Here’s what I found when analyzing **6.3M U.S. flight departures (2024–2025)**:
✈️ **Southwest:** After 3pm, 40% of flights leave late
✈️ **American (IAH):** If you’re delayed, expect ~25 extra minutes
✈️ **Weather:** Only ~5% of delays — it’s usually the airlines
✈️ **LAX:** Consistently among the best for on-time departures (wow!)
* Downloaded raw BTS departure data
* Used **ChatGPT** as my “data engineer” to clean, compile & pivot
* Designed & built the viz in **Tableau**
* Polished the layout in **Figma**
deffjams09 on
But this is data for only these 6 airports? This list of airports exclude other major airports (and even major hubs).
EmptyForest5 on
20% = one in FIVE flights.
Delta builds in an expected 15 delay – they probably all do. Makes total sense to me.
Wirax-402 on
I don’t really know how to react to this, but it’s a little all over the place.
Are these just mainline flights? Or do they include regional airlines?
American also has a pretty tiny presence at IAH. The vast majority of the flights are operated by a regional carrier and only go to Dallas. It makes sense that if there is weather in Dallas, these would be among the first to get cut.
Most egregiously from a quick glance though is the delay data. You state that only 5% of delays are due to weather. However “NAS Alert”, and late arriving aircraft are also primarily caused by weather. For example, if I have a plane that’s scheduled to fly from City A to cities B, then C, and D on a particular day and it gets delayed in A waiting to go to B due to a line of weather that’s limiting the amount of traffic that ATC can handle it’ll be coded as a NAS alert. The next two flights will be delayed and coded as late arriving aircraft. So you’ll have 3 total delays and 0 of which will be officially caused by weather even though they’re all due to weather.
Leave A Reply
Du musst angemeldet sein, um einen Kommentar abzugeben.
![[OC] Ein von drei Flügen in den USA lässt 15+ Minuten zu spät (2024–2025)](https://www.bytesde.com/wp-content/uploads/2025/09/02s4tvasrrnf1-1536x864.png "[OC] Ein von drei Flügen in den USA lässt 15+ Minuten zu spät (2024–2025)")
4 Kommentare
Here’s what I found when analyzing **6.3M U.S. flight departures (2024–2025)**:
✈️ **Southwest:** After 3pm, 40% of flights leave late
✈️ **American (IAH):** If you’re delayed, expect ~25 extra minutes
✈️ **Weather:** Only ~5% of delays — it’s usually the airlines
✈️ **LAX:** Consistently among the best for on-time departures (wow!)
📊 [Tableau Public Link](https://public.tableau.com/app/profile/data.enchilada/viz/TMAIWhyYourFlightDidntLeaveonTime20242025/Dashboard1)
📂 **Data Source:** [U.S. Bureau of Transportation Statistics – TranStats](https://www.transtats.bts.gov/ONTIME/Departures.aspx)
**How I built this:**
* Downloaded raw BTS departure data
* Used **ChatGPT** as my “data engineer” to clean, compile & pivot
* Designed & built the viz in **Tableau**
* Polished the layout in **Figma**
But this is data for only these 6 airports? This list of airports exclude other major airports (and even major hubs).
20% = one in FIVE flights.
Delta builds in an expected 15 delay – they probably all do. Makes total sense to me.
I don’t really know how to react to this, but it’s a little all over the place.
Are these just mainline flights? Or do they include regional airlines?
American also has a pretty tiny presence at IAH. The vast majority of the flights are operated by a regional carrier and only go to Dallas. It makes sense that if there is weather in Dallas, these would be among the first to get cut.
Most egregiously from a quick glance though is the delay data. You state that only 5% of delays are due to weather. However “NAS Alert”, and late arriving aircraft are also primarily caused by weather. For example, if I have a plane that’s scheduled to fly from City A to cities B, then C, and D on a particular day and it gets delayed in A waiting to go to B due to a line of weather that’s limiting the amount of traffic that ATC can handle it’ll be coded as a NAS alert. The next two flights will be delayed and coded as late arriving aircraft. So you’ll have 3 total delays and 0 of which will be officially caused by weather even though they’re all due to weather.