points out, that increase is largely due to better detection tools like Doppler radar, a growing population that reports more storms, and more attention given to severe weather events. The actual frequency of the really dangerous, strong tornadoes hasn’t changed much over the past 55 years.

It’s a great reminder to always look at the underlying methodology and the full timeline when you see a trend. The exchange about the last five years is a valid point, though. A trend that held steady for decades could still be changing in the most recent years, and it’s worth keeping an eye on that.

This kind of careful analysis is also crucial when you’re optimizing a pay-per-call campaign for conversion. You can look at raw call volume and think you’re doing great, but if you’re not tracking the quality of those calls—the actual conversion rates, the caller intent, the source attribution—you could be wasting your budget. It’s the same principle: you need to dig into the details and understand what’s really driving the numbers, not just take the surface-level data at face value.

5 Likes