Remember when I posted that Frontier had seen strong fare gains in Q2 results compared to last year? I’m going to have to take that back. I went on quite a journey to try and get some answers to what actually was happening here, but in the end, I got none. Instead, I built my own model to get as close to reality as I think I can get. And it’s not nearly as good as the initial look would have suggested.

It all started when a friend pointed out to me that these strong gains didn’t really make any sense. After all, in the airline’s press release, unit revenue and revenue per passenger were basically flat, and total revenue was actually down about 4.5 percent on a 2 percent decline in capacity. So this chart I had posted couldn’t fit reality…

Frontier YoY % Chg for Top 15 Markets

^{DB1B/T100 Data via Cirium, airports shown by market size descending from left to right}

Your first thought might be that with the airline’s newish bundling strategy, what gets included in the fare vs ancillary might have changed. It doesn’t seem like that should be the case, but clearly something happened. After all, Frontier’s bundles are just a facade for what is still a base fare + ancillaries. It is just easier to buy it. For example, this $94.98 one way economy bundle from LA to Denver is actually…

I asked Frontier if anything had changed in the way they filed it, and my understanding after a long back and forth is that it didn’t change, though that wasn’t completely clear. Attempts to get clarification from the revenue side of the house failed, but I checked with Cirium to see if there was a data issue — there was not — and I spoke to several others to try to figure out what was going on.

The end result is I have resigned to the idea that without ancillary being reported on a route basis, there is no way to get this perfect. I just hadn’t expected it to be so wildly skewed in my year-over-year look. So, after talking to some experts, I decided to just build my own model based on the data that’s out there.

I took all the route-level data that I had, and then I had to reconcile that with the info in the press release. The DOT data is only domestic, but I figured that international in the airline’s announced results is pretty small. So I ignored that. Then I had to reconcile the data here. For example, Q2 2025 shows a $68 average fare. Frontier says it had fare revenue of $40.94 per passenger plus ancillary of $68.33. In other words, that DOT base fare data includes some of the ancillary. What a mess.

I just had to do math to figure out what ancillary amount was missing. This was easy on the network level, but to get it down to a route level? This is where it becomes more art than science. We know that ancillary revenue is lower on short-haul flights and higher on long-haul flights. So, I created some assumptions that adjusted the ancillary depending upon the route.

In the end, I was able to build a model that I think shows a far more accurate representation than that old chart above. Here’s a look at a corrected chart showing the stage length-adjusted passenger unit revenue (without ancillary) and the stage-length adjusted total unit revenue (with ancillary).

Frontier YoY % Chg for Top 15 Markets

^{DB1B/T100 Data via Cirium, airports shown by market size descending from left to right, TRASM is estimated}

Not so impressive now, huh? I mean, Denver actually looks pretty good, but things west of there didn’t do so hot. There weren’t nearly as many bright spots.

And what if we now look at Vegas in a new light? You’ll remember that Frontier led the pack with an actual unit revenue gain when we first looked while nobody else did. But now, not so fast. An estimated 10.6 percent drop is worse than Southwest now. It’s still not terrible, but it brings the airline back down to earth.

So… forget pretty much anything I had said about Frontier’s impressive performance. Q3 looks fairly similar, but as is always the case with Frontier, big profits are just around the corner… they say.