Spoofing Mimi Anderson's Run Across America On Strava

Follow-up thought: as I said, I 100% agree that Strava data can be readily faked. Anyone with a programming background can see that. It's an interesting technical issue.

But showing that data can be faked is a very different thing from asserting that a specific Strava activity *is* faked. I don't believe Mimi's Strava data is faked in the way this line of research implies. It totally fails an Occam's Razor test, in my opinion. Her crew can't even keep the tracker batteries charged, so it's difficult to imagine them creating fancy data forgeries. Even if they had the necessary technical skills, there are much easier ways to cheat. Perhaps there's something else dodgy happening with the transcon attempt, but I think suspicions of entirely faked Strava activities are falling into tinfoil hat territory.

It's Just Data wrote:

Follow-up thought: as I said, I 100% agree that Strava data can be readily faked. Anyone with a programming background can see that. It's an interesting technical issue.

But showing that data can be faked is a very different thing from asserting that a specific Strava activity *is* faked. I don't believe Mimi's Strava data is faked in the way this line of research implies. It totally fails an Occam's Razor test, in my opinion. Her crew can't even keep the tracker batteries charged, so it's difficult to imagine them creating fancy data forgeries. Even if they had the necessary technical skills, there are much easier ways to cheat. Perhaps there's something else dodgy happening with the transcon attempt, but I think suspicions of entirely faked Strava activities are falling into tinfoil hat territory.

Occam's Razor the simplist solution to a problems the best problem 55 year old women with a gammy knee is running across america a fast time this despite the fact she cant walk up stairs her team have lied about the trakers and someone being arrested conclusion mimis team are doing something shady with the amount of weird stops on her data and the previous time this has been theorized that her team is doing a relay or fake data

These spoofs are not to imply that it is exactly what is happening on Mimi's team. The main purpose of this spoofing tool was to show URC and Mimi's supports that they are wrong saying only the Strava data matters and why their lies about the Race Drone live GPS tracker is such a major issue. It boils down to the lack of trust about Mimi and her team from what they have done.

I agree that there could be easier ways to cheat.

I also find the technical and algorithmic aspects of what it takes to make a convincing spoof very interesting.

It's Just Data wrote:

Follow-up thought: as I said, I 100% agree that Strava data can be readily faked. Anyone with a programming background can see that. It's an interesting technical issue.

But showing that data can be faked is a very different thing from asserting that a specific Strava activity *is* faked. I don't believe Mimi's Strava data is faked in the way this line of research implies. It totally fails an Occam's Razor test, in my opinion. Her crew can't even keep the tracker batteries charged, so it's difficult to imagine them creating fancy data forgeries. Even if they had the necessary technical skills, there are much easier ways to cheat. Perhaps there's something else dodgy happening with the transcon attempt, but I think suspicions of entirely faked Strava activities are falling into tinfoil hat territory.

Correct me if I'm wrong, but up until Scam_Watcheroo showed this to us, there seemed to be considerable debate about whether or not Strava data could be faked with cadence. It seems to me you're moving the goal posts.

The idea that four or five people supposedly can't remember to charge batteries on a trip planned as carefully as this is a huge red flag. It's absurd.

Please enlighten me, what are the "easier ways to cheat"? Thanks.

It just looks like you used mimi's data, ran it through a function to add a little noise and present that as original.

If you really want to show it being possible using a function - try present the upcoming few days worth of data before they are run.

It'll be fun 😀

Tango Charlie wrote:

It just looks like you used mimi's data, ran it through a function to add a little noise and present that as original.

If you really want to show it being possible using a function - try present the upcoming few days worth of data before they are run.

It'll be fun 😀

This is tough to do because no one knows where Marina will run in the coming days. They've changed the course repeatedly, along with turning off trackers or letting them run out of charge, so it has made it very difficult for anyone to meet up with them.

Tango Charlie wrote:

It just looks like you used mimi's data, ran it through a function to add a little noise and present that as original.

If you really want to show it being possible using a function - try present the upcoming few days worth of data before they are run.

It'll be fun 😀

That's a sign I did my job if that's how it looks. A convincing spoof is meant to look similar to the original.

The future runs will be spoofed before they occur when I'm done the resampling function.

For those who doubted the data could be faked, Scam_W's effort is important and valuable. I personally never disputed that Strava data could be faked. Yes there was some debate about it, but not from me, so I don't think I've moved any goalposts.

I wholeheartedly agree that the problems with tracker batteries etc are absurd, and indicate a comical lack of organization at best, or attempts at cheating at worst. And granted there are other issues that raise eyebrows too, like an absence of much photo or video evidence, and weird patterns in the Garmin data.

Nevertheless I still think her attempt is most likely legitimate, but marred by poor data collection, and colored by her apparent narcissism and flair for exaggeration that make LRC readers mistrust her. Her poor data collection should probably disqualify her from claiming any record, but I would stop short of claiming she's cheated.

I do think she (or her team) are doing themselves a disservice by apparently refusing to do some simple things that could help. Like explain why she stops every ~75 minutes, or provide a lengthy running video, or arrange for third-party witnesses to join for part of the run.

The old saying goes "where there's smoke, there's fire". But in this case, I honestly think there's just smoke. Maybe I'm wrong, and if better evidence emerges later I'll own up to being wrong. Right now I think we're just looking at a narcissistic runner who handles criticism poorly and whose crew is badly disorganized.

If she *has* cheated, the "easier way" I referred to is some variation on the relay. So the Strava data itself would be real, just not all hers. That seems much more plausible than data tampering.

Results wrote:

This is tough to do because no one knows where Marina will run in the coming days. They've changed the course repeatedly, along with turning off trackers or letting them run out of charge, so it has made it very difficult for anyone to meet up with them.

Exactly, and they'll keep doing this all the way to the finish.

When people join her at the end, she'll only be able to walk, not from being so tired but because that's the best she can do.

So please tell me why you think you have evidence that the arrest was fake ?

I've got an idea...instead of spending your time being a keyboard warrior why don't you attempt the record yourself ?

Have you also considered that the Guinness World Records people will scrutinise the data and decide whether it is a verifiable world record or not...you don't need to...because nobody cares what YOU think

Saying someone lied implies that you actually have some knowledge or proof of dishonesty ? Are you part of the crew ? No, I think not. Therefore all your keyboard posturing is exactly that....posturing. You're not there and you have no idea of whether they can get a signal or not...so please stop assuming you're the expert. And stop the tinfoil hat conspiracy theories.....you have a very low opinion of other people's integrity...don't judge others by your own standards. Oh, yes and any competent coder can create or recreate running data....means nothing, except someone is good at coding and any competent coder can also look at the code behind data and see if it was manipulated or not...so just calm down

Embrarunner wrote:

So please tell me why you think you have evidence that the arrest was fake ?

I've got an idea...instead of spending your time being a keyboard warrior why don't you attempt the record yourself ?

Have you also considered that the Guinness World Records people will scrutinise the data and decide whether it is a verifiable world record or not...you don't need to...because nobody cares what YOU think

The arrest log for that area was checked and there was no arrest of a driver for following an RV. Total BS.

They claim the fantasy follower was from on here, yet he never said a word about this encounter on here.

I assure you he would had it happened.

Read the other thread. It was shown there was enough signal for the tracker to work. It was turned OFF. Even Richard the tracker owner confirmed this.

Guinness have accepted literally hundreds of fake marathons off Parvaneh and Larry Macon for Guinness records. They cannot be trusted.

WE find out the truth, not goons at Guinness

If a tool like this falls into the wrong hands it could be devastating sites like Strava or mapmyrun. No one could be believed.

Please don't open source it and release it to the wild.

you really need to get out of the basement and into the real world

I have created a spoof of Mimi's run from this morning (Oct. 7, 2017) BEFORE she has even uploaded it to Strava, based on the planned route on her website:

https://www.strava.com/activities/1219731895

Next up, I'm going to rest a little and create the spoof for her afternoon run. Got to make sure I don't get injured.

My Strava spoof of Mimi Anderson's Oct 7, 2017 Afternoon run BEFORE it even finishes is up:

https://www.strava.com/activities/1219781156

Doomsday wrote:

If a tool like this falls into the wrong hands it could be devastating sites like Strava or mapmyrun. No one could be believed.

Please don't open source it and release it to the wild.

Eh, I disagree. No one should be believed right now, because Strava and similar systems aren't proof of anything. When a computer system is this insecure, the best thing to do is advertise that fact far and wide. Then everyone will realize that a posted Strava result is no more "believable" than a Facebook post. And maybe we'll see some progress towards more secure run logging systems, if there's demand for it.

By keeping this tool secret, it just plays into the hands of the few people who've already created similar tools, by making their fake results look believable in the public's eyes. With all due respect to Scam_Watcheroo, building a tool like this isn't that hard. Any programmer with some web services experience could probably do it in a couple of days. There's some lingering misunderstanding that it involves a hack or an exploit of a Strava flaw, but it's far simpler than that. There's no security in the first place.

And there's nothing that Strava can really do about it either. It's not their fault; it's just the way things work. Anyone can create a GPX file full of fake position, heart rate, and cadence data, and Strava has no way of knowing if it's from a real run or is total fiction. Just like anybody can submit a falsified tax return, or any other kind of fraud. Maybe they'll get caught if there's some other evidence that contradicts the Strava data. But if they only fudge the numbers a little, and nobody's paying attention, the lie will probably go uncaught.

Damn, you are GOOD, Scam_Watcheroo! I'm totally impressed (cadence, HR, all the goodies)