Hacker News Reader

Comment by cyberpanther | Hacker News Reader

I created a Strava segment in my neighborhood and started to monitor it. I quickly found a couple cheaters and flagged them. Some I could tell some were just bad GPS data, such as riding a bike in a lake. I usually take mine down if the GPS is totally off. However, some were definitely blatant cheating, such as cycling at 80 mph when the road speed is 40mph.

View on HN

I gave up on Strava long ago and now only use it to "benchmark" myself. How well did i do on this ride vs the last time i took this trip?

The cheating on Strava is just shameful. you often see people with impressive KOM scores, but then check their ride history and it is pathetic and clear they cheated. Someone with 3 rides and an average speed of just 8km/h posting a 55KM climb up a 6 Degree incline? Sure...

As the other guy posted, strava's DEV's need to do more to stop this. Cross-reference the KOM with the rider's history.. or even just exclude things that are simply not possible??

I've had my GPS mess up, and sometimes post that i was traveling hundreds of KM/h and flagged those trips myself.. I would think that an average speed of 833 km/h should be a red flag for strava as well, but it wasnt?

Lastly, i fail to understand the whole point of cheating in the first place. Great, you got KOM on something, and everyone saw this, knows you cheated and so?

I gave up on it, but for a different reason. Strava is basically "done" in that all the roads have been ridden tens of thousands (and even hundreds of thousands) of times. Take away the cheating and the ranking on climbs just becomes ossified over time. It basically started out as a fun activity and turned into a public performance that nobody else cared about.

There are plenty of other apps out there you can use to track your personal progress that don't have the "social" bit attached.

I've been enjoying this one– https://wandrer.earth/

The game mechanic is novel and I'm surprised how many good areas I've been purely because I'm looking for trails I've never been before.

Strava's social aspect is easy to control. Dont follow anyone, and you can even boot others who follow you. You can also mark yourself "follow by request" which seems to facilitate cheating as well (you can see their KOM, but not their rides?).

This speaks to the changes strava needs to make. if the rider is "by invite only" all KOM's should be hidden as well.

Strava does a decent job tracking myself, and once i turned off all the social. media aspects and stopped looking at other riders it became much better.

Dealing with fakers is one of the key problems for any service like Strava. I have a set of ten segments that I do around town, and I constantly see bike times posted as running. I think most are accidents (e.g. forgot to change the activity type on their watch) but I've noticed some folks who seem more deliberate about it. BTW it's really not hard to tell. I've even seen a couple of times that were people in their cars, but those were even more obviously accidents.

To their credit, Strava tends to be really good about taking down times once they're flagged. OTOH, one of my own runs got flagged once, and the weird thing is that it wasn't even a particularly good time or anomalous in any way. If somebody wants to erase one of my below-average runs from the record, I think I'm good with that. :D

The thing is that Strava doesn't even try to automatically flag activities that are obviously fake or cheating. Their developers just seem completely lazy or incompetent. Like it should be trivial to automatically flag any activity that is significantly faster than the world record for that distance.

> developers just seem completely lazy or incompetent

why? you really don't need to go down that road. Are you a developer yourself? because I'm sure you don't enjoy someone saying you're lazy because you haven't implemented a feature that some rando on the internet thinks you should. You must not be a developer because you can't possible think that this decision lands squarely on the shoulders of the dev team.

I've had my GPS go haywire and tell me my average speed was 833km/h. Do you think it is unreasonable to assume strava should have flagged this?

Strava needs to take more ownership of the rampant cheating. It is costing them users, and causing the overall quality of the product to go downhill. I use to be a big strava fan, now the only thing i use it for is to benchmark myself as anything else isnt worthwhile given the rampant cheating that takes place.

There are a few cases that are trivial, but there are a lot more that can be very difficult. I'm sure the devs err on the side of avoiding false positives, because that could really enrage a user.

The problem is that the Strava developers have been too lazy to even catch the most obvious cases where there is zero risk of false positives. Like if someone "runs" a mile in 3:25 on a flat segment, is that really possible? It would be absolutely trivial to automatically flag those cases, but they haven't bothered to even try.

again you call the devs lazy. you really need to sit back and think about what you say. There is nothing more apparent to me that you have no idea what you're talking about when you say that. Obviously you've never worked on a large team or in development at all. Strava probably doesn't think its worth it, and has absolutely nothing to do with the development team.

I have done development work on large software product teams. I know what I'm saying and stand by my comment.

so you're just a jerk. nice.

I'm always loading my bike/skis/etc into my car after a ride and forgetting to turn off Strava for the drive home.

My commute goes past a velodrome that sits next to a short, steep hill. Every day there were new times on that hill that would be difficult for some cars to achieve. It was pretty obvious that many people forget to turn it off, and never bother to trim the data.

Reply on news.ycombinator.com