The word "threshold" appears in more cycling conversations than almost any other training term. And almost every time it comes up, the person using it means something slightly different from the person hearing it.
One rider says threshold and means FTP — the number from their last 20-minute test. Another means the burning sensation in their legs on a hard climb. A coach means LT2, the second lactate inflection point. A physiology paper means MLSS, the maximal lactate steady state. A podcast guest says "lactate threshold" and could mean any of four distinct measurements, depending on which decade they trained in and which textbook was on their shelf.
This is not pedantry. These distinctions reshape the way you set training zones, the way you execute intervals, and the way you structure a training block. Get them confused and you end up training in the wrong place — often for months — without understanding why the adaptations aren't landing.
The confusion problem
Cycling borrowed its threshold language from exercise physiology, but it borrowed selectively and sloppily. The result is a vocabulary where one word covers at least four different physiological and performance concepts:
- FTP — functional threshold power, a performance metric
- LT1 — the first lactate threshold, also called the aerobic threshold
- LT2 — the second lactate threshold, also called the anaerobic threshold or OBLA (onset of blood lactate accumulation)
- MLSS — maximal lactate steady state, the true equilibrium point
These are related. They live in the same neighbourhood of your physiology. But they are not synonyms, they do not occur at the same intensity, and they do not respond to the same training in the same way. Using them interchangeably is like using "speed" and "acceleration" as if they mean the same thing — close enough in casual conversation, catastrophic if you are building an engine.
The problem compounds in online training forums, where someone posts asking about their "threshold" and gets twelve answers referencing four different definitions. Each answer is internally consistent. Collectively they are contradictory. And the rider asking walks away more confused than when they started.
So let's separate them.
What FTP actually is
Andrew Coggan introduced functional threshold power in the early 2000s as part of the Training and Racing with a Power Meter framework he developed with Hunter Allen. The definition is specific: FTP is the highest power a rider can sustain in a quasi-steady state for approximately one hour.
That word "approximately" does more heavy lifting than most people realise. Coggan was explicit that FTP is a practical training tool, not a laboratory measurement. It gives you a number you can anchor zones to, track over time, and use to prescribe interval intensity. It was never designed to represent a precise physiological boundary.
The standard testing protocol — ride as hard as you can for 20 minutes, multiply by 0.95 — was a pragmatic shortcut. Most riders cannot or will not do a full hour-long time trial in training, so the 20-minute test with a correction factor became the default. And that correction factor is where the first crack appears.
The 0.95 problem
The assumption behind the 0.95 multiplier is that your 20-minute power represents roughly 105 per cent of your one-hour power. For some riders — those with a well-developed aerobic engine and a relatively modest anaerobic contribution — that holds. For others, it does not.
Riders with a strong anaerobic system can sustain a disproportionately high percentage of their peak power for 20 minutes. Their anaerobic energy contribution inflates the 20-minute number. Apply 0.95 to an inflated starting point and you get an FTP that sits above what they can actually sustain for an hour.
Tim Cusick, who helped develop WKO5 and has spent decades working with power data, has spoken extensively about this. His analysis of thousands of power files shows that the ratio between 20-minute power and true one-hour power varies significantly across riders — from 0.90 to 0.97 depending on physiology, training history, and fibre-type composition. The flat 0.95 correction is an average, and averages are never individual.
This matters because every training zone you set from FTP inherits whatever error exists in that number. If your FTP is 5 per cent too high, your Zone 2 ceiling is too high, your sweet spot range is too high, and your threshold intervals are prescribed at an intensity you cannot sustain for the target duration.
LT1: the first lactate threshold
LT1 is the intensity at which blood lactate concentration first rises measurably above resting levels — typically from a baseline of around 0.8-1.0 mmol/L to roughly 1.5-2.0 mmol/L. Below LT1, your body clears lactate as fast as it produces it. Above LT1, lactate begins to accumulate, slowly at first.
This is your aerobic threshold. The ceiling of true Zone 2 training. The boundary below which you are riding on fat oxidation and aerobic metabolism with minimal lactate accumulation.
Iñigo San Millán brought LT1 into mainstream cycling conversation through his work with Tadej Pogacar and UAE Team Emirates. San Millán's research at the University of Colorado focused on the relationship between LT1, mitochondrial function, and fat oxidation. His argument — backed by published data — is that training below LT1 drives the mitochondrial adaptations that form the aerobic base everything else sits on. Pogacar's training reportedly includes a substantial volume of work carefully kept below this first threshold.
The significance for self-coached cyclists is this: if you do not know where your LT1 sits, you do not know where your Zone 2 actually ends. And if you are using FTP-derived zones to set your Zone 2 ceiling, you might be above LT1 without realising it. The standard Coggan Zone 2 range is 56-75 per cent of FTP. For many riders, the upper end of that range — 75 per cent of FTP — sits above their LT1, particularly if their FTP is itself set too high.
Stephen Seiler, who has published more research on training intensity distribution than perhaps any other sports scientist, frames it differently but arrives at the same place. His three-zone model puts everything below LT1 in Zone 1 (easy), everything between LT1 and LT2 in Zone 2 (moderate), and everything above LT2 in Zone 3 (hard). In Seiler's framework, the polarised training approach means spending roughly 80 per cent of your time below LT1 and the hard sessions above LT2. The territory between the two thresholds — the moderate zone — is where you spend the least time.
How LT1 shifts with training
LT1 is not fixed. It rises with aerobic training. As your mitochondria become more numerous and more efficient, your body can handle higher work rates before lactate starts to accumulate. This is the adaptation that makes Zone 2 training so valuable — you are directly targeting the physiology that determines where LT1 sits.
A well-trained cyclist might have an LT1 at 65-75 per cent of their VO2max. An untrained person might see LT1 at 50-55 per cent. The gap between those numbers represents years of aerobic development — capillary density, mitochondrial volume, enzyme activity, fat oxidation efficiency. None of it is visible on a power meter screen, but all of it determines how hard you can ride before your body starts running up a lactate tab.
LT2: the second lactate threshold
LT2 is where most cyclists instinctively think "threshold" lives. It is the intensity at which lactate production begins to exceed clearance in a sustained, accelerating way. Blood lactate concentrations at LT2 typically sit around 4 mmol/L — the classic OBLA (onset of blood lactate accumulation) reference point, though the exact concentration varies between individuals.
Above LT2, you are accumulating lactate faster than your body can process it. Time to exhaustion shortens rapidly. The effort becomes unsustainable not because your muscles cannot produce the power, but because the metabolic byproducts of that power production build up faster than they can be cleared.
LT2 sits closer to what most people think FTP represents. And in many riders, FTP and LT2 are within a few watts of each other. But "close" is not "identical," and the gap between them varies with training status, testing protocol, and individual physiology.
The zone between LT1 and LT2
This middle territory — above LT1, below LT2 — is what most models call tempo or Zone 3. It is also where a disproportionate number of self-coached cyclists spend most of their training time.
The reason is intuitive. Below LT1 feels too easy. Above LT2 is hard and requires structured recovery. The tempo zone feels productive — you are breathing harder, your legs are working, you finish the ride feeling like you did something. But from a physiological standpoint, this intensity is neither fish nor fowl. It is too hard to maximise the aerobic adaptations that come from sub-LT1 riding, and too easy to drive the lactate clearance and muscular endurance adaptations that come from working at or above LT2.
Seiler calls this the "black hole" of training intensity. Not because it is useless — tempo riding has a place in certain race-specific preparation — but because defaulting to it day after day produces mediocre adaptations across the board rather than strong adaptations in any direction. The rider who spends four days a week at tempo gets slower at going easy and slower at going hard. The polarised rider who spends three days below LT1 and two days above LT2 gets better at both.
MLSS: the threshold behind the threshold
Maximal lactate steady state is the highest exercise intensity at which blood lactate concentration stabilises over time rather than continuing to rise. It is the true equilibrium point — the hardest you can go where production and clearance are balanced.
MLSS testing is not practical for most cyclists. The gold-standard protocol requires multiple 30-minute constant-load efforts on separate days, with blood lactate measured throughout each effort. If lactate stabilises over the final 20 minutes of a 30-minute effort, that intensity is at or below MLSS. If it keeps climbing, you are above it. Finding the exact boundary requires testing at several intensities across several sessions.
This is what FTP was supposed to approximate. Coggan's one-hour-power concept was designed to land near MLSS — the idea being that the highest power you can sustain for 60 minutes would, by definition, sit close to the intensity where lactate production and clearance are balanced.
For riders whose FTP is well-calibrated — usually through actual hour-long efforts or through careful multi-duration testing — FTP and MLSS are indeed close. Within 5 watts for many athletes. But for riders relying on a 20-minute test with the standard 0.95 correction, the gap can be meaningful. Research published in the International Journal of Sports Physiology and Performance has shown that FTP derived from 20-minute tests can overestimate MLSS by 5-10 per cent in some populations. That is 15-30 watts for a rider with a true MLSS of 280W. Not a rounding error.
Why the distinction reshapes your training
Take a rider — call him Mark — who does a 20-minute test and holds 300W. He multiplies by 0.95 and sets his FTP at 285W. He builds all his training zones from that number.
But Mark has a strong anaerobic engine. His actual MLSS sits at 265W. His LT2, confirmed by a lab test, is at 260W. His LT1 is at 195W.
Using 285W as his FTP anchor:
- His Zone 2 ceiling (75 per cent of FTP) sits at 214W — nearly 20 watts above his actual LT1
- His threshold training zone (95-105 per cent of FTP) sits at 271-299W — the upper end is 39 watts above his LT2
- His sweet spot range (88-94 per cent of FTP) sits at 251-268W — overlapping his actual LT2
Every session Mark does is shifted. His easy rides are not easy enough. His threshold work is too hard to sustain. His sweet spot sessions are actually threshold sessions. He is doing the wrong workout at every intensity, and his training log looks fine on paper.
This is not hypothetical. It is one of the most common training errors in amateur cycling, and it is almost invisible from the outside. The rider is structured, consistent, doing all the right session types. The zones are just anchored to the wrong number.
The cascade effect
When FTP is set too high, the damage compounds over a training block. Mark's Zone 2 rides — now sitting above his LT1 — fail to produce the deep aerobic adaptations that sub-LT1 riding delivers. His mitochondria get a moderate stimulus instead of the clean, high-volume, low-lactate environment where they multiply fastest.
His threshold sessions — now sitting above his LT2 — become VO2max efforts in disguise. He cannot complete them at the prescribed power. He either bails halfway through the second interval or drops 10 watts below target and finishes the session at an intensity that does not match the training prescription. His heart rate drifts upward across the interval set. His recovery between sessions takes longer than expected.
After six weeks, Mark retests. His FTP has not moved. He assumes he needs more volume, or harder sessions, or better recovery. The actual problem is that his training zones were wrong from the start, and every session delivered a different stimulus than the one his plan intended.
This is fixable. But the fix requires understanding that "threshold" is not one number.
How to find each threshold
Finding LT1
The most accessible method is the talk test. If you can speak in full, comfortable sentences without your breathing interrupting the flow, you are below LT1. The moment your breathing starts to fragment your speech — not gasping, just the first noticeable change — you are at or just above LT1.
For a more precise estimate, a structured aerobic threshold test works well. Ride at a steady, easy effort for 30 minutes and record your average heart rate and power for the final 10 minutes. Then increase the intensity by 10-15 watts and ride another 10 minutes. Continue stepping up. LT1 sits roughly at the intensity where your heart rate begins to drift upward relative to power — the point where the linear relationship between heart rate and intensity starts to steepen.
Iñigo San Millán's preferred method involves blood lactate testing. A finger-prick lactate measurement at each step of a ramp test identifies LT1 as the first sustained rise above baseline — typically the step where lactate moves from below 1.5 mmol/L to above 2.0 mmol/L. Portable lactate meters from Lactate Plus or Lactate Pro make this feasible outside a lab, though the technique takes practice and the strips are not cheap.
Finding LT2
A well-executed 30-minute field test provides a reasonable approximation. Ride the hardest effort you can sustain for a full 30 minutes — paced evenly, not a fast start that fades — and take the average power of the final 20 minutes. This lands closer to MLSS than the standard 20-minute test because the longer duration reduces the anaerobic contribution that inflates shorter tests.
With lactate testing, LT2 shows up as the second inflection point in the lactate curve during a ramp test — typically around 4 mmol/L, though individual variation means some riders see it at 3.5 and others at 5.0. The absolute number matters less than the shape of the curve: the point where lactate concentration starts to rise steeply rather than gradually is LT2, regardless of the exact millimolar value.
Heart rate at LT2 is also a useful anchor. For most cyclists, LT2 sits at roughly 85-92 per cent of maximum heart rate. Combined with a field test, this gives you two reference points — a power number and a heart rate number — that together produce more robust training zones than either one alone.
Finding MLSS
True MLSS testing is impractical for most self-coached riders. It requires multiple 30-minute constant-power efforts across several days, with blood lactate measured every five minutes. The intensity where lactate stabilises rather than rises is your MLSS.
The practical workaround is to use a longer field test (the 30-minute protocol above) and compare the result against your 20-minute FTP estimate. If the 30-minute number sits more than 5 per cent below your calculated FTP, your FTP is probably set above your MLSS and your zones need adjusting downward.
Tim Cusick's work with WKO5 offers another approach: modelled FTP derived from a power-duration curve built from multiple maximal efforts at different durations. This mathematical model estimates the power you could sustain at the deflection point of the curve — typically closer to MLSS than a single 20-minute test produces. It requires a good spread of recent maximal efforts across durations from one minute to sixty minutes, which accumulates naturally over a racing season but takes deliberate testing if you do not race.
What this means for your training
The practical consequence of understanding these four thresholds is not that you need to test all of them with laboratory precision. It is that you anchor your training zones to the right reference points.
Zone 2 should be anchored to LT1, not to a percentage of FTP. If you know your LT1 — even approximately from a talk test — use that as your Zone 2 ceiling. If you are using FTP-based zones, check that the upper boundary of Zone 2 does not exceed your LT1. If it does, cap your Zone 2 rides at the lower number.
Threshold work should be anchored to LT2 or MLSS. The threshold intervals you prescribe yourself should target the power at which lactate production and clearance are roughly balanced — your MLSS — or just below. If your FTP sits above your MLSS, doing intervals "at FTP" means you are doing intervals above threshold. Scale back.
The space between LT1 and LT2 — tempo — should be used deliberately, not by default. If you find yourself riding in this zone most days, you are in no-man's-land: too hard for aerobic base building, too easy for threshold adaptation. Seiler's polarised model suggests spending the minimum necessary time here. Easy days below LT1. Hard days above LT2. The middle only when race-specific preparation demands it.
A practical framework
For a rider who has done a 20-minute test and wants to validate their zones without access to a lab:
-
Do the talk test on your next easy ride. Note the power and heart rate where full-sentence speech starts to fragment. That is your approximate LT1. If it sits below 75 per cent of your FTP, your FTP-derived Zone 2 is probably too high.
-
Do a 30-minute field test. Ride the hardest steady effort you can sustain for 30 full minutes. Take the average power of the final 20 minutes. Compare it to your 20-minute FTP estimate. If the gap is more than 5 per cent, adjust your FTP downward.
-
Cross-reference with heart rate. Your LT2 heart rate — from the field test — should sit at roughly 85-92 per cent of your max. Use this as a secondary check during threshold sessions. If your heart rate at "threshold power" sits above 92 per cent of max within the first interval, the power target is probably above your LT2.
-
Retest every 6-8 weeks. These markers shift with training. As your aerobic base develops, LT1 rises. As your threshold fitness sharpens, LT2 and MLSS rise. Zones set in January may not be accurate in April.
You do not need perfect numbers
I want to be clear about this because the physiology can make the whole thing feel more complicated than it needs to be. You do not need a lactate meter. You do not need a lab. You do not need to know your LT1 to the nearest watt.
What you need is an understanding that "threshold" is not one thing. That FTP is a tool, not a truth. That the number from your 20-minute test might sit above the intensity your body can actually sustain in a steady state. And that if your easy rides feel harder than they should, or your hard sessions always leave you wrecked rather than productively fatigued, the problem might not be your fitness — it might be your zones.
The science has finally caught up with what experienced coaches have known for years. San Millán's work with Pogacar made LT1 a mainstream topic. Seiler's research gave us the data to back up polarised training. Cusick's power modelling gave us better tools to estimate threshold from real-world data. The information is available now in a way it was not a decade ago.
The practical application is simpler than the physiology suggests. Ride easy below LT1. Ride hard above LT2. Be suspicious of any training plan that puts you in the middle most of the time. And if your FTP was set from a 20-minute test and you have never validated it against a longer effort, do the 30-minute test. It takes half an hour and it might rewrite your entire training block.
If you want to go deeper on this — discuss your own test results, compare notes with riders working through the same questions, and get direct input on how to set your zones properly — the Roadman Cycling community on Skool is where those conversations happen daily. Real riders, real data, no fluff.
The good news is that this is fixable. All of it. Wrong zones, misaligned thresholds, months of grey-zone riding — it all corrects itself once you anchor your training to the right reference points. The understanding you now have of what threshold actually means, and what it does not, is the first step.