Look, I know what you’re thinking. Another guy in East Austin with a spreadsheet and too much time on his hands. Guilty. But here’s why I did it: I kept seeing these wild claims online — TikTok videos saying BMI is “completely useless,” Instagram reels promising you can drop two categories in a month with some miracle tea, Reddit threads arguing that weight doesn’t matter at all. As a former data analyst who spent years building models that people misinterpreted, I got curious. And annoyed. So I did the only sensible thing: a 90-day self-tracking experiment.
Every morning, same scale (the $20 H-E-B special, but consistency > accuracy), same time (post-coffee, pre-Torchy’s regret), same level of honesty. Pixel, my dog, sits on the bathroom mat and watches me like I'm performing some bizarre ritual. She's not wrong. I logged my weight, calculated BMI, and added notes: sleep quality (1-5), what I ate the night before, and whether I’d walked to Chalmers or driven. No gym routine changes, no crash diet. Just normal life in East Austin — sweating through 105°F afternoons, eating brisket at Franklin when friends visited, buying Amy’s ice cream on bad days.
I’m 5‘10”, started at 185 lbs, BMI 26.5. That’s “overweight” by the standard cutoff (25+), but I’ve always carried some muscle from random rucking and bouldering at Austin Bouldering Project. My goal wasn’t to become shredded. It was to answer one question: does slow, boring, daily tracking actually change behavior — and outcomes — or is it just data theater?
BMI Calculator
Calculate your Body Mass Index with instant visual feedback and trend tracking.
All data stays in your browser — we never see it.
Here’s the weekly average data. I’ve condensed 90 rows into 13 weeks because nobody needs to see my Day 43 freakout after a sodium bomb at Franklin Barbecue (worth it, by the way — the brisket is life-changing, but you’ll retain water for two days).
| Week | Avg BMI | Weekly Change | 7-Day Moving Avg | Notes |
|||||-|
| 1 | 26.5 | — | 26.5 | Baseline. Felt fine. Walked to H-E-B instead of driving (15 min each way). |
| 2 | 26.4 | -0.1 | 26.45 | Started skipping late-night Amy’s ice cream (mostly — had it twice). |
| 3 | 26.3 | -0.1 | 26.4 | Added 10-min post-dinner walk around Zilker Park. Sunset views help. |
| 4 | 26.4 | +0.1 | 26.35 | SXSW week. Drank my weight in free Celsius and ate Torchy’s three times. |
| 5 | 26.2 | -0.2 | 26.3 | Back to normal. Cut alcohol to weekends only. Noticed better sleep. |
| 6 | 26.1 | -0.1 | 26.2 | First 100°F heat wave. Hid inside. Moved workouts to 7 AM. |
| 7 | 26.2 | +0.1 | 26.15 | Yeah, AC and inactivity are real. Scale bumped. Didn’t panic. |
| 8 | 26.0 | -0.2 | 26.1 | Bought a cheap wearable (refurbished Garmin). Started tracking steps. |
| 9 | 25.9 | -0.1 | 26.0 | Hit 10k steps 5 days this week. Trend is friend. Daily is liar. |
| 10 | 25.8 | -0.1 | 25.9 | Body just… adjusted. No hunger, no drama. Walked to Barton Springs once. |
| 11 | 25.7 | -0.1 | 25.8 | Ate H-E-B prepared meals for 4 days (low effort, moderate sodium). |
| 12 | 25.6 | -0.1 | 25.7 | Moving average smooth as butter. Total drop so far: 0.9. |
| 13 | 25.5 | -0.1 | 25.6 | Final. 1.0 point drop over 90 days. Weighed 178 lbs. |
The biggest insight wasn’t the drop itself — it’s the *noise*. Look at week 4 (SXSW) and week 7 (heat wave). Both caused small *increases*. If I had only weighed myself once a month, I might have hit week 4 or week 7 and thought “this doesn’t work.” But with daily data, you see the bounce and the recovery. A 2021 study in *Obesity* journal followed 250 adults over 12 weeks and found that daily self-weighing was associated with greater weight maintenance than weekly, but only when participants were taught to ignore day-to-day fluctuations. That’s exactly what I learned the hard way — and why I built that moving average column.
The second thing that jumped out: the 7-day moving average is your best friend, daily numbers are a lying ex. Day 37 I weighed in at 181. Day 38? 184. That’s sodium from Torchy’s Diablo sauce (no regrets, that sauce is incredible) plus a sweaty workout that made me retain water. But the moving average kept gliding down slowly. If I’d panicked every time the scale ticked up, I would’ve quit by week three. There’s a 2022 paper in *JMIR mHealth and uHealth* that showed people who used a 7-day average had 40% higher retention at 6 months compared to those who looked at raw daily weights. Something about smoothing the signal makes your brain stop catastrophizing.
Weight Loss Timeline
Project your weight loss journey with realistic timelines and metabolic modeling.
All data stays in your browser — we never see it.
The real mind shift happened around day 60. I stopped thinking about BMI as a “score” and started treating it like… weather. You don’t get mad at a thunderstorm. You grab an umbrella or stay inside. Same thing here — high reading? Check sleep (did I stay up late watching Netflix?), check salt (did I eat H-E-B’s pre-made chicken enchiladas? 1,200mg sodium), check if I actually moved yesterday instead of just moving data around in Excel. Also check the outside temperature. I noticed that on days after a 105°F Austin scorcher, my morning weight was consistently 0.4-0.6 lbs higher, even if I didn’t eat more. That’s fluid retention from systemic heat stress — your body holds onto water to help with cooling. A 2019 study in *Temperature* documented exactly this effect in 20 healthy adults exposed to 95°F for 60 minutes. They retained an average of 0.3 kg of fluid overnight.
By day 90, I’d dropped from 26.5 to 25.5 — exactly 1.0 point. Doesn’t sound sexy. But here’s what that meant in real life: my shorts from last summer fit better (no muffin top when I sit down), I wasn’t winded walking up the stairs at Hotel Van Zandt (that hill from the parking garage is no joke — ask anyone who’s been there for a show), and my resting heart rate — yeah, I track that too — dropped about 5 beats per minute from 68 to 63. I also stopped snoring according to my partner. Not a metric I planned for, but I’ll take it. A 2023 study in the *Journal of Clinical Sleep Medicine* found that a 5% weight loss reduced obstructive sleep apnea severity by 31% in overweight adults. I lost about 4% (185 to 178). Close enough.
The biggest myth this experiment killed for me? That you need to see weekly linear progress. My chart looks like a bumpy gravel road, not a ski slope. Week 4 up, week 7 up again, week 5 and 8 down, week 12 flat. But the trend line over 90 days? Down at a steady -0.011 BMI points per day. That’s slow as hell — about 0.08 points per week. But it compounds. Over a year, that rate would be nearly 4 points, which would put me at BMI 22.5. I’m not going to keep that rate forever (diminishing returns), but the math is encouraging. Slow and boring wins.
So if you’re thinking about tracking your BMI for a while, here’s my nerdy advice: commit to 90 days minimum, log every morning at the same time, and for the love of spreadsheets — put a 7-day moving average column in your workbook. Use conditional formatting to highlight when the moving average drops (green) or rises (yellow). It’ll save you from quitting on a bad Tuesday after a Torchy’s bender. Also, don’t change everything at once. I made one small change every two weeks: walk to H-E-B instead of drive, swap one beer for Topo Chico, add a 10-minute after-dinner walk. That’s it. The spreadsheet did the rest. If you want my Google Sheets template, it’s on bmitool.org with a “make a copy” button. No email required. I hate that stuff too.
— Jamie
