|
Oct 2019
8:56am, 3 Oct 2019
8,881 posts
|
larkim
The comparison of those two HMs as I'm reading the data Baz says the following:-
82:57 is worth a raw performance score of 5.6, 83:29 is worth a raw score of 5.9, so it's inherently 0.3 "better".
The SSS adjustment for those races were 0.5 (SHM) and 0.6 (AHM), so your raw scores produce net scores of 5.1 (SHM) and 5.3 (AHM), so 0.2 "better" at SHM.
But then the "recentness" adjustment factor for SHM is the full 1.5, whereas for AHM it has decayed to 0.9. So the figure affecting your RBR ranking (from which the average of the 5 is taken) is 3.6 (SHM) compared to 4.4 (AHM)
The opaque bit is how it comes up with the SSS adjustment, but as you can see from the scores above the biggest impact is that "recentness" adjustment factor. It definitely still considers the SHM run to be better though, it will always be at least 0.2 better than AHM as the "recentness" adjustment decays away.
Your current five scores are 3.6, 4.0 (18:38 parkrun), 4.1 (5 miler), 4.4 (AHM) and 4.7 (10k) which actually averages 4.16, so either there's some invisible rounding in there or it rounds down. To improve to a 4.0 if you could replace the 4.7 score with a 4.0 (fast parkrun at the weekend?) which would bring it down enough I think.
|
|
Oct 2019
9:44am, 3 Oct 2019
5,356 posts
|
jda
no that's not how it works. The faster HM gets a better score, but the calculation of the handicap takes account of the runner's "PB" which has improved in the meantime. Which makes it appear like not such a strong performance.
("PB" here may include performance at similar distances, and only counts back a finite time period)
|
|
Oct 2019
9:57am, 3 Oct 2019
8,883 posts
|
larkim
Sort of, but also not sort of. Commutative law etc, it works both ways around.
The SSS score is consistent for every athlete, so if you simply take the "Basic score" and deduct the SSS score you get the Net score.
You can get to the same score in a different way by comparing the runner's "PB" score and the adjustment factors, but the answer is the same.
The runner's PB score and their performance relative to it is really necessary only for the system to calculate the race's SSS score. Once you know the SSS score, everyone's actual outcome will be Performance score - race SSS score - "recentness".
|
|
Oct 2019
9:57am, 3 Oct 2019
8,884 posts
|
larkim
(Though if you can find some flaw in my maths I'd be happy to see it!)
|
|
Oct 2019
10:08am, 3 Oct 2019
8,885 posts
|
larkim
Actually, just read your blog and it seems we're agreed, unless I'm missing something?
"Note that we can also get there by just subtracting the event SSS of 1.4 off from my performance of 5.1, ie it's just my score adjusted for event difficulty.
So, each performance gets a net score. There is then a time penalty applied, which starts at -1.5 for recent performances and after a 2 month period of stability then increases gradually over time. "
|
|
Oct 2019
10:13am, 3 Oct 2019
5,357 posts
|
jda
The point is that bazo's vSSS scores makes it look like the slower HM was better. Which is not the case (unless you define "better" to mean relative to his "PB" at the time. Which in this case I think was actually a 10k).
Bottom line is you can't judge your performance on vSSS. Or indeed on any other numbers on the main result page. As I blogged about.
|
|
Oct 2019
10:21am, 3 Oct 2019
8,886 posts
|
larkim
Agreed - I completely ignore the vSSS figures.
|
|
Oct 2019
10:38am, 3 Oct 2019
15,320 posts
|
Bazoaxe
At the end of the day, both were PBs and I loved them both
|
|
Oct 2019
10:42am, 3 Oct 2019
5,360 posts
|
jda
I agree with larkim that a decent parkrun would knock your handicap down a bit. A couple of quick repeats of that fast one would take you below 4. And your sub-3 marathon will help too 
|
|
Oct 2019
10:48am, 3 Oct 2019
15,321 posts
|
Bazoaxe
when I stick an 18:38 into the predictor it still says 4.1
I think though that this ignores the fact that Edinburgh is slightly long and an 18:38 is actually worth a faster time, so in reality if I run well on Saturday I might get under 4.0
|
