I think this is feels a pretty useful gap. Seems very useful for low-key, grassroots gravel races, etc. I'd use it if I were still in the RD business. But I'm burned out of that. :)

I know that at least early versions of BWR kind of hacked Strava to get ordinal lists of finishers by time. I don't know exactly how they did it, but whatever it was it was probably frustratingly labor-intensive. It's one of those head-scratching things that I wonder why Strava never bothered trying to support (along with casual ride organization, which people mostly still use Facebook Events/Messenger for).

I actually think you could find a very eager user base by actual professional race directors. As a former RD, there is *always* a need for independent backup data. When your RFID system and/or cameras goes down for 10 minutes because someone kicked the power cable to the generator, and you have to reconstruct finisher order for that 10 minutes while people are screaming at you as if it's the Olympics.

I could even see it being useful as an integrity insurance at major events. E.g. having a RasFre kiosk after the finish line, and if you upload your GPX file within, say, 20 minutes of crossing the finish line, then protect yourself against issues with accidentally missed timing pads or something.

I didn't try out an actual pretend race, but otherwise it looks like a nice start.

One thing I didn't really mention is the live tracking feature. If you use it, anyone can see race progress as it happens. So for an RD, just by telling people about the site, they get free "live athlete tracking" for anyone who chooses to use it.

Overall, I think you are right. I hoped my target audience would be end users; like you say, grassroots and gravel. But it may well be RD's who use it to augment their results.

That, and for triathlon, they could use the Replay to validate drafting calls. If people are interested in using the site, my plan is to add in drafting detection. Then in the results, I'd highlight people with a flag that spent too much time drafting. The RD could then decide what to do....

(Note drafting detection still has the GPS accuracy limitations. I could not guarantee someone was within the 15m box for more than 15 seconds, but I could flag "this person spent X minutes in what appears to be drafting position".)

---

2015 USAT Long Course National Champion (M50-54)

You can't use GPS for drafting. We often have GPS "offset" in location between people on group rides making it look like someone is 1/8 mile up road, when in reality they're on the back of the paceline. That won't work and shouldn't be used for that. GPS is wildly inaccurate for a call like that.

As for using this for actual racing, I can see it for grassroots. With a big BUT. This can't be some Strava like free for all. You'd have to accept licensing from people wanting to do it or something. Otherwise there may be some liabilities opened up that Strava avoids.

I think it won't work, IMO, because if someone is short on cash to do results/chips/etc.......all they have to do is put in the signup "you must upload your results to Strava on the day of the event" and then yank the data from the Strava "single day" filter on the actual day.

It's basically a work around to not pulling the data the day the event occurs. Also, you can create a segment for the event itself so that you will essentially never wind up with any results not from the "day of" unless a rider goes back and re-rides or re-runs the exact long route later.

Lastly, you'd still have to have volunteers verify actual finish order in groups. Unless it is a tri or TT, position matters most for finish place. Not time. Tris and TT's use time elapse as they often do wave starts and 30-second stagger starts. So you can't use position of finish.

I'd see that matter for something like a gravel grinder.

I dunno. It makes sense, but I'd just wipe the Strava data the day of the event.

I deleted a long wonky reply to Paul on that topic just now (too wonky). But I wouldn't be so quick to discount GPS. (My day job is in the area of precision navigation systems).

A briefer version of my wonk:

1) you can estimate the error. When you know the error, you can know when data is useful and when it's not.

2) Most GPS error is not random. It only appears random when looking at one track. Most is related either to the geometry of the satellite constellations (which appear pretty much the same to all the devices in the vicinity of a race course) or local occlusions/multipath (which appear the same to devices in a local area (e.g. a drafting group).

If you were to hand out small GPS trackers that logged the raw GPS data (pseudoranges and ephemeris), you could, (with the help of data from a GPS base station at a known ground truth position), post-process data to the get the error down to the order to 10's of centimeters (or less) the majority of the time. Basically something that approximates a tactical GPS with a dedicated correction service.

It's much harder with just GPX data from a bazillion different GPS chips and antennas, all mounted in different positons. But I bet you could still get a lot done with post-processing.

And since RasFre here is *already* post-processing data, it'd just be another step.

All the source code is open source. It's wonky, but open source.

"You can't use GPS for drafting. "

You can't use it blindly, and for some tracks (low accuracy), maybe not at all. But my experience, having looked at a lot of tracks, is that they are usually quite accurate.

So it would work for some of the people, some of the time.

And, as I said, I think the utility is being able to flag people for RD inspection. I.E. the RD is watching the result, live, and sees a 10 person pack. OK - redirect your draft marshals to that location.

But really, this is getting off topic - I think the utility is for grass roots, club, group of friends events where (hopefully) none of this is a problem. That is why I didn't even make draft detection a part of my initial code.

---

2015 USAT Long Course National Champion (M50-54)

Cool concept, I'd give it a try. I can envision using it for grassroots mountain bike and gravel races, or adventure-type races. Not so much for road running or cycling.

I like it!

I thought about putting on a "climbing challenge" that included dozens of climbs in the local area. The goal would be to maximize VAM (with a minimum total) in a specified period (say 6 hours). For instance there might be 5k and 10k ft categories. Only specific designated climbs (segments) would count. The riders can choose which segments they want to ride, and in any order they like. The format allows people of different abilities to ride together from one segment to the next if they wish; a feature I think would appeal to many. Families, friends, and clubs could spend most of the day together.

Strava would work for this, but your system seems better.

Oh yea - That is exactly the kind of thing RāsFrē is made for.

The results processing is only checking when people cross the markers specified. So you could create the different enduro segments, and it would not care in what order people did them. It would just score the segments as it found them in the incoming GPX file.

---

2015 USAT Long Course National Champion (M50-54)

Interesting concept. Can see the appeal of GPS based race timing, cut out the hassle of manual timing for low key or ad hoc racing.

Trying to imagine my running club using it- although maybe we could not because not everyone runs with GPS (yet)

Slick feel- accessed via Android phone- nice.

In FAQ I'd like to see cost. Is it free to use, is it subscription. Maybe you have said but I missed it

"although maybe we could not because not everyone runs with GPS (yet)".

Did you notice the Live feature? You can use your phone to stream GPS data to the website directly. The screen does need to stay on for recording (you cannot lock the phone), and that does mean you need a way to carry the phone where you are not accidentally interacting with the screen. I.E. carry the phone in an arm band, not your pocket. Without a dedicated phone app, that is the best I can do. And I won't invest in dedicated apps until it seems like there is interest in the idea in general.


It is free to use for now, and will be for the foreseeable future.

Long term, it needs to be paid for somehow. My preference would be to have no adds, no selling your data (of any type), and have a simple and low cost subscription fee.

---

2015 USAT Long Course National Champion (M50-54)

For any of you who liked the idea I'd certainly appreciate it if you would share a link on your favorite social media. I need grassroots engagement to make this work.

---

2015 USAT Long Course National Champion (M50-54)

I'm giving this a "covid-19 caused my race to be cancelled" bump.

Also, I see the virtual race interest popping up. This IS NOT virtual racing - you race on the same course.

But it does allow TT starts, so you can keep socially distanced....

---

2015 USAT Long Course National Champion (M50-54)