Heading out the door? Read this article on the new Outside+ app available now on iOS devices for members! Download the app.
The recent Ironman debuts of Norwegian teammates Kristian Blummenfelt and Gustav Iden have not only seen world best times come tumbling down—they’ve also led to plenty of jaws dropping and folks asking: “How on earth did they do that?”
Good question. We reached out to exercise physiologist, data analyst, and coach Alan Couzens to help us look at what it took from a training and physiology standpoint for Blummenfelt to nail a 7:21 debut Ironman and Iden a 2:34 debut Ironman marathon. The data comes from their publicly available Strava files and training logs. We also asked Blummenfelt’s coach for some additional insight. Spoiler alert: Setting aside their phenomenal talent, the answers, from a training and physiology point of view, are far simpler than you might think…
Blummenfelt’s Cozumel bike performance
At Ironman Cozumel on Sunday, Blummenfelt posted an impressive 4:02.40 bike split in conditions that were less than favorable for fast times. Some parts of the bike course were water-logged with up to eight inches of water, obviously causing athletes to slow down. Olav Aleksander Bu, Blummenfelt’s coach on the ground and a coach with the Norwegian team, said this exacerbated an issue his athlete already had with tires going into the race: “We were not able to get the correct tires due to an error and we ended up having to fit butyl tubes inside the tubeless tires, which added to the rolling friction. Under ideal conditions, with a dry track, it is our belief he could have gone more than five minutes faster on the bike.”
And from a training and aerodynamics standpoint—as is the case for many first-time Ironman athletes—the Norwegians believe there are still further gains to be made: “We know we can still optimize the bike further,” Alexsander Bu said. “We can open Kristian’s hips more and flatten his back to help lower his CdA (coefficient of drag) and at the same time increasing his highest sustainable power output.”
Additionally, if his Strava file is to be believed, the bike course was a mile long. You can see the full bike file here.
And his run…
A lower CdA and improved position would arguably help slice a chunk of time from the Olympic champ’s marathon split too—although to be fair, his 2:35:24 was far from shabby, especially for his maiden voyage at the distance.
Alexsander Bu said: “We had trained him to run ~2:30 for the marathon, but as it was his first time racing the distance it’s hard to know how hard you’ll be able to go on the run, especially after the swim and bike. As Kristian felt more confident towards the end of the run, he increased the pace gradually over the last 5K, finishing with ~3:30 min/km pace (5:39 min/mile pace).”
As with the bike course, conditions on the run, he said, were not optimal for world-best times. “Under ideal conditions, we are confident now with some more optimizations in the program, that Kristian can go closer to 2:25 on the marathon.”
His run file—though he says he hit lap for two short port-a-potty stops—is here.
Iden Under Analysis
Prior to Blummenfelt taking Cozumel by storm, his countryman and the two-time 70.3 world champion Iden had already raised eyebrows with his own debut into long-course racing. His 2:34 marathon split at Ironman Florida and 7:42 finish on Nov. 6 left competitors wondering how he made the transition to Ironman racing look so seamless. Given that Iden has posted almost all of his run training for the three months leading into the race on Strava, we asked Alan Couzens to check it out and report his findings. The results are interesting…
“Of course, the data is limited to what he shared on Strava,” Couzens said. “That said, he was sharing approximately 90K (55 miles) worth of running each week, so one would have to assume that we are getting the vast majority of his run training within this sample.”
The training zone distributions are set by Strava’s algorithm, but for Iden they worked out to be:
- Active Recovery (>4:00/km or 6:26 min/mile pace)
- Endurance (3:26-4:00/km or 5:31 min/mile – 6:26 min/mile pace)
- Tempo (3:05-3:26/km or 4:57 min/mile – 5:31 min/mile pace
- Threshold (2:53-3:05/km or 4:38 min/mile – 4:57 min/mile pace)
- VO2max (2:43-2:53/km or 4:22 min/mile – 4:38 min/mile pace)
- Fast (<2:43/km or <4:22 min/mile pace)
That means that Iden’s Ironman pace was ~3:40/km or 5:53 min/mile pace, which falls in his training Zone 2. And most of his runs in Zone 1 were at ~4:30/km or 7:16 min/mile pace (approximately 80% of his Ironman pace). According to Couzens, Iden’s best recent 5K (that can be found online) is 14:07 (~2:49/km or ~4:30 min/mile pace), i.e. smack bang in his VO2max zone as we’d expect.
Couzens explained that with an Ironman pace of ~5:50 min/mile pace, we might expect to see a good proportion of his run volume at or faster than this pace in training. But the training data does not show this. Instead, it shows that two-thirds of his weekly run volume was actually performed in his Active Recovery Zone—that is, easy jogs at only about 80% of his Ironman race pace or 60% of his 5K race pace. Of his weekly run volume (from Strava), about 10 miles of it was at Ironman pace, about three miles of it was at tempo pace, and less than two miles of it was at threshold. There was very little mileage at anything above this.
“For comparison, for a competitive amateur running 3:30 off the bike, this would be equivalent to doing the bulk of their run training at 10:00 min/mile pace,” Couzens said.
The pattern held throughout much of the three months leading into Ironman Florida, with Couzens noting his data shows lots of easy, short, frequent jogs coupled with sporadic (every five days or so) key sessions that were either a hard long run or a speed session incorporating a healthy dose of sub-threshold up to threshold work.
Couzens was quick to add: “Now, this is not to say that all of Gustav’s running was easy. For example, on Oct. 13 (23 days before the race), he did a key session on the track—15 miles total at 5:40 min/mile pace (a little quicker than his Ironman average pace) with 3 x 5K descending one to three from ~5:10 min/mile pace to 5 min/mile pace! So that’s 23 minutes of tempo leading into 27 minutes of threshold work, all in the middle of a solid 90-minute run!”
The greatest takeaway here—if you’ve not seen it already—is that his training approach is remarkably polarized. “Gustav truly embodies that old principle: ‘Keep your hard days hard and your easy days easy!'” said Couzens. “This is true to the advice of pretty much all the great run coaches and it should be no surprise that it led to such a remarkable performance!”
Leave It To Science
While many triathletes know about polarized training and its benefits, many still struggle to truly land their volume at either end of the spectrum, but Couzens said there’s nothing new about this approach. “About 40 years ago, a classic exercise physiology paper was released by Sandra Harms and Robert Hickson, that concluded low levels (intensities) of physical activity (approximately 50% VO2max or less) are sufficient to induce most of the total possible increases in mitochondria of fast-twitch and slow-twitch red muscles,” he said.
In plain English, this means that training at approximately 60% of your 5K pace is all that is needed to maximally stimulate aerobic adaptation in the “red” slow twitch and fast oxidative fibers, Couzens said. And to develop the higher threshold fibers, sprinkle in less frequent but higher intensity “key workouts.”
The conclusion here? Aside from having a ridiculous amount of natural talent (and an impressive work ethic, no doubt), the Norwegian secret to running a 2:34 in your debut Ironman marathon has a lot to do with keeping your hard days hard—and your easy days very, very easy.
Takeaways For Us Mere Mortals
Of course, very few of us will ever roll up to our debut Ironman and bust out a 2:34 marathon off the bike, but that’s not to say we can’t all learn lessons from the way Iden, Blummenfelt, and the Norwegian team train. Couzens said he is often criticized when he applauds the polarized training approach, because he’s often met with the question: “Well, yeah, that’s great if you’re able to run 50+ miles a week, but what about us age-groupers training half of that?”
He said: “The implication here is that if you don’t have the time to go as long, shouldn’t you just go harder? And I think this is how many (or most?) age-groupers train, especially on the run. At the risk of generalizing, I’d say the typical age-grouper, with a four-hour Ironman marathon to their name, is doing the bulk of their running in that upper Zone 2/lower Zone 3 range—that is, faster than their Ironman race pace. This would be equivalent to Gustav running 30-minute 10K pace for the bulk of his training!”
Couzens cited a study from 2014, in which Stephen Seiler and colleagues looked at this very question in time-crunched runners training only ~four hours per week. One group—the low intensity “polarized” group—did 73% of their training below 77% of maximum heart-rate. The higher intensity group adopted the more typical distribution and did more than half of their training above this mark. Despite training considerably harder, the high intensity group actually performed 1.4% worse than the lower intensity group.
So what’s the greatest thing for you to apply to your training? “Remember that the assumption that if we train at a higher intensity we get more aerobic benefit is fundamentally flawed,” Couzens said. “There is a ‘correct’ intensity for our easy sessions that gets us the maximal adaptation—and any energy expended above this point is wasted energy that could be better directed towards our higher intensity sessions. This is the lesson that Gustav and his coaching team totally get and, in my opinion, it’s a lesson that many age-groupers would benefit from adopting, even if they are not training at the same volumes or paces as the speedy Gustav.”
You might not be as fast as Iden or Blummenfelt, but you can still train like them.