Replying to @realshannonwong
I have a 42 that we are planning to fly around the world next year. Would love to connect.
3
12
1,401
Weighted CASM at ORD:
• UA: 11.19¢/ASM
• AA: 11.73¢/ASM
AA has a 4.6% per-seat-mile cost handicap before revenue enters.
Now stack revenue and cost together:
1
186
UA runs 86% of ORD passengers on mainline metal. AA runs 76%. Convert to departures and it's ~69% vs ~58% mainline.
Now apply Form 41 CASM by aircraft type:
1
203
COST SIDE. Regional jets cost more per seat-mile than mainline (fewer seats spread same crew/handling). So the mainline-vs-regional mix matters more than the paint job.
DB1B records the operator behind each ticket:
1
173
UA yields 26.6¢ per revenue passenger-mile. AA yields 24.0¢. That's +10.5% on nearly identical trip length. UA collects ~$20 more per passenger on essentially the same trip.
Obvious question: same routes head-to-head?
1
209
REVENUE SIDE. Pulled 2025 Q2 DB1B (2.1 GB, 8.5M rows), filtered to 203k ORD-origin domestic tickets, attributed to the ticketing carrier so regional flying (Envoy, Republic, SkyWest) rolls up to AA and UA correctly.
1
1
214
Every article I read on the American vs United fight at Chicago O'Hare was surface-level. UA's Scott Kirby says AA loses ~$1B/yr at ORD, AA denies it, no reporter checks. I had @claudeai pull the data and do the math🧵
Replying to @Grady_Booch @nvidia
I have a couple or A100 heat sinks. I got Jensen to sign one…..
2
455
Amateur private pilot here who has done it in an A320 full motion sim with pic to prove it. Air speed to the middle of the white arc, descend 700’ per minute…
Replying to @camolNFT
I know this is engagement bait but you’ve activated my autistic interest card!
I’ve been taking lessons on small craft, and love a good flight sim, but anyone saying yes here is insane.
If there was no other experienced people on board I’d try but the odds are baaaad.
With ATC or a trainer pilot on the comms, most student or novice pilots could probably understand the instruments enough to program and activate the autoland for a safe landing (probably) but that assumes you’re landing at an airport that supports the system, outside of major US airports most airports don’t have the system.
It also assumes no bad wind variability, a runway that can clear in time, and being able to go through all the steps in time.
Someone who isn’t familiar with a 737 cockpit or flight terminology is going to have a very rough time getting that configured even if it’s avail.
But if you have to actually manually land…good luck.
ATC risk analysts have modelled stats on this for a 747.
Civilian with no flight experience, and ATC guidance has an estimated 1% chance of survival.
A small craft student pilot, around 10%
A licensed amateur small craft pilot, around 15%
An experienced professional small craft pilot, still only 40%.
The civilian will likely pull on the yoke, feel like nothing is happening, pull again, then the plane will pitch up hard, they’ll over correct on the push and that’s it. Once that starts, even professional pilots probably can’t get you out of the spin (basically what happened to Aeroflot Flight 593)
For a student pilot, you can have ATC be your systems manager, which is good, but you’d have to deal with response lag, not knowing when to flare, and no way to gauge runway speed.
Plus on a Boeing with the jets under the wings, adding power pushes the bottom forward causing the nose to pitch up, cutting pitches down
On small craft we’d add power to try and stay airborne, but you’d find yourself in a rapid pitch-up and stall… with that size and weight of craft I don’t know how it works… I know a small craft can with enough altitude and time, escape a stall and spin. But I don’t even know if an experienced pilot would want to do that in a large craft.
So yeah, even if you’ve done a ton of flight sim, still probably cooked!
302
Replying to @antoniogm
Passing through Iowa City yesterday, as good or better than any rib eye I’ve had in NYC or SF
199
Replying to @TMFAssociates
Sitting on a plane with my phone and Gemini…. Converted the chart into % growth per day to normalize for uneven reporting periods. Growth might have slowed but Q4 2025 was significant…
1
4
1,383
Strange world where investors are selling the sticky consumer subscription product to buy a testing company… I guess the robots need to test their chips and already know Spanish…
Instead of NYC gov run grocery stores how about TJs runs the MTA?
TJs in BKLYN b4 storm. Line curve almost the entire store. 32 checkout lanes running full speed. Assuming 3min/co and 100$ avg tkt: 10$/s revenue and perhaps 1$/s profit? @NYCMayor @mattyglesias @Noahpinion
211
























