So going through this journal article it looks like FTP is equivalent to 4mmol/L, but not to the 4 standard methods for determining lactate threshold used in the article. These yield values of 89%, 92%, 83%, and 89% of FTP respectively. Three of those values just happen to be right at the peak of Coggan’s Sweet Spot model which suggests it as being one of the most efficient places to train to raise threshold power. So if we assume that there is a fairly important lactate threshold inflection point at around 89-92% of FTP (can we assume this?), is there any reason (outside of race specificity) to train exactly at FTP? Is 95-105% of FTP a grey area in the same way that tempo work is often considered? Does anyone else set their target Threshold training power in this range and only go above it for Vo2 max? Would you apply it to running as well and why/why not? Should I bring popcorn?
Abstract:
Jeffries, O, Simmons, R, Patterson, SD, and Waldron, M. Functional threshold power is not equivalent to lactate parameters in trained cyclists. J Strength Cond Res 35(10): 2790-2794, 2021-Functional threshold power (FTP) is derived from a maximal self-paced 20-minute cycling time trial whereby the average power output is scaled by 95%. However, the physiological basis of the FTP concept is unclear. Therefore, we evaluated the relationship of FTP with a range of laboratory-based blood lactate parameters derived from a submaximal threshold test. Twenty competitive male cyclists completed a maximal 20-minute time trial and an incremental exercise test to establish a range of blood lactate parameters. Functional threshold power (266 ± 42 W) was strongly correlated (r = 0.88, p < 0.001) with the power output associated with a fixed blood lactate concentration 4.0 mmol·L-1 (LT4.0) (268 ± 30 W) and not significantly different (p > 0.05). While mean bias was 2.9 ± 24.6 W, there were large limits of agreement (LOA) between FTP and LT4.0 (-45 to 51 W). All other lactate parameters, lactate threshold (LT) (236 ± 32 W), individual anaerobic threshold (244 ± 33 W), and LT thresholds determined using the Dmax method (221 ± 25 W) and modified Dmax method (238 ± 32 W) were significantly different from FTP (p < 0.05). While FTP strongly correlated with LT4.0, the large LOA refutes any equivalence as a measure with physiological basis. Therefore, we would encourage athletes and coaches to use alternative field-based methods to predict cycling performance.
Edit: so as a lot of people have rightly pointed out, it does not appear that the 5’ all out effort was likely performed before the 20’ max effort, which makes the FTP numbers potentially invalid. Since it has been probably a decade since I had done an FTP test, I forgot completely about the 5’ effort beforehand (I usually estimate my FTP from either a sprint triathlon or just looking at my training data)
Speaking from my own experience, LT2 (true FTP) is not necessarily 4.0mmol, it is highly individual. In fact for me, it’s not even close on the run. I’m lower on the run and higher on the bike than 4.0. All depends on how you derive “FTPâ€. I take a realistic approach to FTP for a power I can actually hold for about an hour and this number lined up perfectly with LT2 in my lactate test (literally spot on). I do my “FTP†intervals now within that tight range. Based on how they’re defining FTP and using 95% of a 20 min test, I’d say that study is as good as useless.
How did you determine LT2? Why one hour and not 2 hours?
Lactate step test; 6mins per interval. My LT2 lined up to exactly what I estimated (and WKO5) I could do for ~1 hour. Essentially made me think WKO5’s mFTP is good enough rather than spending $ on lactate strips.
Lactate step test; 6mins per interval. My LT2 lined up to exactly what I estimated (and WKO5) I could do for ~1 hour
How did you choose that inflection point? Did you follow a specific protocol or a set lactate accumulation target? 1 hour seems low for what I would expect LT2 to be.
I used exphyslab to input the data and my LTP2 and modifiedDMax values were within a watt or two of each other. The test data point right before where it calculated these values is a pretty clear inflection point, it isn’t rocket science.
I find my LT2 to be 4-6% lower with 8 minutes steps compared to 5/6 minutes steps, same onlinecalculator as you.
I also find that the LT2 I find with 8 minutes steps, is pretty steady state for let’s say 6x10-12 minutes with 2 minutes rest at same lactate value (2.7 mmol for me). Much more power and I quickly see numbers above 3 mmol and going up after each interval (also when on top of hydration)
I used exphyslab to input the data and my LTP2 and modifiedDMax values were within a watt or two of each other. The test data point right before where it calculated these values is a pretty clear inflection point, it isn’t rocket science.
In the article above, they observe that Dmax values were 89% of tested FTP/4mmol/L LA. This would seem to me to be a power that could be held for 2-3 hours, so I’m a little surprised that yours gave you a value for only 1 hour. What lactate accumulation was that at?
Edit: the wattkg.com power data above does seem to suggest that the Dmax values they found in the study could be closer to a 1 hour power than 2 hour power if they did not do the 5’ all out effort before the 20’ effort as specified in Allen’s protocol.
Thanks for this chart. It looks like 1 hour power is 87% of 20’ power in their 75th percentile projection, which would in turn suggest 92% of a 95% 20’ effort (if the 5’ all out interval was not done before the 20’ effort, which is not clear in the study)
Makes sense, the longer the interval the better for sure. Will have to try with the longer steps in future.
I tend to have the same heart rate for all of my lactate testing at LT2 on the bike and so in full disclosure I have a power target of course when doing intervals but also pay close attention to my heart rate and keep it in the right range.
As an aside, I find knowing your power to HR correlations important as a great first indicator of systemic fatigue as if I notice my heart rate is fairly low at a certain power (whether threshold or otherwise), I know I am overreached.
I know they mention both DMax and ModifiedDmax in that study, although to clarify I use ModifiedDMax as I find it correlates better based on other studies I’ve seen. The test data point I’m speaking of just before where it calculated my LT2 is 4.2mmol and it calculates my LT2 using ModDmax as 4.9mmol.
Baseline lactate up to what I count as LT1 was 0.8mmol (right before first rise in lactate) and my max lactate after the last step was 12.9mmol.
Rightly or wrongly, that’s how I use the values.
Looking at that chart, couldn’t we infer that 95% PAvg of a 20’ effort to estimate FTP may be a high estimate? In the P10 group we are looking at about 86% of the 60’ effort.
I know they mention both DMax and ModifiedDmax in that study, although to clarify I use ModifiedDMax as I find it correlates better based on other studies I’ve seen. The test data point I’m speaking of just before where it calculated my LT2 is 4.2mmol and it calculates my LT2 using ModDmax as 4.9mmol.
Baseline lactate up to what I count as LT1 was 0.8mmol (right before first rise in lactate) and my max lactate after the last step was 12.9mmol.
Rightly or wrongly, that’s how I use the values.
Thanks for the info. Do you typically do threshold intervals at that Modified Dmax, over it, or at a percentage of it?
Where I’m trying to go with this discussion is to talk about is there an optimal threshold training spot, where that lies in the lactate accumulation curve, and if there is reason to train above/below it. And I’m also curious where exactly these lactate threshold inflection points fall in Coggan’s famous “sweet spot” graph (assuming that data is still relevant?). It seemed to me that they lie right at the peak of it, but if they are much closer to the FTP values derived in his testing, then would that mean we should be setting target threshold training power even lower?
The most I’d ever consider using with an athlete is 88-90% - that’s adjusted based on their cycling history.
If I can, I’ll validate with lactate to get a few data points.
Pretty interesting.
I will add my N=1 datum. I use Trainer Road’s ramp test to estimate my FTP, and last year I decided to do lab testing (lactate and ox mask) to get a better idea of where I was… turns out the result from TR was pretty spot on vs what I got from the lab testing.
I don’t own a lactate meter, so I continue to use the ramp test to estimate FTP and training zones.