Allegiant’s History Of Scary Midair Mechanical Failures
November 3, 2016 4:31 PM   Subscribe

Thousands of people flew Allegiant last year thinking their planes wouldn’t fail in the air. They were wrong. All major airlines break down once in awhile. But none of them break down in midair more often than Allegiant. A Tampa Bay Times investigation — which included a first-of-its kind analysis of federal aviation records — has found that the budget carrier’s planes are four times as likely to fail during flight as those operated by other major U.S. airlines.
"The FAA requires airlines to fill out forms called service difficulty reports when planes have serious breakdowns. Carriers also have to file monthly 'mechanical interruption summary reports' — logs of failures that cause delays, diversions or cancellations. But the federal government doesn’t regularly check these documents for accuracy or completion, and it leaves it up to airlines to store them.

As a matter of policy, the FAA also doesn’t compare airlines’ records to search for warning signs. Instead, FAA inspectors examine each airline independent of the others. It’s a policy that has continued even though a majority of the FAA’s own inspectors say it should be changed."


Via
posted by cynical pinnacle (34 comments total) 12 users marked this as a favorite
 
tl;dr; Allegiant's fleet is almost entirely composed of MD-83s, and MD-88s, which are nearing the end of their useful lives. That portion of the fleet is due to be replaced over the next 4 years with an order of new A320s. The two other airlines that still fly those types have better numbers, but the sample size is small.

There's a lot of stuff in the article that's cause for concern. I won't be flying on Allegiant any time soon. But... I'm not sure how alarmed we should be by the sensational headline.
posted by schmod at 4:44 PM on November 3, 2016 [5 favorites]


It's interesting that according to the bar chart in the article, Southwest and Spirit have the fewest unexpected landings per 10,000 flights of US airlines (and Allegiant have the most). So being a low-cost airline does not automatically mean maintenance problems.
posted by Triplanetary at 4:56 PM on November 3, 2016 [2 favorites]


I'll be sticking with Baltia, thank you very much.
posted by Stonestock Relentless at 5:00 PM on November 3, 2016 [41 favorites]


Forty-two of Allegiant’s 86 planes broke down in mid-flight at least once in 2015

Sample size or no, that sounds terrible.
posted by zippy at 5:00 PM on November 3, 2016 [1 favorite]


I flew Allegiant a few weeks ago and now I'm feeling very lucky.
posted by Pope Guilty at 5:16 PM on November 3, 2016


I flew Allegiant a few weeks ago and now I'm feeling very lucky.

Samesies. And Pittsburgh is like the opposite of a hub so not very many carriers fly out of here--Allegiant is one.
posted by soren_lorensen at 5:20 PM on November 3, 2016


It's interesting that according to the bar chart in the article, Southwest and Spirit have the fewest unexpected landings per 10,000 flights of US airlines (and Allegiant have the most). So being a low-cost airline does not automatically mean maintenance problems.

Could that just be the phenomenon that small sample size makes extreme results more likely? You know, if the actual probability of needing to make an unexpected landing was the same for everyone, it would be the smallest airlines that had the lowest and the highest sample rates, just because they'd had fewer flights on which to regress to the mean. (This phenomenon must have a name...)

I admit, the bar chart doesn't really look like a case of that, though.
posted by finka at 5:43 PM on November 3, 2016 [2 favorites]


Obligatory Bob Newhart
posted by flabdablet at 6:16 PM on November 3, 2016 [3 favorites]


As a matter of policy, the FAA also doesn’t compare airlines’ records to search for warning signs. According to a 2013 report, three in four inspectors surveyed by the U.S. Department of Transportation said comparing airlines would make air travel safer.

Service difficulty reporting isn't meant to detect problems with an operator--it's meant to find problems with aircraft designs by providing evidence of a systemic problem. It is also explicitly intended to be non-punitive. If you start tying punitive actions against operators (or manufacturers) to their event reporting, it will lead to more covering up of problems, and ultimately more accidents.

This is great data mining by the TB Times showing strong evidence of maintenance problems at Allegiant, but I wouldn't want to rely only on the testimony of FAA inspectors for the interpretation of the data and on suggestions of how to fix the system--they often have their own axes to grind, and they are usually only interested in one piece of the overall aviation safety picture.
posted by cardboard at 6:16 PM on November 3, 2016 [3 favorites]


Given the large number of MD-8xs and Boeing 717s still in service, I don't think you can say it's a sample size issue if Allegiant is having more incidents per aircraft.

Most of them are pretty high on the cycle count, though, since they primarily fly short haul routes that would be done on RJs if the airlines could fly RJs. (I believe most majors still have scope clause limitations with their pilot groups that make it noneconomic to fly them, which is why they haven't been replaced with more fuel efficient large RJs) Basically, the MD-80 series is the smallest plane the pilots will allow them to fly, aside from the Fokker 100 AA flew until its worse-than-MD80 fuel efficiency killed it.

Ironically, Allegiant flies them farther than most anyone else, so in many ways they ought to have fewer problems with them than other airlines.
posted by wierdo at 6:21 PM on November 3, 2016


When I was younger, I figured that, while budget airlines might cut corners on things like comfort and not losing your luggage, the FAA would maintain a strong floor on safety. The ValuJet crash changed my mind on that. I just won't fly a U.S. budget airline. I even try to avoid the regional airlines like American Eagle.
posted by praemunire at 6:22 PM on November 3, 2016 [2 favorites]


Forty-two of Allegiant’s 86 planes broke down in mid-flight at least once in 2015

Sample size or no, that sounds terrible.


Of course it's supposed to sound terrible. As a back of the envelope calculation, if each plane is making 300 flights a year or so, that implies well under 1% of flights having any issue. And in fact the "unexpected landing rate" is down around 0.1% in the full article.

I'm a bit torn about this. There's a lot of stuff in the article that leads you to draw bad inferences*, on the other hand it seems they are legitimately doing worse and publicly calling them, or regulatory agencies, out isn't all bad. Doing the math I could fly Allegiant as the carrier my life and I'd be unlikely to ever have an incident, but why would i want to when there are safer airlines? Trying to sort out how much i should worry puts me in mind of Kahneman's claim that we're really bad evaluating low-frequency events--usually we don't think of them at all, but when we do we tend to vastly overestimate the likelihood.

finka: Could that just be the phenomenon that small sample size makes extreme results more likely?

wierdo:Given the large number of MD-8xs and Boeing 717s still in service, I don't think you can say it's a sample size issue if Allegiant is having more incidents per aircraft.

Actually, finka has a good point here about sample size that's different than yours, wierdo. If you divide something up randomly into a sets of very different sizes, then statistically you are very likely to find the ones with extreme characteristics are the small ones. It has nothing to do with some underlying difference, it's a statistical artifact--the small sets are more impacted by random variations in the partitioning (ie, bad luck.)

Since Allegiant also has old planes, I don't think it's likely the entire problem is an artifact. It could easily make it look more extreme though.





*For example, the bar chart makes it look like there's a big gap between Allegiant and the next bad airline--as if it's a qualitative difference--but that's not necessarily true. Most airlines are not shown and the ones that are, are dissimilar to Allegiant.
posted by mark k at 6:38 PM on November 3, 2016 [1 favorite]


Trying to sort out how much i should worry puts me in mind of Kahneman's claim that we're really bad evaluating low-frequency events--usually we don't think of them at all, but when we do we tend to vastly overestimate the likelihood.

Well, in these cases it's not quite so much the frequency that is the issue, is it? If flying Allegiant were an act necessary to leading a normal everyday life, it's possible you'd find those numbers tolerable when weighed against the benefits. But, for the most part, no one has to fly Allegiant to get anywhere or do anything.
posted by praemunire at 6:42 PM on November 3, 2016


When I was a lad, an old pilot told me "Fly Boeing. You'll have the occasional electrical fire, but you can always land them. Lockheed can't design a diaper without it failing."
posted by Sphinx at 7:00 PM on November 3, 2016 [2 favorites]


I even try to avoid the regional airlines like American Eagle.

I had to fly on a small-ish prop plane from Philly to New Haven a few weeks ago on American Eagle. I didn't mind the size of the plane so much (I've flown smaller) but the age of the plane kind of wigged me out. The armrests actually had ashtrays in them, which doesn't exactly inspire confidence! I mean, that makes them at least, what, 30 years old? You don't see all that many vehicles of any type that age.
posted by lunasol at 7:04 PM on November 3, 2016 [2 favorites]


It's also my understanding that they generally require less experience from their pilots, have them on more strenuous schedules, and pay them less. So.
posted by praemunire at 7:05 PM on November 3, 2016 [2 favorites]


I'm due to fly Allegiant in a couple weeks so... thanks for making me shit myself metafilter! 😭
posted by supercrayon at 8:17 PM on November 3, 2016 [3 favorites]


Well the good news is short haul pilots get a lot of experience in the actual things that are mostly dangerous (IE: take offs and landings) as opposed to level flight supplemented with autopilot.
posted by Mitheral at 8:26 PM on November 3, 2016 [2 favorites]


"Fly Boeing. You'll have the occasional electrical fire, but you can always land them. Lockheed can't design a diaper without it failing."
posted by Sphinx at 9:00 PM on November 3


You're giving me flashbacks to the old L-1011's which, actually weren't too terribly bad. However, I still think it was wise of Lockheed to exit the commercial, aka passenger, aircraft business.
posted by InsertNiftyNameHere at 9:00 PM on November 3, 2016


Well, in these cases it's not quite so much the frequency that is the issue, is it?

If the problem isn't so much the frequency then I don't know what it is. All airlines have some problems. The whole issue is Allegiant has them at a higher frequency, although still at a very low absolute rate.
posted by mark k at 9:01 PM on November 3, 2016


Jessica Stoffel was so afraid on Allegiant Flight 175 over Mesa, Ariz., that she grabbed the stranger next to her and squeezed his hand. “I was terrified and honestly did not think we were going to make it,” she said.

30 years ago, flying from ME to DC often involved small planes. One seat either side of the narrow aisle and if you leaned you could see what the pilots saw through the open cockpit. It was usually wondrous, cruising above the Hudson, but not so far above it that you couldn't look the liberty statue in the eye or not appreciate the skyline and it made about half the passengers nervous, but not me.

I'd night-landed in Quito before over the burning trash fires but I was not prepared for a loud thunk and the pilot shouting "WTF KIND OF BIRD WAS THAT?" And then we dropped and even the lone stewardess looked green. I held hands with a complete stranger before things were under control again. Not sure who reached for who first.

We are really not supposed to be flying through the air in machines and deep inside we know this.
posted by Mr. Yuck at 9:09 PM on November 3, 2016 [4 favorites]


This was an interesting article to see on my phone while sitting at the gate waiting to board my Allegiant flight home. But we made it!

Although a few days ago for the flight down we did have to wait forever for a mechanic, and then eventually have a replacement plane flown in, because of some problem with a cooling fan in the avionics somewhere. But at least that was a pre-takeoff mechanical problem. And they gave us pizza while we waited.

I sort of figured to take what you can get with an MD-80 in this decade.
posted by traveler_ at 9:11 PM on November 3, 2016


If the problem isn't so much the frequency then I don't know what it is.

You've suggested that in evaluating the safety issue one might fall prey to a cognitive bias that exaggerates the likelihood of a disastrous event. That may well be true. But when deciding to take a potentially risky action, one does not consider the likelihood of disaster alone; one weighs that risk against the potential benefits. So even if we tend to overrate in our minds the risk of Allegiant crashing, as there are many airlines to fly besides Allegiant, the benefits may still not be worth the actual risk.
posted by praemunire at 9:28 PM on November 3, 2016


@praemunire, sure, if all other things are equal it'd be a no brainer to take an airline with fewer mechanical issues, because you'd expect the flight to be safer.

What I have no idea of is what if Allegiant was going to save me $100 or 3 hour layover. Or if they had a flight at a reasonable hour and the only alternative was at 4 AM. Is it more dangerous than other travel choices I've made, like taking a puddle jumper over mountains, or driving home tired at night in bad weather?

The difference isn't that I'm weighing the risks differently, it's that you talk about people weighing risks and benefits and I think it's nigh impossible in this sort of situation.

The Tampa Bay article obviously wants you to decide they are qualitatively worse, and thus unacceptable. Because that's how they have to write articles about this sot of subject. I don't know that they're wrong--maybe the FAA should shut them down if they don't improve.

The Tampa Bay article obviously wants you to decide they are qualitatively worse, and thus unacceptable. Because that's how they have to write articles about this sot of subject. I don't know that they're wrong--maybe the FAA should shut them down.
posted by mark k at 9:57 PM on November 3, 2016


I've only flown Allegiant for one trip, to Las Vegas. It was memorable not for any mechanical issues, though the plane was obviously well aged, but because of the delicious contrast between the rowdy passengers on the first flight and the incredibly hung over and subdued passengers on the return flight.

Compared to routine decisions I make every week like being a passenger with a tired driver, I doubt Allegiant'a mechanical lapses even register.
posted by Dip Flash at 11:15 PM on November 3, 2016


Despite being in Pittsburgh, which as soren_lorensen notes has a sort of anti-hub airport with Allegiant being one of the main airlines left, I've avoided Allegiant flights ever since I saw their pilots' union's Open Letter to Passengers about a year ago, and reconfirmed that based on further warnings a few months later. I'm also subscribed to a flight anxiety support newsletter written by a retired airline pilot who's written that Allegiant is one of the few airlines he will not board; between all that and this latest story I'm pretty confident saying that I'm just going to keep on avoiding these guys.

There's enough out there suggesting to me that Allegiant is more dangerous than it needs to be; everyone has their own threshold but for me, this is enough.
posted by DingoMutt at 5:50 AM on November 4, 2016 [1 favorite]


You can find a list of Allegiant's incidents here.
posted by knackerthrasher at 7:07 AM on November 4, 2016


praemunire: "When I was younger, I figured that, while budget airlines might cut corners on things like comfort and not losing your luggage, the FAA would maintain a strong floor on safety. The ValuJet crash changed my mind on that."

From TFA:
Southwest had the lowest rate of problems last year. It had about one in 10,000 flights end in unexpected landings.

JetBlue was in the middle of the pack. It had about three flights end in unexpected landings.
So, current data seems to suggest that the biggest "budget" carrier actually has the best record, and the second-biggest also fared better than the legacy carriers.

(I agree with you on the regional airlines though – American and United don't run those flights directly because of liability reasons, which should tell you something.)
posted by schmod at 7:17 AM on November 4, 2016


When I was a lad, an old pilot told me "Fly Boeing. You'll have the occasional electrical fire, but you can always land them. Lockheed can't design a diaper without it failing."

But MD-8x's / 717's are Douglas designs. Evolved DC-9's.
posted by ROU_Xenophobe at 7:37 AM on November 4, 2016


The Tampa Bay Times is doing some incredible data-based investigative journalism.

You'll recall, this is the same paper that produced:
-The stunning Walmart crime story.
-The Pulitzer-winning Pinellas County failure factories investigation.
-And a deep look at the price of mental health cuts in Florida - another Pulitzer winner.

Support your democracy: Subscribe to a newspaper today!
posted by sixpack at 10:24 AM on November 4, 2016 [1 favorite]


We flew Pittsburgh to St. Petersburg, FL in February, and as fate would have it, the girlfriend of one of my family members was a flight attendant for many airlines, including over a decade with Allegiant. She said everything about them was terrible and got worse as time went on, including their approach to safety, and that you couldn't pay her enough to fly with them.

She's with Southwest now and loves it.
posted by tonycpsu at 1:09 PM on November 4, 2016


"When I was a lad, an old pilot told me 'Fly Boeing. You'll have the occasional electrical fire, but you can always land them. Lockheed can't design a diaper without it failing.'"

I was taught to always, always fly Boeing, and that it was much safer... but being raised in Seattle may have had something to do with that. ;)
posted by litlnemo at 2:01 PM on November 4, 2016


(I agree with you on the regional airlines though – American and United don't run those flights directly because of liability reasons, which should tell you something.)
It isn't liability. It's cost. The ticket price would be insane if they flew turboprops and RJs at the same payscale that mainline pilots and FAs get, which is what their union contracts require.

I'm not saying that I'm comfortable with how low regional airline crew pay is, mind you, but it obviously can't be as high as it is for larger jets with more seats. Combine the contract issues with the lower return on capital and it's pretty much inevitable that the majors have zero interest in flying small planes to serve smaller markets.

The really unfortunate part is that the people in those smaller markets are less able to absorb the higher prices that are necessary to pay crews more reasonably and have less ridiculous work rules. Add to that the fact that even with the state of the regionals being what it is, flying on them is still by far the safest form of transportation into those communities, so increasing cost enough to make a difference to crew pay and work rules will literally kill people and it's a damn sticky problem.
posted by wierdo at 7:11 PM on November 4, 2016


soren_lorensen: "And Pittsburgh is like the opposite of a hub so not very many carriers fly out of here--Allegiant is one."

Thanks again for screwing us, USAir (née Allegheny Airlines).
posted by Chrysostom at 9:48 PM on November 4, 2016


« Older The Children of "Runaway Train"   |   A Carmen Medley........... Newer »


This thread has been archived and is closed to new comments