ROAR ANALYSIS: Porsche, Cadillac on top at Daytona test

The Roar Before the 24 is Daytona's traditional pre-season test event for both the 24 hour race, and the IMSA season in general. It's the first chance we get to properly see the 2026 cars and gives just an inkling, a glimpse, of what may be in store for the year ahead.

And yet the test is just that: a test. Just as in other forms of motorsport, where all testing results have to be taken with a large pinch of salt, it's exactly the same here. Maybe with an even larger pinch of salt. For example, last year, BMW were fastest in the Roar test, but Porsche won the race and undoubtedly had the quickest car overall.

Client Challenge

So, it's hard to draw meaningful conclusions here, especially as there's every chance the Balance of Performance figures will change before the race anyway.

But nevertheless, it's interesting to look and see how various teams did and how the race may look from this point in the week leading up to the 24 hours.

SPONSORED BY the racing line

Download The Racing Line on iOS today and get 70% off a year's subscription!

Check it out

Porsche 1-2-3 on single lap pace

The first thing to note is all 3 Porsches, the 2 factory cars and the customer JDC-Miller machine, had the fastest single laps of all in the 2026 Roar test.

These 3 were all set in the first session on Friday. The the #7 Porsche of Felipe Nasr was fastest from Nico Pino in the #85 JDC-Miller car, followed by last year's winner Laurens Vanthoor in the #6 Porsche.

There wasn't much to choose from between them; Pino was just 4 thousandths of a second slower than Nasr while Vanthoor was 0.108 off his Brazilian teammate.

The #10 Wayne Taylor Racing Cadillac of Filipe Albuquerque was the fastest non-Porsche, but the Portuguese veteran was just a smidge over 4 tenths off Nasr. The fastest BMW was the #24 in the hands of Sheldon van der Linde, just 0.011 slower than Albuquerque.

The fastest Meyer Shank Acura over a single lap, meanwhile, was the #60 in the hands of Tom Blomqvist. The Briton, though, was almost a second off Nasr.

Long run pace topped by Cadillac

But, seasoned endurance racing or sportscar fans will know single lap pace doesn't really mean all that much: it gives you pole position but that's about it.

Long run pace is what matters here. And, based on the 7 sessions from the Roar, Cadillac narrowly beat Porsche to the punch here. Filipe Albuquerque, in the #10 and the fastest non-Porsche driver in single lap results, was fastest in the long runs, when you take every car's longest run and average it out using 20% pace.

He set an average of 1:37.029 over 13 laps in the Wayne Taylor Racing machine, over two tenths quicker than Julien Andlauer in the #7 Porsche. The Frenchman set his time in an earlier session, the third session of the test, doing 18 laps.

Despite being at midday, the third session was cooler than the fourth. This matters because, in previous years, the Porsche has generally performed better in cooler temperatures or at night. See Le Mans last year when they were faster than the winning #83 Ferrari during the night.

Client Challenge

So, Cadillac may have genuine long run pace. At this stage it is still too early to tell, but it's an interesting indication.

Cadillac were third fastest too, the #31 Action Express Whelen-sponsored car with guest driver and rising NASCAR star Connor Zilisch at the wheel. He did a 14-lap stint also in session 5, a couple of laps after Albuquerque finished his, but was over 6 tenths slower than his Cadillac stablemate. Perhaps understandably given this is Zilisch's first time in a GTP car!

The fastest BMW and Acura were next up, separated by just 7 thousandths of a second. Dries Vanthoor in the #24 car was the slightly faster of the two, a 97.768 set over 23 laps in the fifth session, overlapping with Zilisch's.

#93 Acura's Nick Yelloly was only very slightly slower, also overlapping his stint with Zilisch and D. Vanthoor.

The slowest of the 5 manufacturers was Aston Martin with the Valkyrie. The Valkyrie has never raced here, the only GTP car this year never to have done so. It makes sense, then, that Aston are still finding their feet at Daytona International Speedway.

Marco Sorensen's 15-lap stint in the 6th session, the only GTP car to do its longest stint on Sunday, let alone in that specific session, culminated in an average of 1:38.727. This is around 1.8 seconds off Albuquerque's best stint.

20% averages put Porsche on top

However, a slightly different picture emerges when instead of taking long runs, you take all laps from every car and apply the 20% average filter to them.

This means qualifying simulation laps are included, which may skew the picture somewhat, but it's an interesting comparison nonetheless.

Porsche emerge on top here, quite considerably so as well. The #6 Porsche, of Matt Campbell, Kevin Estre, and Laurens Vanthoor, is fastest here by around two tenths. The fastest non-Porsche is the #10 Cadillac, followed in manufacturer terms by the #24 BMW.

Acura are next, followed by Aston Martin bringing up the rear; again, perhaps understandingly so. The Heart of Racing and Aston Martin, though, learn quickly – we saw evidence of that last year – and I'd expect they'll be right on pace come race day. And if they aren't, we'll get to enjoy that glorious V12 anyway.

A 'roaring' pinch of salt needed

In conclusion: this may all be meaningless. Which is nice.

The Roar is just a test. The teams will have programs to complete and setups to trial. They won't, at this stage, be too worried about ultimate pace, although it may be playing on their minds. And it's probably nice to be at the front for Cadillac and Porsche.

But we don't know full levels, we don't know what tyre compound each car is on, or if its new, scrubbed, or used. We don't know how much hybrid energy the teams are using each lap or engine modes.

So: huge pinch of salt. Massive. But we'll come back to this after free practice to see how the picture has changed, or maybe stayed the same.

Feature image: Porsche Penske Motorsport