This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

How are "Smart" eat/drink alerts determined?

I have been wanting to get more accurate guidance on when and how much to eat and drink on very long rides.  So, Garmin's "smart" eat/drink alerts seem very promising to me, but I have some questions.  

What assumptions does Garmin make in triggering these alerts? For example, does Garmin assume that I start the ride fully hydrated? Does it just take into consideration calories burned, or does it look at other factors like power output, temperature, heart rate, etc?  Based on the online manual, it appears to take several of these into consideration, but the manual does not give any details about how or when the alerts are triggered.

I started using the "smart" alerts this past weekend, and it seemed to work ok -- but seemed to recommend drinking a little less than what I would normally do.  Has anyone been using these "smart" alerts for any extended period of time?  How is it going?

Are there any IQ apps that have similar (but more accurate) in-ride fuel/hydration guidance?

  • Well there must have been an update as that information does show up when you drill down to the activity in both the web and mobile UI.

  • Wish there were some way to account for calories in the drink. I usually have about 100 calories in my water bottle so I eat actual food only at every other eat reminder.

    Would be helpful if there were some way to set your parameters around how many calories per hour you want to comsume and how many calories may be in your drink type things...

  • I started using smart drink alert feature since 2020 (with Edge 1030) and I still continue to use it just now (with 1030+), since I think this is very useful (vs a fixed alert).

    I never used smart eat alert instead (now I’m just starting to use CIQ “EatMyRide - Carbohydrate Burn/Intake Balancer” data field).

    I would like to know more about algorithm implemented by Garmin (specifically to be able to plan in advance drinking strategy refill and not only to follow what’s prompted in real time).

    I agree that main parameters should be calories (much better with a power meter than with heart rate monitor only) and temperature.

    Unfortunately, while temperature by 1030 is quite reliable, this is not true with the 1030+ (3-4 °C higher than 1030 and Garmin Tempe), and of course I suppose this can affect smart drink alerts too.

    Doing reverse engineering on my past rides, I tried to obtain a relationship between drink over calories ratio and temperature.

    First of all, at least in my case, the transition/neutral point is not in the range 16…21 °C; maybe this is strongly dependent on rider personal data…and so very different among people.

    The second point is that resulting transition/neutral point seems changing over the time: approx. 23 °C for 2020 rides, 25 °C for 2021 rides, and 28 °C for 2022 rides….

    I don’t think this is due to a sort of adaptive/AI-based algorithm based on historical data, but maybe to changes done by Garmin in the algorithm over the years.

    Furthermore, with equal temperature value, sometimes the drink over calories ratio results in different values (even if in limited range), and this seems to indicate that other parameters are considered in the Garmin algorithm.

    What’s your experience/guidance about what reported above ?