Any way of tuning race predictor?

My max HR is set at 178. That's the peak HR from a 5k TT a few weeks ago. Average HR on a 5k TT is typically 161 with a resting HR of about 45. My age is 55. If it's any use, Stryd CP is 260W. Have been running > 70km/week since beginning of March. Previously was about 50km/week. I do a bit of cycling too. I'm not really training for anything specifically.

All I'd like is for my Garmin to show an achievable prediction that I could go out the door and aim to get somewhere in the region of the time shown, like within 30s.

9256409 said:

All I'd like is for my Garmin to show an achievable prediction that I could go out the door and aim to get somewhere in the region of the time shown, like within 30s.

And all I want is unicorns that **it diamonds for me.

That sounds really impossible to get right on that level for the whole userbase. 

The simplest solution would be to allow the user to enter a correction factor.

The algorithm is dynamic, it's hard to see that you don't need to tweak the correction factor also.. and what predictor it's then if you tweak the values to be something that you like? I don't get it..

On Runalyze, I can mark a session as a race and then use that to calculate a vo2max correction factor which it then applies to all your race time predictions. It seems a pretty simple thing to implement.

In that case, you choose itself "this is my best racing pace", and even then you need to give it correction factor? Doesn't sound impressive.. Just compare how Garmin does it and you maybe understand the difference.

Ok, you probably won't. On Runanalyze you mark "this is my best", and even then add correction factor, to get it match your X result.. then the others are calculated based on known factors for others distances and your happy. Yes, that's called usually predictor but it really doesn't predict, it just uses known factors to convert it. Converting is not predicting. You tell it that this is my pace at X, and then you want to see it for other distances. You can just put your time here and get it converted to distance you would like. https://www.runnersworld.com/uk/training/a761681/rws-race-time-predictor/ 

The Garmin (aka Firstbeat) algorithm then is a totally different thing. It looks your activities.. and does it's magic and we really don't know exactly how it works, but yes, it tries to predict and I would call it predictor and the others as converters. 

Yes, but when I get fitter (or less fit), the factor gets applied to my new fitness and I get a new prediction. Perhaps I do have a vo2max of 50 but don't have the running efficiency or even the tenacity to reach the typical percentage of max HR that other people do. The correction factor would then allow the prediction time to track my actual personal potential. That's how Runalyze works and it seems more useful than Garmin's fantasy (for me) prediction times.

I don't know how exactly RunAnalyze does it? Do you? It might just be VO2max lookup tables with your mentioned correction factor. Not much predictor again.

"RACE PREDICTION

Based on your effective VO₂max and a unique marathon shape calculation, you will always get an up-to-date prediction for every of your favorite distances from 3k to (ultra-)marathon."

Unique marathon shape calculation is your correction factor? And VO2max is just the classic tables? 

I think you would be happy with the VO2max tables, apply your corrections and convert them to distance you like.. Not feeling this as algorithm that tries to predict anything. Just lookup tables and conversions. 

And don't look what Garmin's race predictor gives you. 

Problem solved.

So based on that the problem is not the race predictor but the VO2max difference.