I’m wondering how Xertonline does the calculation how many grams of fat and carbohydrates I’ve burned. I’m asking because I do from time to time a lab test including indirect calorimetry. So at least for that lab ride I know exactly the amount I burned and Xertonline’s estimate is way off. It claimed that for a 45 minutes testing I burned 26 g fat. That would be nice but the number is much lower and the CH in turn much higher.
Is it possible to feed back the results of those lab tests into Xertoline? That would make these calculations useful.
I have exactly the same question, because I recently had a spiroergometry. This not only showed how much fat and carbohydrates I burn at what wattage, but also revealed my pathetically low fatmax (approx. 100w), with an FTP (20min test, 5% deducted) of 301w. In other words: I actually only burn carbohydrates. It would be cool if I could give Xert these values for the calculation and the program could develop my fatmax range in the background.
This “data” always amuses me because I am keto/low carb. My ride last night claims I burned 79g carbs, while my total carb intake yesterday was about 3. I probably haven’t eaten 79g of carbs in the last month or 6 weeks combined. Needless to say, those fields got dropped from my Garmin in favor of something actually useful and accurate.
GNG a bit? There can be differences between how much you burn, how much gets converted from fat/lactate to glucose and how much you eat. You can burn a lot of carbs without having to eat them back.
Well put! We all have just to multiple weight by body fat percentage to understand the (theoretical) size of our body’s depot.
Still I’m curious to understand what algorithm/heuristic Xertonline uses to project the number of consumed grams of fat and carbohydrates and how you can control it to come up with meaningful data.
As no one has so far come up with an answer, is that perhaps proprietary information the vendor doesn’t like to disclose? On the other hand, typically all that is based on scientific research. Most companies in the business refer to some papers that provide the scientific background.
It’s all based on pretty standard science. If we assume LTP is ~LT1, then efforts at LTP will start to burn carbs (by definition of lactate accumulation). The rate increases with intensity to the point where there’s enough lactate that fat combustion drops off to near ~10%. This is near TP (again based on existing research). We have a mathematical relationship of this that we derive based on your signature. It’s presumptive but far better than guessing from pure caloric counts. The rate of carb combustion does change (decreases) as your rides get longer and longer and we don’t factor that into things yet. We have a improved fat/carb combustion model that I’d like to implement.
What some of the science doesn’t quite have answers to is the mechanistic basis for how glycogen is spared and how gluconeogenesis (GNG) of lactate and perhaps fat for ketogenic athletes generates glucose that can be used during exercise and to maintain blood glucose levels. As @ntnunk points out, you can rely on very little carb intake yet still generate a lot of lactate, hence burn a lot of carbohydrates during exercise. Your brain needs glucose too to function so the liver will dump glucose into the blood stream either from its own glycogen or converting lactate, fat or worst case, protein to glucose in order to maintain blood glucose levels. The liver can also dump additional converted glucose into the blood in support of exercise. In effect,the liver can produce glucose for anaerobic glycolysis without you having to eat any carbs.
I can only say that the numbers don’t match the readings from the lab. I recorded the lab session and was provided with the raw data so I know exactly how many grams fat and CH I burnt. There is a hugh misaligned to Xertonline’s projections. I’d guess, Xertonline’s model would require more feedback but I’m not sure how I could enter the lab readings into the system.
Peak power: 538 W,
High intensity engery: 14,7 kJ (manually adjusted based on lab result, was 10 kJ by the algorithm)
Threshold power: 185 W
Fat burnt: 26 g
Carbs burnt: 27 g
In reality (measured in the lab) I burnt:
Fat: 2,94 g
CH: 64,26 g
The test was a with
15 min warmup: 60 W
9 min idle (lab setup was prepared), 0 W
5 min warmup: 60 W
3 min: 80 W
3 min: 100 W
3 min: 120 W
3 min: 140 W
3 min: 160 W
3 min: 180 W
3 min: 200 W
3 min: 220 W
38 seconds: 240 W
14 min: 50 W
Test was far too short. 1 hour or so? Were you fasted? You’ll always burn more carbs at the start of a ride and fat increases proportionally as the ride progresses. If 95% of your energy was from carbs for an entire ride, you wouldn’t last very long and you’d bonk far too early.
I’m not sure I understand. Do you want to say Xertonline’s algorithm to predict burnt amount of fat and CH works only for longer rides? Because I did this ride and this is exactly what I burnt (based on the analysis of my breathing gas).
Speculating that I might have burnt more fat if the workout had taken 2 or 3 hours is not really relevant to my question.
Because time is money. Many labs do shorter interval lengths to make the tests shorter and more affordable. A number of coaches recommended at least 5 minutes steps to get more accurate results.
It’s nice that you are all concerned about whether the test was meaningful but then again, it’s not relevant to my question because the readings what amount of fat and CH I burnt are not based on projections but are actual measurements.
How many years have you been cycling?
How much historical data have you loaded into Xert?
What is your current Training Load value or Status Stars count?
How do you hope to utilize the fat/carb estimate?
You’ll find several Xert articles describing the process along with detailed discussions on the forum including feedback from users successfully using the fat/carb data field to monitor their values (or ratio) during rides and events.
Like any algorithm applied to the general population the results are subject to a bell curve.
Couple that with a range of fitness levels, physiology, and a variety of diets and fueling strategies and naturally the same tool doesn’t always work well for everyone.
Some are expecting the estimate to do what is was never intended to do. Others rely on it as a rough gauge. Some focus on the numbers from a weight loss/gain perspective. Some use it for fueling purposes.
Speaking of bell curves, adjusting your HIE to 14.7 has pushed that signature value from the normal range into outlier territory as indicated below. That’s certainly a possibility, but further signature validation may prove otherwise.
I’m sorry to say but your message sounds more like an interrogation than a reply in a forum to a user asking for help. Let’s quickly recap the situation. I’ve got a ride where I know exactly the amount of consumed fat and carbohydrates. That ride has been imported into Xert besides my rides of the last few years.
Xert’s prediction of the amount of burnt fat and CH are significantly different to the real results. Hence I asked what could be wrong? Perhaps Xert requires me to provide more data etc. So far the only thing for me to check was the signature that I’ve adjusted but the small adjustment had no impact. That doesn’t come to me as a surprise as fat burning is not relevant to the watts that exceed the CP and my W’ capacity shouldn’t have an impact.
To my surprise almost all replies I get are people how want to criticize the testing setup of the ride (as if that were relevant) or I’m asked why I actually want to know or get links to most generic articles but of course there are numerous others available but no one ever provided a link (shouldn’t be easy if there were so many, should it)?
So my conclusion is not to expect any meaningful help. Perhaps that’s that sort of community were you need to be known or have a reputation to be taken seriously.
So, please don’t waste your your time any longer, matter closed.
There isn’t an option to change how the fat/calc estimate is calculated but you can manually adjust your signature as you have done.
We are trying to establish whether your signature numbers are valid which in turn determine the fat/carb estimate.
I linked to a similar discussion in my first reply.
Here’s a link to multiple threads in this forum – Search results for ‘fat carb calculation’ - Xert Community Forum (xertonline.com)
Sorry if you feel miffed. We are trying to help and context would be useful to proceed further.
This is my understanding
As someone once said All models are incorrect but some are useful
The amount of CHO and FAT used depends on a number of factors.
Your gross efficiency.it can be between roughly 18 & 25%. Xert, I believe, uses around 21% It has no way of knowing what an individual’s is. If you are better or worse than that you will produce more or less calories.per kJ of power at the the pedals? Your met test will take this into account.
On your step test you will use progressively less CHO as the step progresses. 3 minutes will probably not get you to steady state so your CHO may be higher.
Finally the met test will include your resting metabolic rate (mine is approx 1800 kCals/day or 75/HR which will obviously include CHO. Xert, I believe, does not included that.
Add all this together and factor in that any statutory information on foods can be out by 20% (UK) you can see that things aren’t as precise as we would like or think they are.
