Peter Attia MD331‒ Optimizing endurance performance: metrics, nutrition, lactate, & insights from elite performers
CHAPTERS
- 0:00 – 3:24
Olav's Applied, Engineering-Driven Approach to Elite Athletes
Peter Attia welcomes Olav back and asks him to recap his work. Olav describes his engineering background and the 15 years of in-depth, longitudinal testing he has done on two of the fittest athletes in the world, including developing technology along the way. He says his athletes span triathletes, cyclists, runners, track and field athletes and sailors.
- •Olav's background is in engineering and guides his applied approach
- •15 years of extreme in-depth, longitudinal studies on two elite athletes
- •Technology sometimes has to be developed to get more granular understanding
- •Athletes include triathletes, cyclists, runners, track and field athletes and even sailors
- 3:24 – 13:23
Defining FTP, Critical Power, and Anaerobic Threshold
Olav explains the original FTP protocol attributed to Andy Coggan, how it has been diluted by variations, and how critical power differs. They discuss why condensing power into one number loses the duration dimension, and where anaerobic threshold sits relative to these metrics.
- •FTP originally meant a five-minute effort, a pause, then a 20-minute effort minus 5%; consistency of protocol matters most
- •Critical power comes from multiple all-out efforts and mathematical extrapolation
- •A single number drops the duration axis, so tracking one-, five- and 15-minute efforts is more informative
- •Anaerobic threshold is not a switch to anaerobic metabolism and normally sits a few percent below critical power or FTP
- 13:23 – 25:44
Lactate Threshold, Testing Protocols, and Lactate Concentration
They discuss why lactate threshold reflects measurement limitations, how step protocols work, and why lactate concentrations vary with hydration, blood volume and athlete type. Olav explains that elite endurance athletes can have a second turn point below 2 mmol and that utilization of VO2 max is very high in elite marathoners and the Ironman athletes he works with.
- •Lactate is produced all the time; the threshold depends on when it appears in blood
- •Step length matters less when lactate concentration is used as a field guide
- •Hematocrit changes of over 10% during a session can influence lactate concentration
- •Elite endurance athletes may show a steady-state turn point below 2 mmol
- •Elite marathoners work at roughly 94-95% of VO2 max at threshold
- 25:44 – 38:48
What VO2 Max Is, Field Versus Lab Testing, and the Breathing Trick
Olav defines VO2 max and Peter shares how he tests it on hills and a StairMaster. They digress into Formula 1 heart rates and muscle blood-flow restriction, then compare VO2 max across cycling, running and swimming. Olav explains that restricting breathing can artificially inflate VO2 max numbers without making an athlete fitter.
- •VO2 max is the maximum oxygen consumed over a minute, and field efforts can exceed graded lab tests
- •High heart rates in Formula 1 may partly reflect G-forces and reduced preload
- •VO2 max differences between swimming, cycling and running have disappeared in Olav's athletes through training
- •Restricting breathing near exhaustion can push VO2 max above 70 from a baseline of 60 but is cheating the protocol
- •Measurement location and reproducibility matter more than absolute numbers
- 38:48 – 48:07
How to Get a Reliable VO2 Max Test: Preparation and Warm-Up
Peter asks what people should look for when getting a VO2 max test at a public facility. Olav recommends a standardized approach: being fueled, hydrated and rested, testing at a consistent time of day, and using a structured warm-up with short efforts progressing toward VO2 max.
- •Keep a consistent pre-test routine, including food, hydration and sleep, without over-standardizing life
- •Timing of day should stay consistent because of circadian body-temperature rhythm
- •A warm-up of about 20 minutes with easy, moderate, threshold-feel and short VO2 max-type efforts
- •Carbohydrate before the test matters little for maximal efforts, though it can influence submaximal RQ
- 48:07 – 57:43
VO2 Max Instruments and Repeated Testing in Elite Athletes
Olav lists the metabolic devices his lab uses, including mixing-chamber systems and the VO2 Master, and describes validating them with a lung simulator. He discusses how his athletes' VO2 max fell when they moved to Ironman, why VO2 max is one-dimensional, and how repeating tests ten minutes apart can yield higher values.
- •Devices used include the Jaeger Oxcon Pro, Parvo, and VO2 Master; machines are independently validated with a lung simulator
- •Kristian has exceeded 7.7 liters per minute in absolute oxygen uptake, roughly 100 ml/min/kg
- •VO2 max declined when the athletes moved from Olympic to Ironman focus
- •Capacity is the best predictor but too impractical to test, so VO2 max is used
- •Second and third tests ten minutes apart often give higher VO2 max than the first
- 57:43 – 1:02:32
Beetroot Concentrate and an Adaptogen's Effect on VO2 and Efficiency
Olav discusses beetroot concentrate and nitrate as vasodilators that tend to help amateurs but show no effect on VO2 max in his elite athletes. He then describes a pine-bark-derived adaptogen from Plasmid that raised VO2 and lowered RQ in Kristian while efficiency worsened and heat production rose.
- •Nitrate converts to nitric oxide, aiding vasodilation, with roughly 5% effects in amateurs taken with a grain of salt
- •No VO2 max effect from beetroot was seen in the elite athletes
- •The adaptogen led to higher VO2 and lower RQ, around 0.95 instead of 1.05
- •Efficiency declined, with VO2 rising more than power and higher heat production seen via temperature pill
- 1:02:32 – 1:14:04
Interpreting RQ, Lactate's Energy Contribution, and Fatigue in Long Races
They discuss why RQ-based fuel tables become unreliable in elite endurance athletes, since lactate can contribute a significant share of energy. Olav also explains that VO2 max and thresholds shift over the course of an Ironman, which is how athletes can hold a very high fraction of VO2 max.
- •RQ above about 0.92-0.93 in extreme endurance athletes means lactate becomes a meaningful energy contributor
- •Standard RQ tables stop at 1.0 but Olav is skeptical even in the low 0.90s for elite athletes
- •Isotope tracers are used to see how much ingested carbohydrate is utilized
- •VO2 max and thresholds decline over a race, so race pace becomes close to the athlete's current max
- •Ironman specificity means de-prioritizing high one- and five-minute power
- 1:14:04 – 1:18:14
Triathlon Distances, Elite Metabolism, and Swimming Inefficiency
Peter lays out Olympic, half Ironman and Ironman distances and asks whether one athlete can be elite at all. Olav says the triathletes' VO2 max matches or exceeds specialists, but their efficiency is lower because training is split across three sports. He gives an example of an elite swimmer using 25% less oxygen than Kristian at the same swim speed in a flume.
- •Distances covered: Olympic, half Ironman and Ironman, with Kristian holding the Ironman debut record of 7:21
- •Triathletes' VO2 max equals or exceeds specialists, but efficiency is lower
- •Swim efficiency is especially poor relative to specialists
- •A bronze medalist swimmer used 25% less absolute oxygen than Kristian at the same speed in a flume
- 1:18:14 – 1:33:33
Drag in Swimming, Biofeedback, and Running Power Meters
Peter proposes that swimming has the most potential for radical gains through drag reduction, like aero changes in cycling. Olav turns to biofeedback tools in cycling and running, including Stryd, and explains why running power is still debated, why he prefers raw mechanical numbers, and how power and VO2 max relate across modalities.
- •Peter's thesis that drag reduction could transform swimming, drawing on cycling's aero history
- •Cycling has power meters and GPS; running now has insole force-plate or foot-mounted motion-capture power meters
- •Olav uses only raw numbers rather than normalized power or modeled metrics
- •Running power lacks an agreed standard for what to output
- •At VO2 max, watts per kilo and VO2 max roughly match, but not at submaximal intensities
- 1:33:33 – 1:46:35
Pacing by RPE, Heart Rate and Power, and the Heart's Use of Lactate
Peter asks how athletes prioritize RPE, heart rate and power in an Ironman. Olav says RPE is most important in racing, illustrating how perceived effort changes with duration, and explains why a heart trained for the wrong demands is inefficient. They touch on the heart's use of lactate as a fuel.
- •In long races athletes should not be slaves to numbers; RPE is the most important
- •Duration is a hidden dimension of RPE, as shown by the Kipchoge example at the same pace
- •Max heart rate cannot be reached at the end of a race
- •The heart is an energy-hungry muscle, so excessive VO2 max training can be inefficient
- •The heart likely uses lactate richly as fuel
- 1:46:35 – 1:55:34
Lactate as Fuel and Buffering with Bicarbonate Hydrogel
Peter explains why hydrogen ions rather than lactate cause trouble and asks about buffering. Olav describes bicarbonate delivered in a hydrogel, Gustav's highest VO2 max under bicarbonate, and the large differences in blood lactate response between Gustav and Kristian. They speculate about MCT transporters, plasma volume, stroke volume and compensatory systems.
- •Lactate is a potential artificial fuel that could also act as a buffer in salt form
- •Maurten's hydrogel packaging lets large amounts of bicarbonate be delivered without gastric problems
- •Some athletes show nearly double blood lactate with bicarbonate while Kristian shows almost no change
- •Blood lactate concentration does not equal glycolytic contribution
- •Kristian and Gustav differ massively in stroke volume, plasma volume and hemoglobin mass yet compensate to similar performance
- 1:55:34 – 1:58:35
Acetaminophen and Heat: Manipulation Versus Natural Adaptation
Peter asks about high-dose acetaminophen for heat performance. Olav says his team doesn't use it, preferring natural adaptation to heat and pain over acute manipulation of one system, and notes cycling got faster after doping scandals thanks to other techniques. Peter suggests testing it in training, and Olav asks where heat from mechanical power comes from.
- •Olav's team does not use acetaminophen
- •He prefers natural heat adaptation built into protocols over acute interventions
- •Peter says the proposed mechanisms are blunted temperature and pain governors
- •Olav notes heat arises from mechanical power and possible reduced work could carry a capacity cost
- •Peter suggests alternating weeks with and without it as a placebo-style test
- 1:58:35 – 2:17:34
Nutrition Evolution, Rider Weight, and Fueling at 160+ Grams per Hour
Peter contrasts today's carbohydrate intake and smaller riders with the Lance Armstrong era. Olav says nutrition is the main reason racing is faster and predicts athlete weights may rise again. He describes athletes taking 160 grams per hour and even utilizing over 240 grams per hour, trained like any muscle, with a 40/60 fructose-glucose hydrogel mix.
- •Lance Armstrong said they didn't track carbs and ate when hungry, while today's athletes fuel precisely
- •Olav expects athlete weights to rise again as fueling supports growth
- •Kristian's watts per kilogram rose as his weight increased
- •Gut tolerance for very high carb intake is trainable; 240+ grams per hour of utilization was measured
- •Pure glucose and fructose is chosen to save oxygen for propulsion
- 2:17:34 – 2:21:31
Ketones and the Limits of Alternative Fuels
Peter asks about beta-hydroxybutyrate. Olav explains that stoichiometry alone is insufficient, and that the fuel demand of racing means carbohydrate replacement is the aim. He says ketones can't practically replace fuel at the needed quantities and instead induce a state in the body.
- •Stoichiometry isn't enough; enthalpy, Gibbs free energy and entropy matter
- •Racing strategy aims to replace as much carbohydrate as possible so more power comes from the same source
- •Ketone intake of the needed magnitude could create a bonk-like, diabetic-feeling state
- •Ketones act more as inducing a state than as a replacement fuel
- 2:21:31 – 2:31:09
Olympic Lessons: Tokyo to Paris, the Seine, and Drafting
Peter asks about Kristian's Tokyo win versus his Paris result. Olav says they became sloppy with intensity control in the last year and that the Seine's heterogeneous current plus drafting forced Kristian to push about 30% harder for much of the bike, costing him on the run.
- •Kristian won in Tokyo and finished 12th in Paris after moving back from Ironman
- •A big shift in intensity control drove progress from Rio to Tokyo; complacency crept in before Paris
- •The current and heterogeneous flow of the Seine changed the swim dynamics
- •Short-course drafting forced about 30% more power over roughly 25 km to rejoin the group
- •He lost about four seconds per kilometer on the 10K, the gap between first and 12th
- 2:31:09 – 2:41:04
Using AI in Elite Training and Research
Peter asks where Olav uses AI. Olav describes an agentic system with language models as the interface, argues AI can supercharge research and coaching with better consistency and proactive adjustments, and compares its potential impact to Norway's winter sports technology.
- •Olav's team has built agentic AI systems since around 2020 to highlight important new data in real time
- •AI helps surface patterns, such as Kristian's utilization rising too high before the Olympics
- •AI struggles at edge cases requiring extrapolation, and the user still has to produce the questions
- •It would make coaching more precise rather than replace it, with gains from consistency
- •Biggest hoped-for gain is being more proactive, for elites and amateurs alike
