Ionic cardio fitness suddenly dropping

[BillN26.2]

Had a question about the cardio fitness score from the Ionic. The first 3 weeks I had it in gradually increased from 44 to 49. Over the past week it dropped 4 points back to 45. Any idea why this might happen?

I'm 31 year old male, fairly active, I run 9 or 10 miles a day averaging 7:45 per mile pace. It doesn't seem like there should be a lack of data. So does this just mean the 49 was inaccurate and it has gradually settled on a more correct number? Or can it keep dropping? A bit surprised as I'm on the cusp of falling out of the "good" zone.

[Venkats]

The fitness score is perhaps estimated from heart rate boundaries and we know there are some inaccuracies to that effect. That may be skewing your numbers.

For now dont worry about it and hopefully it will get resolved in future updates.

[bbarrera]

@BillN26.2 Fitbit app will initially show a range of scores. I'm assuming you received an actual score (single number) early on in using the Ionic, given you are an active runner.

The 'cardio fitness' VO2max estimation is based on HR vs Pace. A company called FirstBeat in Finland has a patent on some aspects of this, and as far as anyone knows Fitbit did not license the technology. Therefore Fitbit is using a relatively simple technique, where you need a 10+ minute run on flat terrain for it to do a good job at estimation. 

For reference, the 12 minute Cooper's test for 7:45 pace estimates VO2max at 44.4:

44.4 = 35.97 * (12 / 7.75) - 11.29 

If you ran a faster steady pace for 12 minutes, your Cooper's VO2max estimate would go up.

Over on Charge 2 forums there are reports of fluctuating scores, after months of use. I think that is an artifact of Fitbit's estimation technique, and doesn't truly reflect your actual score.

The reason I say that is fluctuating between 44 and 49 is some pretty wide swings for someone that is a regular runner. I've got a bike computer with the FirstBeat technology, ride about 6000+ miles a year for an average of say 5+ hours a week over the course of a year (lately 8-10 hours/week). My score will slowly increase and decrease as my weight and fitness level changes. Here is a 12 month chart:

The peak was in April and May (900 mile month), I had been training since December for my first 200 mile double century in late May. After that event, I took a two month break and gained weight so my score dropped. For me the May score of 43 dropped to 41 just by gaining 8 lbs by end of June, and reduced riding levels in the summer dropped it from 41 to 39 (I'm 55, and 39-43 is good category). Just losing the final 20 lbs to get to 'race weight' will bump my score into excellent category, so keep in mind these are relative scores and can be influenced by weight.

Back to your point, I've increased my riding this month and VO2Max over last 4 weeks is fluctuating between 39 and 40. 

Hope that helps.

[BillN26.2]

@bbarrera Thank you very much for your reply. Very insightful. Seems like the 45 is roughly accurate.

I ran 10 miles today at 7:32 pace and it popped back up to 46. So will continue to monitor as my performance (hopefully) improves.

[Julia_G]

It also factors in your resting heartrate, so if there is a change in that it will affect you cardio fitness score.

[bbarrera]

---

@Julia_G wrote:

It also factors in your resting heartrate, so if there is a change in that it will affect you cardio fitness score.

---

I thought that was only for the range, and not the single "more precise" score:

If your Fitbit device connects to GPS, we can provide a more precise estimate
of your score using the relationship between pace and heart rate during your
runs. This is because individuals with higher VO2 Max have a lower heart rate
while running at the same pace compared to individuals with lower VO2 Max. 
If you're able to run at a comfortable pace for at least 10 minutes, track a
run with GPS. (Charge 2 and Blaze customers should use multisport mode with
connected GPS; Ionic customers should use the Exercise app). We recommend
running on flat terrain as much as possible since only the flat sections of
your run count towards your score estimate. You may need to go on several
runs that are at least 10 minutes in length to affect your score.

Source: http://help.fitbit.com/articles/en_US/Help_article/2096/?q=cardio+score&l=en_US&fs=Search&pn=1#estim...

More detailed background on the technique is found here https://assets.firstbeat.com/firstbeat/uploads/2017/06/white_paper_VO2max_30.6.2017.pdf - Page 3 shows calculation steps, it only involves pace (output) vs heart rate (input). The picture at bottom of page 3 is a good summary if you have some technical background.

In any case, for an experienced runner like @BillN26.2 that regularly runs long distances (you said 9-10 miles a day) I would expect VO2max to remain somewhat constant unless you did some very dedicated VO2Max training (or lost/gained weight). Your max aerobic capacity is genetic, and declines with age or when you go thru a period of detraining. Its possible to make minor gains with dedicated VO2max interval training, however performance gains at longer distances are normally accomplished by raising your lactate threshold (increasing sustained pace as % of VO2max), or improving your running economy.

[Julia_G]

Hi, @bbarrera, I am pretty sure that the RHR is factored in in both instances.

I noted, for instance, that my RHR (as calculated by Fitbit), has risen significantly this week (not sure why).  My previous RHR was 57 and it has gradually increased to 71 yesterday. I generally run once a week about 8 miles and have been doing this for some time.  For a while I used a Blaze, which would give a VO2 max score, but the "connected GPS" worked so inconsistenly I gave up on it, so only have recent data for the month or so that I have been using the Ionic.

My score is still rated "excellent" for a woman my age (63), but I noted that after this morning's run it dropped one point (from 38 to 37).  The only difference I could see was the calculated rise in RHR (which is, of course, nonsense, but that is a different matter...). Otherwise for this morning's run my average heartrate was a bit down, and my speed a bit up, so I am pretty sure that RHR is a factor in the calculations.

[bbarrera]

---

@Julia_G wrote:

Otherwise for this morning's run my average heartrate was a bit down, and my speed a bit up, so I am pretty sure that RHR is a factor in the calculations.

---

Just going by Fitbit algorithm description, and how its done. The issue with Fitbit is that others have reported long runs may decrease the score. We don't know what run data Fitbit uses, or if they use multiple algorithms. In the case of a long run, as cardiac drift sets in later in the run, if Fitbit uses the last 10 minutes then your score will decrease due to cardiac drift. The FirstBeat patent is on selecting only "good" segments and rejecting "bad" segments to perform the analysis. As I'm a cyclist its not possible for me to get a precise score with Fitbit. What I do know from 12+ months of data, is that I can go on long rides, short rides, tough rides, easy rides, and in all cases my score barely moves. It will slowly change over time, as shown in the graph, but even then my cycling performance is more easily correlated to other metrics.

[shipo]

@bbarrera, yeah, I've wondered about the long run vs. shorter/higher speed run thing as well.

In my case I have bobbed back and forth between 49 and 50 since first getting my Ionic in early October; while I am in good shape, I personally think my score is a tad optimistic.  The thing is, most of my runs have either been in the ten to fourteen mile range at a pace between 9:00 and 9:30, however, I have also gone on some shorter runs leading some of the horse owners in our barn out through some very hilly and technical trails; these runs are at more like a fifteen minute pace.

The only reason why I can figure the Fitbit calculation gives me such a high score is the combination of my very low resting heart rate (low 30s at night, low 40s for my Fitbit calculated daily RHR) and the fact I have felt frisky near the end of some of my 10-milers where I've pushed the pace down into the high-seven/low-eight minute range for the last two to three miles.

Regardless of the reason, I'm thinking my score for a fairly heavy (190 lbs. and 5' 8") 60 year old is unreasonably high.

[bbarrera]

Coopers test of 7:03 pace for 12 minutes gives a VO2max estimate of 50. And 7:10 pace is score of 49. Regardless of age or weight. 

[shipo]

---

@bbarrera wrote:

Coopers test of 7:03 pace for 12 minutes gives a VO2max estimate of 50. And 7:10 pace is score of 49. Regardless of age or weight. 

---

Thanks for that.  The last time I ran that quickly was a two mile race a couple of years ago; per the results I ran it in 14:00.9 for a 7:00.45 pace.  I'm thinking I'm probably in slightly better shape now then I was then, but I've been concentrating on longer distances, so it's a crapshoot as to whether I could repeat those results (but I'm sure I can get pretty close).  Said another way, maybe a 50 isn't so far-fetched after all.