I used MFP to drop a bunch of weight a couple years back, and go back to it when the "off season" is done to get a better handle on it. I agree that the MFP calories burned can be overstated, but I think it's closer when you use a HRM (and i have my garmin connect and MFP accounts linked, so the calories are from garmin).

That said, yes I added the calories burned into the total equation but do not always take the full amount. Depends on the day and the types of workouts. If a long weekend day, it's rare that I get to replace all of them. But typically I'd shoot for a 500 Cal deficit, taking into account the garmin estimate.

Thanks....my calc is coming from my Garmin HR. Encouraging to know that is more accurate than MFP.

I also use myfitnesspal. What has been working for me is setting up my caloric intake based on my lifestyle without training (i.e., sedentary desk job) and with a 500 cal/day caloric deficit on that. Then for any calories burned in training, I'll allow myself to have them all but 500 calories (i burn 900 calories, I add 400 cals to my diet for that day). This seems to work for me for steady weight loss (1-2lb/week) while still having enough energy for workouts. I think it also helps with the tendency for workout calorie expenditure to be over-estimated by apps. I should mention that I still get a lot of calories because I am tall/heavy so with this I still eat 1900 cals/day minimum. Everyone's "numbers" might look a bit different but I think it is a reasonable approach.

I'm not sure I understand exactly what your asking. I'll try and restated what I THINK you mean...in reverse order:

#2. What should you use as an estimate of the calories burned in a session: MFP or Garmin, or a percentage of one or the other?

I haven't used MFP. I used Loseit! when I lost my weight. Actually, I used a combo of Loseit! (for intake tracking) and Google Fit (for calorie burn estimates). Google Fit uses MET based formulas for calorie burn. That actually seemed pretty accurate. I lost 55 lbs, and tracked food-intake/exercise-burn daily, along with actual weight loss vs. deficit-predicted weight loss in a spreadsheet. I also tracked bio-impedance BF% (and implied fat mass). I actually lost weight slightly faster than the calculators predicted that I should.

Garmin uses some HR based formula. I bought my Garmin just as I was reaching my weight goal. But, I found that the Garmin calorie burn estimates to be higher than the GoogleFIT estimates by a generous amount. I don't recall by how much, but I remember it being significant...maybe around 20%.

#1. Should you add your exercise calories back into your daily available INTAKE-budget?

I did not do it that way. But, most of the phone/web app based trackers (MFP, LoseIt!, etc) all WANT you to. It seemed dishonest to me, and felt like it was enabling the bad habits I was trying to break.

In the end, my approach was more of a closed loop control system. I tracked daily body weight, body fat, exercise calories, and intake calories. I compared predicted weight loss to actual weight loss on a weekly basis using the typical 3500 cal = 1lb fat-loss. If I lost less than that the deficit-predicted amount, then I assumed that I was eating more (or exercising less) than estimated. In those cases I made an adjustment to compensate.

One approach would be to do as you suggest and apply a de-rating factor to your exercise calories.

My approach was to cut back on food intake by the deficit error. Eg, I had an estimated deficit over two weeks of 7000 cals = 2 lbs predicted loss. If I only lost 1.5 lbs, then I had an error of 1700 calories over 14 days (120 cals per day). So, I'd adjust my daily intake target downwards by 120 calories.

Iit doesn't take long to nail it, if you truly log everything.

I use my Garmin HRM to monitor calories burned... seemed like before I used the HRM, Garmin was WAY too generous, but with it, it was closer. I also put in the 500 calorie/day deficit goal and tried to stick to that. On heavy workout days, I knew I'd go over the base calorie allocation, but tried to not use all of the additional calories burned leaving 500-1000 calories additional deficit. Over time i lost 1-2.5 lbs per week. Seemed to work well.

I'm sure there is a LOT of YMMV in all of the approaches. That's why I went with the closed-loop approach of using the error in real weight loss to inform my future deficit predictions. After a few cycles through the feedback loop, it all stabilizes out.

For the run I add what my Garmin 920/935 says ~ 80 calories per mile (I'm fairly small)
For the bike I add half of what my Garmin says (I have a PM but I think it's still over-estimated)
For the swim I add 150 calories per 1000 yds/mtrs

No. But maybe add 100 KCals per day, so an increase of 700 KCals total, are you still losing weight? Good, hold that for another week and then add another 100.

---

Washed up footy player turned Triathlete.