When Did "Bike for Show, Run for Dough" Stop Being True?

Originally published at: When Did “Bike for Show, Run for Dough” Stop Being True? - Slowtwitch News

Top runners can make headlines with quick marathon times in IRONMAN racing, but they don’t always get the best overall results. Photo: Kevin Mackinnon

Compared to many other sports, triathlon is pretty young. It incorporates sports that have long been contested around the world, but combining all three sports is a relatively new idea. Because of this, up until quite recently, elite triathlon was dominated by athletes who excelled at cycling or running and could simply hold their own in the other two sports.

There are still plenty of professionals who fit this description, and every athlete will, of course, have a preference for one sport over the others, but as triathlon has matured, more and more people are entering the sport as triathletes first – not just swimmers, cyclists or runners who decided to take a stab at the multi-sport event. This change occurred much quicker in the draft-legal space before reaching long-course racing. Today, though, even half- and full-distance super-runners have to face a harsh truth: they can not — in the pro ranks, at least — steal the win with a stellar run after an average, or sub-par, bike.

A strong, solid run is of course needed to win any race, but there is a balance that must be struck between the bike-run split for optimal results. The most consistent athletes on the pro tour have that balance dialled in, pushing hard enough on the bike to stay at or near the front of the race heading into T2, all while saving their legs enough to hold on for the win or podium on the run. (And also giving themselves enough of a buffer over the strongest runners so they won’t be caught before the line).

Tamara Jewett routinely posts the fastest run split when she races. Photo: Kevin Mackinnon

We want to find out where the line is in this balance. How much time can an athlete give up on the bike before the gap to the front is too big? What is the minimum that long-distance triathlon’s super-runners have to do between T1 and T2 in order to give themselves a legitimate chance at a come-from-behind win on the run?

Comeback Runners

Three-time IRONMAN world champion Mirinda Carfrae was a super-runner just before the scales tipped in long-distance racing. In 2010, when she won her first IRONMAN world title, she was 11 and a half minutes behind the leader after the bike, but a 2:53 marathon lifted her to the top of the podium. When she won for the second time in Kona, in 2013, she was seven and a half minutes back as she entered T2. A 2:50:38 marathon (13 minutes faster than second place’s run split) carried her to the win.

It was the next year, however, when Carfrae laid down one of the most incredible comebacks in triathlon history. After a one-hour swim and a 5:05 bike, she was fourteen and a half minutes behind Daniela Ryf. Even for a runner like Carfrae, this deficit seemed insurmountable. But she ran yet another 2:50 marathon on Hawaii’s Big Island to catch Ryf and win by two minutes.

A few years later, in 2017, once again in Kona, the men’s race saw a phenomenal comeback on the run. Lionel Sanders out-biked the rest of the top contenders, posting a 4:14:19 split, and he set out on the marathon course looking like he had finally cracked the code in Kona. He ended up running well, recording the fourth-best marathon of the day with a 2:51:53, but Patrick Lange ran 2:39:59 to come from nine minutes down to win the race.

Patrick Lange has used speedy runs to win his three IRONMAN world titles. Photo: Kevin Mackinnon

Comebacks on the run will of course always be an occurrence in triathlon, and from time to time a super-runner will lay down a remarkable result like Carfrae or Lange to steal the win. But instances like those in 2014 and 2017 in Kona are happening less and less — especially at the biggest races on the pro circuit.

Just A Little Short

At IRONMAN Brasil in June, Manoel Messias entered the first full-distance race of his career. An Olympian and three-time Pan Am Games medallist, Messias is a strong runner, and he made this clear while racing on home soil for his IRONMAN debut, when he ran an incredible 2:26:50 marathon to finish his day. His run split was 11 minutes faster than the next-best marathon of the day, and his bike was strong, too, as he posted the fourth-fastest time of the race. He still lost.

Argentina’s Luciano Taccone won the race after riding the 112-mile course in 4:00:09 and following that up with a 2:37 marathon. Messias’s run was unrivalled, but he didn’t even come close to the win, crossing the line more than five minutes behind Taccone.

In July, Canada’s Tamara Jewett ran a 2:40:05 marathon at IRONMAN Lake Placid. Her split was six minutes ahead of the second-fastest that day and a whopping 15 better than third-best. Her run helped her pick off multiple women before the finish, but she couldn’t make it to the podium, finishing in fourth, two minutes behind third place.

Sam Laidlow is consistently able to back up his top-tier riding with solid run splits. Photo: Kevin Mackinnon

Twice in 2025, American Matt Hanson has run nine minutes faster than the next-best runner in the men’s field, but on both occasions, he finished well behind the podium. The first instance came at IRONMAN Cairns in June, when he ran a 2:30:21 marathon. This run lifted him to fifth overall, but his bike split was close to 17 minutes slower than fellow American Matthew Marquardt’s, who won that day and recorded the second-fastest marathon with a 2:39:47. Less than a month later, at Challenge Roth, Hanson ran a staggering 2:28:03 marathon, but he finished ninth overall. France’s Sam Laidlow won in Roth, out-biking Hanson by 18 minutes and still managing to run a 2:37:19 marathon.

What do these results all have in common? The super-runners were not just bested on the ride by a significant amount. If that had been the case — if it were a matter of super-bikers versus super-runners — then they would have had a shot at overcoming those deficits and potentially winning. It’s more than just falling behind on the bike, because in each of these cases, the winners of the races all had solid runs. The runs were minutes slower than those of Hanson, Messias or Jewett, but they were still strong performances that kept their leads safe. The athletes winning big races don’t get to the run after a top-five bike split and simply hope they can hold on for the run — they know they have the legs to hold off any charging runners coming up behind them.

Where’s the Line?

The limit for how far behind a pro athlete can be as they leave T2 in a full-distance triathlon will, of course, vary from race to race. If Hanson or Jewett show up to an obscure race with no top professionals, they could come off the bike 20 minutes back and blow by everyone on the run course. (Of course, if the field was that weak, it’s unlikely they would be so far back off the bike to begin with.) But at the biggest races on the pro circuit — the continental championships, the worlds, iconic events like Roth — the fields are always going to be stacked. There is no room for a weakness at at these races.

Looking at the run splits from at major full-distance events in the past decade, the limit for a comeback on the marathon seems to be around 10 minutes. In Kona in 2022, Gustav Iden was down six minutes to Laidlow after the bike, but his 2:36 marathon lifted him past Laidlow, who still ran a solid 2:44. Iden won by two minutes. The next year in Kona, Lucy Charles-Barclay won her first IRONMAN world title, holding off a charging Anne Haug on the run. Charles-Barclay had a lead of about 12 minutes going into the marathon, and although Haug ran a race-best 2:48 split, the Brit was able to hang on with a 2:57 run to win by three minutes.

At IRONMAN Frankfurt in 2021, Denmark’s Kristian Høgenhaug was up eight and a half minutes going into the run, but Patrik Nillson of Sweden used a 2:39:39 to come from behind and win by a minute. In 2023 at IRONMAN Texas, Kat Matthews was nine minutes off the lead as she exited T2, but the fastest marathon of the day shot her up the rankings to the top of the podium.

Once again, this is not a hard rule for racing. There will likely be another day when someone pulls a “Mirinda Carfrae” and comes from 14 minutes down to win a big race on the run. But if pros want to have a shot at winning the most competitive races time after time, they need to consistently get off the bike at least within 10 minutes of the leaders. Otherwise, those super-runners may find themselves running amazing times and managing to go from 12th to fourth, or maybe even making it onto the podium, but more often than not, they will run out of room and rarely be able to complete their late-race comebacks to take the win.

2 Likes

Related question - how far can you get without a front pack (or maybe 1.5 pack) swim?

On the Men’s side, IMO the only real contender who isn’t exclusively “front pack” is Ditlev, and he’s better described as being in the 1.5 pack. He makes slow front pack swims, but gets ejected if the pace gets turned up too quickly - or stays close enough that he’s under a minute.

I guess Chevalier in 4th last year shows about as high as you can get?

On the Women’s side Laura Phillip was fine giving up 4 mins on the swim.

Still true.

1 Like

Super interesting topic and purely eye test the mid pack runners are now still quite electric runners. Laidlow being the prime example.

It’s also why I think Ditpev will never win a world champs he loses too much on the swim he’s chasing immediately.

Lange might be the last single sport hero to achieve the big one and even then he’s only 5% slower on the bike. So I think it’s probably again just eye test thinking of him as just a mega runner.

I missed this article when it was published in August last year, before both Nice and Kona. Prescient with regard to Kona last year: Loevseth limited the gap and pressurised the two leaders to ride hard and run faster than they could sustain. Matthews let the gap balloon to 9 minutes to Loevseth and even running 2:47 couldn’t quite make the catch.

Thorsten has just shared the bones of his proper analysis on exactly this adage on insta: “Bike for Show, Run for Dough — Old Wives’ Tale or Backed by Data?
https://www.instagram.com/p/DWGeiTZDKuR/
and a Triathlete mag article, but here’s his graph (go read the insta):

Of course there’s shading but the 2024 and 2025 outcomes (podiums) of both Nice and Kona seem to me to be ‘run for dough’. We also saw in Marbella that at least the men’s podium was defined by ‘run for dough’, even though there was serious climbing on the bike.

Of course one swallow a summer doesn’t make, but, looking forward to Nice and Kona this year (with hors d’oeuvres at Texas and Roth) the run will decide the races. If Laidlow can avoid overbiking (for him; and not have a water panic) and establish a lead, with his improved run (see 2:37 at Roth), he can stay ahead of the chasers, or at least make them decide whether to overbike (for them) to keep the deficit down: is this ‘riding for show’? Have to expect Matthews to ‘do a Loevseth’ and keep LCB and Knibb within range, not letting Loevseth get away, and ‘run for dough’.

I don’t believe this statement to be true at all..The pressure on the two up front came from them, and I doubt there was more than a fleeting concern on what was going on miles behind them. And likely that fleeting thought went to Kat and was quickly dismissed once she was so far behind.

Their race was with each other, and any bike pressure(on Lucy) came from Taylor and her inadvertent penalty. And any run pressure on both came from each other while they still had control of their own paces. By the time a thought of Loevseth came into any of their heads, it was far too late to do anything about it. The die had been cast in the first half of the run on how it was going to end for both of them. Loevseth was just there to pick up the shattered pieces. To them it really was a two woman race and they only had to look to the side or right behind to see who the threat was for the win..

A lot of people of course use hindsight to handicap how the race went, but of course neither had that at the time..

1 Like

I don’t buy that, because Kat and Laura were never far enough behind. Kat Matthews arrived in T2 14 minutes behind Taylor Knibb, but then followed it up with a 2:47:23 marathon. Taylor Knibb’s PB for an Ironman marathon is 3:04:43. If she had run that same time in Kona 2025, that 14 minute head start out of T2 still wouldn’t have been enough to avoid being caught by Kat Matthews before the finish line.

10+ minutes may sound like a lot, but even without the benefit of hindsight, just going by previous Ironman run splits, Taylor Knibb must have known that that wouldn’t be enough to guarantee that she would be safe from Matthews and Philipp in the run. Even if you make Lucy Charles-Barclay a DNS, Knibb could not have afforded to bike and run much slower. Knibb ran the first 10 km of the marathon at a 4:07 min/km average, on pace for a 2:54 marathon, which we now know was too fast for her to sustain. And yet Matthews and Philipp still made 2 minutes good on her over that same 10 km.

Like I said, you cannot look at the race with hindsight, that’s why you can only bet before the race

You don’t need hindsight to know that Haug’s course record was 2:48:23 and that both Matthews and Philipp had in in them to beat that record if the race was going well for them. And if you’re Taylor Knibb, or Lucy Charles-Barclay, and want to win, you have to assume that the race will go very well for Philipp and Matthews.

It is with the benefit of hindsight that we know that it wasn’t Philipp’s day and that Matthews beat Haug’s record by a minute. But Matthews’s run time wasn’t a black swan event. It was very well within what everybody expected her to be capable of. Without the benefit of hindsight, Lucy Charles-Barclay and Taylor Knibb couldn’t be sure that neither Matthews nor Philipp would run an even faster marathon.

I’m not sure I believe that this chart is telling us - namely that the swim is less important in championship events. All the other data points we have seems to tell us the opposite - namely that those athletes who struggle in the swim tend to do less well at championship events (particularly on the men’s side).

Part of this is that by doing well in the swim, you’ve allowed yourself to slot into the main pack, and so have a good bike - which is more a feature of championship events than run of the mill events.

But what the data would then show is that the bike and swim are more highly correlated at championship events - leading to a reduction in the swim’s importance if you’re just taking the segment times at face value.

Pray what are “all the[se]other data points” to which you refer?
Bear in mind there are fair few pinos racing pro in the ‘overall’ category whereas Kona is ‘by qualification’ so there is a quality filter (position in regular season races) working there.
The “relative importance” is basically the regressors in a multiple linear model: iow how much of the final result order can be determined by each of the swim/bike/run/transition order.