Threshold & VO2max Differences Between Cycling and Running

Rich Stanbaugh

I'm starting this thread partially in response to @Dave Tallo's post. This post doesn't quite fit as response to the topic he started (Turning airborne, contagious lemons into lemonade: A Time for Experiments), but there is some overlap and he motivated me to share some cool stuff I've learned recently).

I've had an ongoing discussion with someone who shall remain nameless (to protect me!) about bike performance vs run performance. This person has made super run improvements, always places well running, but biker performance isn't on the same scale. The COVID-Break has given me a some time to put data to the question "what should me bike performance be relative to my run performance?"

I pulled some data from 4 Athletes from 2019 (table below). A couple observations:

• Athlete's A, B & C had run FTP at/below bike FTP, Athlete D (the one in question) was reversed.
• Athlete's A, B & C had run Oxygen consumption at/below bike VO2, Athlete D was reversed.

I found this to be really curious... for an athlete who trains for both running and cycling, why would oxygen consumption be materially different? I found a really interesting paper on the subject: Physiological Differences Between Cycling and Running (Gregoire P. Millet, V.E. Vleck and D.J. Bentley - paper attached). The paper is a meta study, looking at results from other studies on runners/cyclists/triathletes, men/women, different ages, different ability levels.

Physiologicsl Differences Between Cycling and Running.pdf

Some take-aways:

• VO2max is specific the the sport (you can have different numbers for each sport)
  • The cyclists had higher VO2s on the bike than on the run, the runners had higher VO2s on the run than on the bike and both were able to make improvements in their "off" sports when they trained.
• Cross training helps, however, running benefits cycling more than cycling benefits running
• Triathletes, generally have no material difference between VO2max for cycling vs running. Looking at the data, running was typically ≈2% higher than cycling.
  • The EN athletes A, B & C each had bike VO2/FTP 5-10% bigger than run VO2, D's run was ≈30-40% greater than the bike.
• HR is different between running and cycling for both maximal and sub-maximal activities. Running is always higher (we knew this).
• Ventilatory fatigue / deteriorating mechanics measured in all activates, however, it was more impaired in cycling than in running.
  • It was specifically noted that running after biking is an issue because cycling creates ventilators fatigue. Evidence that this can be improved with training (ie, do bricks). 

Putting the WKO data I pulled together with the take-aways from the paper:

1. Having power data from running (Stryd) as well as from the bike gives us an opportunity for insights into our fitness and training that we haven't had historically.
2. It can be useful to do an analysis that compares VO2 and FTP from biking and running. As triathletes, these numbers should be within a small % of each other. To the extent that they are different, eg - run VO2 is much greater than bike VO2, it could indicate an area to focus training. Doing some quick math from the table above, with training focused on raising the oxygen consumption on the bike to approach the level demonstrated on the run could result in FTP improving from ≈155w to 210w!!
3. If the 2019 data from Athletes A, B & C is somehow representative of our team, we are all probably leaving something on the table in terms of run performance and we may wan to consider working harder at our threshold runs!

Thoughts?

Shaughn Simmons

Very nice of you to put in the time to pull this together for us @Rich Stanbaugh . Looks to be a good strong, meaningful data set which you have summarized well, IMO.

2 thoughts for the readers to consider.

An athlete has one size set of heart and lungs relative to his/her extremities with which to pump oxygen and blood. Some of us more gifted than others. Biking, the act of engaging the larger quad and hamstring muscles at the same time vs. running may not allow that pump to push the blood and oxygen through both of those larger muscle sets as forcefully as it can through the running muscles given the hams and quads while biking are using so much blood and oxygen, hence a lower V02 ability biking vs. running using the same pumping system.

Pushing the run harder on the threshold runs to leverage this capacity might send us two steps backwards given the exponential risk of injury the run poses as intensity increase. Why not use the bike to safely work the pumping system (heart and lungs) into those extreme ranges, gain the fitness and decrease the risk injury otherwise inherent in threshold runs?

Dave Tallo

This really is amazing, Rich, and thanks for these data and conclusions. And for the paper. There's a lot to chew on.

My immediate curiosity about Athlete D (as an outlier) and tes A, B and C (as representative) is the role of absolute body mass on run FTP and v02 versus bike FTP and v02. More to follow, as I acknowledge the question is about the difference between oxygen uptake between the two.

tim cronk

Great stuff @Rich Stanbaugh

My question would be. How would the V02 comparison run vs. bike be if it were derived via Heart Rate and not Power? I'm guessing the outcome would still be similar ?

While I really believe the data calculations via STRYD, it is not a real PM strain gauge. The beginning (shorter time frames) of data in the power curve would be in question. The spread in run power curve is a lot smaller than bike power curve?

Rich Stanbaugh

@Shaughn Simmons thanks for the thoughtful response, you brought up some good thinking points.

An athlete has one size set of heart and lungs relative to his/her extremities with which to pump oxygen and blood. Some of us more gifted than others. Biking, the act of engaging the larger quad and hamstring muscles at the same time vs. running may not allow that pump to push the blood and oxygen through both of those larger muscle sets as forcefully as it can through the running muscles given the hams and quads while biking are using so much blood and oxygen, hence a lower V02 ability biking vs. running using the same pumping system.

The study discussed this effect (§1.1.1) and noted that exactly this phenomena occurred in cases with untrained subjects, however in runners having any prior experience in the cycling, they were able to attain VO2max equal to or approaching their running VO2. I would guess that you are exactly right about the muscle mass, but that when you switch sports, there is an associated vascularity / capillary development that overcomes the issue. All of the data in the study suggested that with training, the numbers become very close to the same for all/most athletes.

Pushing the run harder on the threshold runs to leverage this capacity might send us two steps backwards given the exponential risk of injury the run poses as intensity increase. Why not use the bike to safely work the pumping system (heart and lungs) into those extreme ranges, gain the fitness and decrease the risk injury otherwise inherent in threshold runs?

I wholehearted agree with risks associated with running injuries. Having said that, you cannot become a fast runner unless you practice running fast.

The questions become - how fast do you need run and how do you balance that with the risk of injury? When I was young, the answer was "run as fast as I can" and "injuries?" Now, about the fastest I ever run is my "aspirational" 5k speed (works out to about 110-115% of threshold). By "aspirational," I mean the 5k speed that is associated with the my target race pace. I achieve the VO2 development by controlling the number of intervals and the rest between intervals. So, even if I am doing 400m repeats, I will do them at the 5k pace, increase the repeats and do jogging recoveries for 200m. This a a much different approach than on the bike where I regularly learn lessons in humility!

My immediate curiosity about Athlete D (as an outlier) and tes A, B and C (as representative) is the role of absolute body mass on run FTP and v02 versus bike FTP and v02. More to follow, as I acknowledge the question is about the difference between oxygen uptake between the two.

Athlete C has a lower body mass than Athlete D

Rich Stanbaugh

Thanks @tim cronk

My question would be. How would the V02 comparison run vs. bike be if it were derived via Heart Rate and not Power? I'm guessing the outcome would still be similar ?

From my understanding, there are lots of different ways to estimate VO2. Traditionally, the most accurate way has been to go to a lab, jump on a treadmill/bike and do a maximal effort protocol while a technician periodically pricks your finger and measures your blood. This has been/probably still is the gold standard.

There is also a test where you wear a mask and they derive VO2 from the amount of CO2 in your breath. When I was in track in HS, we ran fixed distances at maximal effort and the coach looked the time/distance up on a table and estimated it that way (he also times us with a mechanical stop watch!).

Garmin uses an algorithm from Firstbeat to estimate VO2max based on heart rate data. Studies that I have read put this data within 5%-10% of lab data. The number that Garmin predicts for me tends to be slightly higher than the number that WKO predicts for me.

WKOs estimate is based on some complicated math on power and the PDC based on maximal efforts. When Dr, Coggan was developing this (circa 2013), cyclist were submitting their lab results and they were comparing the WKO calculations to the Lab results. The results were pretty remarkable, ≤5% difference (from memory).

Steve Palladino worked with WKO to adapt the calculations for running and achieved similar results. I believe that some of the adaption that they had to account for was to account for the short duration powers being less developed. If you do not have maximal efforts in the VO2 data ranges building out your PDC, the VO2 estimates will not be good.

I know that the four sets of data that I used were good data sets based on maximal bike efforts.

In the end, I am not certain how "accurate" any of the tests are. If the precision is good (repeatable results), then the tool become useful I think. It is especially useful because every time I load a new workout it is 'up to date' and it impacts decisions that I make about my next training session.

Rich Stanbaugh

Thanks @Dave Tallo

My immediate curiosity about Athlete D (as an outlier) and tes A, B and C (as representative) is the role of absolute body mass on run FTP and v02 versus bike FTP and v02. More to follow, as I acknowledge the question is about the difference between oxygen uptake between the two.

I considered looking at the data in relative terms (normalized to body weight). If I was going to compare absolute numbers from athlete to athlete, I think that the data would have to be normalized.

In the end, I was trying to answer the questions "how much oxygen can this person consume?" And "do they consume about the same running as biking?" So, the comparisons that I was making were either within the same person, A vs A, B vs B, C vs C and D vs D, or they were comparing ratios across athletes, eg. VO2(bike)/VO2(run).

The largest body mass is ≈140% of the smallest body mass.

Dave Tallo

@Rich Stanbaugh - so what's the work prescription for Athlete D? I'm trying to advise on a similar scenario: an experienced masters-age triathlete, lean, ectomorph, with similar time (~12 years) in his run and bike legs. I want him to take 12 weeks in the fall to do a package of bike FTP work. Thoughts?

Rich Stanbaugh

@Dave Tallo - I've spent a lot of time on this subject since April. I have "learned" a ton, but I haven't tested it on myself yet, so take this with a bigger-than-usual grain of salt. Also - there is more here than you asked for... I've been meaning to come back to this post and you gave me the impetus.

Starting with VO2 and adding to thoughts above

• Cardio Output (measured in L/min) is defined by the Fick Equation:
  • VO2 = Max Cardio Output = HR x Stroke Volume = arterial-venous O2 difference (AVO2)
  • AVO2 is the difference between O2 in the blood leaving your heart and the blood coming back into your heart (ie, how much O2 you have consumed)
  • So - the more muscle mass that blood goes through, the more capillary development that one has, the more O2 will be consumed.
  • However, in general, humans are not limited by their ability to consume O2 (right half of the Fick Equation), but rather by their ability to "deliver" O2 to the muscles (left half of the Fick Equation).
  • The reason that I started with this point is that we really should not expect our VO2 to be "identical" from one sports that engage significantly more muscles. For example, nordic skiing engages arms and legs and trunk, whereas running may use all those muscles, but really is more dependent on legs and so will have a lower VO2 potential. Similarly, biking engages fewer muscles than running and should have a slightly lower VO2 potential than running.
  • The data suggests that run and bike VO2 for triathletes is very close, so from a training perspective, I would target trying to achieve a bike VO2 that is approaching the Run VO2. If the Bike VO2 is higher, you are probably leaving something on the table with the Run. If the Run is substantially higher, then I would focus on the Bike.
  • Plasma Volume. There is some great research relating blood volume to performance that shows increased volume is "necessary" for performance, but it is not sufficient for elite performance. The study I saw demonstrated that increasing the blood volume (plasma only - not adding red blood cells) of untrained cyclist was sufficient to increase their VO2, however adding plasma only had no effect on the VO2 of trained cyclists.
  • There is actually a condition called 'athletic anemia' that happens when training volume ramps causing plasma to increase. Until the RBCs catch up, athletes can become anemic because of relatively low volumes of RBCs.
  • This actually lines up really well with our experience... the beginning of any training cycle is typically a few weeks of easy training to establish a base. Physiologically, one of the important things that happens during this phase is that blood volume increases. This leads to immediate short-term training increases and it provides the blood volume needed for further gains.
  • Depending on the current fitness of the person you are advising, it is important to have established several weeks of routine training to prepare the body for further adaptions. Going directly to the "hard" work without building the training base will delay any performance benefits from targeted intervals. Best to spend a few weeks building the volume and preparing the body for the harder work.
• This leaves Stroke Volume as the main driver for increasing performance. I have a ton more to learn about this - I am fascinated and this is what I am focused on right now.
• The left half of the Fick Equation is all about how much blood-O2 your heart can pump to your muscles. Without going into a ton of detail, there are really two factors that determine this: Blood Changes and Stroke Volume.
  • Wile training aerobically, there are two main changes to our blood. The volume increases and the absolute number of red blood cells - RBCs - increases). The average adult has on the order of 4.5-5.5L of blood and 40-50% of that volume is from RBCs; that volume increases 35-45% for highly-trained endurance athletes. First the plasma increases; over time the RBCs increase.
  • Unlike other muscle types, cardiac muscle is all one fiber type and all the cells are fired by the same nerve triggers. How hard they fire is related to the signal they receive, but they all receive the same signal. As opposed to skeletal muscle that has the Types I/II/... and is ever only partially activated
  • There is another mechanism called the Frank-Starling relationship that says cardiac output is directly related to venous return... my simple way of thinking about this is that your stroke volume (Ventricle) is directly determined by how much blood is in the Atrium waiting to be pumped.
  • So - if we want to train the stroke volume (output), we need to load up the venous return (input). Just like all other workouts, we need to design a workout that targets (stresses) the system we want to improve. In order to improve VO2, we need to target stroke volume improvements.
  • Clearly, we can do this by doing intervals... classic VO2 intervals are 5 minutes on / 5 minutes off. What I have learned though is that there is nothing magical about 5 minutes. VO2 intervals can range from about 3 minutes to 8+ minutes (how long can you tolerate?). Some people are inherently better / more efficient in the 3-5 minute zone and others gravitate to the 5-8 minute range. From a cardiac remodeling perspective, all these time ranges will trigger the same development. It is a matter of choosing intervals / reps that the athlete can tolerate. 8x3, 5x5, 12x2. The only real difference is that each interval has a 'ramp-up' phase in the beginning work is being done but the athlete is not in the VO2 zone. So, shorter intervals may require more repeats to hit the TiZ targets.
  • Traditionally, we would have stopped the interval when power dropped. Traditionally, we would have built the interval around a target power. I am suggesting to start the interval at a power above the reasonable power, maintain the power as long as possible, then let the power drop and hold it for the entire planned duration at MAX RPE and near-max HR.
• One more change... we should do these intervals at 110+ RPM.
  • The faster RPM will increase the muscle-pump from the legs. Increased muscle pump will increase the venous return which will (Frank-Starling Law) increase the stroke volume.
  • The high RPM will lower the power that we are able to maintain (high RPM burns some ATPS differently than lower RPM).
  • The high RPM will let us maintain the power longer because it will recruit primarily smaller motor units (Type I fibers).
  • Our goal is to target stroke-volume development and high rpm will increase the cardio pre-load. This will increase the cardio output. If we are at max HR, the only way for cardio output to increase is for the stroke volume to increase.
• Another key learning: There is no single power associated with maximal oxygen uptake. I know that we have been doing this for years... we call our 5' max VO2, or we look at WKO's Power at VO2 and refer to that as our VO2 power. The reality is that a person can achieve maximal oxygen uptake at different powers... and once we have achieved maximal oxygen uptake, we can maintain that state while our power changes. ( one note - the VO2 oxygen consumption estimated by WKO seems to be relatively accurate and compares to lab estimates. The power WKO shows for VO2 is certainly a power that can be used to achieve VO2 for the indicated time... it is just not the ONLY way to reach VO2. Max oxygen uptake (VO2) can be reached with different power-time combinations.)
  • How do we "measure" that we are at VO2? When we are "gasping like a fish out of water." Max HR is an indicator too - the problem with HR is that Max HR can vary from day to day. My HR typically goes up pretty quickly, then it starts to plateau. If my daily max HR is going to be 180 bpm, the time it takes to get from 176-180 is much much longer than the time to get to 175 bpm.
  • One possible way to design and manage a VO2 interval is to choose the maximum power that I think I can maintain for the interval, add 20-30% and start and try to maintain that power. My HR will go to whatever it is going to max at that day... then my power will start to drop. The objective is to KEEP PEDALING MAX RPE, maintain the max HR within a few BPM and IGNORE THE POWER. My body will be at max cardio output even though the power is dropping. Shift, do whatever is necessary to maintain the RPE and near-max HR.
• New VO2 Intervals
  • Accumulate 15 minutes in the target VO2 Zone
  • Plan roughly 30" of the interval to reach max output, so ≈4x5', 6x3', etc in order to get 15' in zone
  • Recovery ≤ 1:1 with interval. Shortening the recovery gets us back into the zone faster
  • Enter the interval 20-30% harder than you can can be maintained for the interval length
  • Target 110 RPM for the entire interval
  • Once near-max HR is reached, maintain max RPE and HR - power is irrelevant

12-Week Plan

• To your question - I would train Athlete D in the same manner that I would train myself. Since D's bike performance (VO2) is substantially lower than the bike, I would put together a bike-focused plan
• I would make certain that a decent weekly volume base was obtained before the 12-week plan (4-ish weeks riding at least 4-5x each week
• I would break the 12 weeks into:
  • 1 week testing
  • 6 weeks VO2
  • 5 Weeks FTP
• Testing: I would do 3 days of maximal testing to get good estimates for the WKO model so that I had a base to design workouts and measure improvement. Day 1: Pmax, 30", 1', 3', Day 2: 5', 20' and Day 3: 60' TT, other days recovery rides
• first 6-week block into a VO2 focused block. 2 VO2 workouts each week, add in some Z1/Z2 recovery and volume. Start with 6x3' and progress towards 2x10'. Find out what works best for the athlete.
• Last 5 weeks do an FTP progression: 3x10', 3x12', 2x15', 20' test, 1 hr Time Trial

One last point - I have read a ton, talked to a ton of people, listened to a ton of podcasts to try learning about this. It is hard to know who all I should credit - Kolie Moore's Empirical Cycling and subsequent conversations with him is a key source of my learning. Some re-reading ancient postings from teh WKO forum, some from reading anatomy texts and a ton of papers. The concept of creating pre-load to drive stroke volume came directly from him. Some of the ideas are mine and some from other reading. I want to make sure to give the impression that this posting is a smorgasbord :-)

Coach Patrick

Wow, this is super great stuff @Rich Stanbaugh and @Dave Tallo .

Presentation is also really good...I like the functional changes (per workout prescription).

Random additions...just doing some performance testing with INSCYD and the data really gives depth to VO2 (range allows for short or long intervals) and the VLaMax metric (maximum glycolytic power) number tells me which VO2 you suck at (low VLaMax struggle with high VO2 numbers).

Note the the testing approach is built on overachieving early, as you have described. So a 3' interval needs to be paced as a 1' interval and then you hang on.

It also struck me we could do some "primed" intervals, and I have written those...both Primed + Threshold, as well as Primed + Lactate Shuttling. I even have a VO2Mix workout where the shorter/harder stuff "primes" the athlete and then they can do longer more sustainable efforts albeit at a lower power).

#overlap !! 😉

~ Coach Patrick