This is great news, and long overdue. Hopefully, not being able to capture past activities “yet” means this is coming at some point.

Much appreciated and still waiting for the past activities Under AI control too (haven’t had much success flagging KOMs with one-minute/km pace segments and you know how annoying that can be 😄).

Great to see that progress is being made, however, from recent experience the algorithms don't work as well as they should.  I just flagged someone who "walked" 8 miles in 24 minutes (top speed was 57 mph) and a few days ago I flagged a "mountain bike ride" that was doing up to 106 mph on a flat road (where the speed limit is 70 mph).  Yesterday I created a new segment on a bike path that runs next to a railway and when the leaderboard was compiled the top 5 places were all above 60 mph ! (I understand that some people like to monitor their commuting time - maybe there should be an activity of "vehicle" so these people don't feel the need to record it as a bike ride.  Clearly it wouldn't be appropriate for "vehicle" activities to have a leaderboard). 

I noticed something slightly different today - someone in my local area typically does daily "bike rides" with improbable speeds (some over 100 mph) causing some distortion of the Segment leaderboards.  None of these Activities seem to have been auto-flagged.  However, today I noticed that one of the individual Segments was marked as "excluded" due to suspicious high speed (it had a speed of 42 mph).  The "ride" still has multiple spurious KOMs which were not excluded so the revised algorithm still needs more work.  The user had a top speed of 52 mph up a 3% slope - such an infeasible speed should have automatically excluded any adjacent Segments (that is, without there being a pause in Activity that might have allowed a transition from vehicle to pedal bike).  

Thanks so much for this feedback, Ian. You're absolutely right that a 52mph speed up a slope should have been excluded. We are continually evaluating and improving the algorithm and I will pass this along to our team.

I'd also like to point out that if the person in question continually uploads activities as regular rides when they are using a motorized vehicle, please send us a support ticket so we can investigate, reach out, and take action if they don't correct their behavior. It is very useful if you also include:

• Links to activities that back up this claim


• Specific details about that activity that point to cheating
  
  
  
  • This can include all or some of the following: links to segments within the activity that are particularly troublesome, screenshots, zoomed in/highlighted portions of the analysis graph, comments on the activity indicating foul play, etc
  
  
  
  


• Links to and names of their Strava profile(s)


• Dates of uploads, if this is a relevant piece of evidence


• A timeline of events, if this is a relevant piece of information

The more evidence you can provide, the faster and better equipped we will be able to handle your request. Thanks once again!
 

✔️We’re doubling down on catching bike rides (or downhill ski runs, car rides, etc.) marked as runs with new run-specific parameters that will flag activities based on their distance and pace data.

Pretty sure this is not working as intended.

Effort thresholds such as sub-30 min. 10k, sub-15 min. 5 k, 4 min. miles and sub 1 min/km downhill runs (obvious rides uploaded runs), should be fairly easy to auto-flag.

Activities below were both recorded and uploaded after this alleged improvement was put into place :

https://www.strava.com/activities/9455702660/overview

https://www.strava.com/activities/9464929005/overview

Above are only two of several activities that triggered multiple "Un oh!" notifications for myself and several other athletes, claiming top spots on several leaderboards.

These two "runs" were very fast but not in the impossible area as they were slower than world records. That can't be catched with algorithms.

What about that long stretch of running downhill at 1 min./Km? Also, anything "near" a world record should prompt the athlete, upon upload, to question their default activity type. Not put the onus on the larger community, which seems apathetic in my area, to flagging obvious abuse.

With sections hitting 1 minute per mile, and 40-second sections under 3 minutes per mile, it’s clear that the auto-flagging algorithm needs work. Extremely obvious cheats, especially when you look at Paul’s previous activity, which is a “Run” that was later corrected to a bike ride. Then you look at Corina’s previous 3-mile runs at ≈11:20/mi, and it’s clear that she used the wrong activity type. She, too, has a “Morning Run” that she later corrected to a bike ride.

But yet here we are, after Strava is "... doubling down on catching bike rides (or downhill ski runs, car rides, etc.) marked as runs with new run-specific parameters that will flag activities based on their distance and pace data."

You don’t even have to consider world records. Beating your personal record by 6 minutes per mile, over a course 3× as long as your previous best performance, should be an automatic flag, or at least prompt the user to confirm the activity type.

Plus, there are things like accelerometer data that can help determine whether an activity is a run. Syncing a run from a fitness tracker that includes very fast GPS data with no cadence information (from wrist-based arm swing or phone bounce) should be suspicious. I understand that phones may require additional permission to share motion data that not every user allows; however, more can be done on Strava’s end to legitimize activities.

It's nice to see this.  There are a lot of leaderboard items that don't even appear feasible.  case in point, the July Cycling Challenge - the leader has over 11,000 km in 25 days.  That's 450 km/day, every single day.  The Tour de France athletes are about 170 km/day an their active days.  There's a lot of work to analyze data, but it should be reasonably straight forward to add some validation by activity type for maximum feasible limits per time period to trap and exclude the outliers.  -  or is that out liars? 😉 😉  

Great news.

Segments are a main motivation for many many riders as you know.

(We) The community helps by marking or tagging, but facilitating cleanup with tools that detect easy-to-identify cases is something that users appreciate.

Now all that remains is to fix the inequality of fighting in the classifications if you are an athlete riding alone VS a riders group (amateur or even a professional race) and I have contributed an idea about that in a separate topic.

Except their tools are not working as intended. Case in point:
https://www.strava.com/segments/18212621

It would be better to disqualify the entire activity even if it is feasible that might have paused it and been legitimate at some point. Then the user can crop the activity and there may be a process to re-qualify it once all driving has been dropped from the GPX file

There's still way a lot of work to be done. Since this was posted, I have seen loads of people in leaderboards with totally unfeasible PRs. The AI should take the whole ride into account.

For example (it is real), a person did a 54km ride in 4h15, meaning that her average speed was 12,8 km/h. Then in a 0,76km 1.9% slope, she managed to get an average of 31,7 km/h. It's obvious that there's an error here. Her whole activity should be discarded for the public leaderboard.

Another thing to get into account is if the person used a heart monitor. Lots of these fake leaderboards entries are done by people without a heart monitor, something that (I think) most serious users do use.

---

@mal wrote:

✔️We’re doubling down on catching bike rides (or downhill ski runs, car rides, etc.) marked as runs with new run-specific parameters that will flag activities based on their distance and pace data.

---

What about car and motorbike rides that are marked as bike rides? I catch tons of these.

An example of an activity from July that was partly done on a car, but the new system didn't catch it:

https://www.strava.com/activities/9408292731/analysis

Check the analysis for the wattage. For 30km or so, Watts were mostly up to 1000W, but then it went down to 100-150. An absurd discrepance like that should automatically flag the activity. 

Soren, please stop patronizing us with copy/paste responses. Everyone here knows how to flag an activity. Telling users to flag activities, after you claim "We are continually evaluating and improving the algorithm and I will pass this along to our team." simply erodes trust in the integrity of leaderboard and your resolve to action user feedback.

Six months after the launch of your new and improved leaderboard that you are "continually evaluating and improving", and my local leaderboards are still cluttered with impossible feats. Like the first two activities on the following segment - recorded and uploaded after changes were deployed: 

https://www.strava.com/segments/22136894?filter=overall

No I won't flag them. Strava's AMAZING algorithm should take care of these glaring abuses of your terms of service (one of them, by a Strava approved PRO, btw.)

But yet here we are...

The lack of response from Strava and the continued presence of impossible activities at the top of leaderboards is very telling. Strava is clearly putting it's head in the sand. But hey, at least you can DM someone now!

Your "algorithm" fails to automatically flag activities such as these:

https://www.strava.com/activities/9766107413
https://www.strava.com/activities/9362748285

There are thousands of activities like these - obviously made in a vehicle. They are easy to spot, and you don't need your users to find them for you. Unfortunately it looks like it's of no interest for you.

The algorithm only works (if it works) for newly uploaded activities. Old activities have to be flagged manually.

It does not. I should have taken a screen cap. Positions 1 and 2 (now gone) were recorded and uploaded *after* alleged improvements were rolled out. They must have been flagged. Manually. By people. Not by 'algorithm'. Both were done by XC skiers, who uploaded their activity as a run. Well above word record 400m pace. At zero cadence, on snow, on a downhill.

Also, please scroll up or hit the 'Load more replies' button at the bottom to see multiple links to world record beating activities recorded and uploaded after summer 2023, when 'new and improved' algorithm was deployed.

@mal  @Soren : could you please update us as to what specifically has been done since June 2023 wrt the following statement:

"We are continually evaluating and improving the algorithm and I will pass this along to our team."

Hello,

Thanks to everyone who recently commented on this topic and provided examples of invalid activities.   We hear you.  Segment and Challenge leaderboards are an important part of Strava and we understand your passion around  leaderboard accuracy.  

We don't have any updates to share right now, but your input and examples have been passed on to the Team.  In the meantime, if you see invalid times on segments, the first step is to use the activity flag tool. If the Athlete doesn't correct their behavior after getting their activities flagged, you have the option to report that Athlete to Strava Support so we can follow up.