Another run across America underway - is it legit? Boots currently on the ground

You are rude. You also repeatedly can't answer simple questions. I haven't got a PhD. Never claimed to. You're just being called out by soany people you can't tell your tail from your ear.

It says: Sneakers has hijacked a thread to try and emphasise his importance to the world and has repeatedly failed to do so. I have harnessed the whole power of AI and still don't know what else is going on. I think there is some suspicion about Transcon veracity. I'm certain about the first bit., though. When we get control of SKYNET I promise we'll send the T units to his zip code first.

Classic Sneakers. Still at it. To be clear, i'm on the fence on this one. Goodge cheated, we know it, he knows it, Sneakers knows it.

It was clear from early in this thread Sneakers was either on team Goodge, or, closely connected. Since this thread fired up again, Sneakers is back at it. By 'proving' everyone else is wrong on this one, they were clearly wrong with Goodge as well.

2 birds, one stone.

Again, I'm on the fence, but the most alarming for me is not releasing data. Why? If it was me, I'd have a live stream, great Starlink sponsorship opportunity, hide nothing, the only time I wouldn't be visible would be on the toilet because nobody needs to see that!

Yet you were. Despite the fact that from the outset I said I wasn't suspicious.

For the rest of you, I am cooking dinner while doing this, here a quick back of the envelope scatter graph from my mac version of Excel:

ibb.co

pj-run

Image pj-run hosted in ImgBB

Link:ibb.co/JkQdhwf

To re-iterate, I made a scatter graph. This is not analysis.

Interestingly I noticed lots of missing cadence points, I am not sure if this is typical on runs. I grouped the cadence points together and created a table to show the count of each grouped cadence point and their association with speed. See below:

ibb.co

Screenshot-2024-03-21-at-19-14-19

Image Screenshot-2024-03-21-at-19-14-19 hosted in ImgBB

Link:ibb.co/WggVRg7

Now I am here to have a legitimate discussion. Please no insults. Everything in this post is in good faith.

And I should explain, for the "average of 100 sector speed", it's a relative measure. For each data point it is the mean of the 50 prior + 50 following data points "cumulative distance", so basically the average change in distance over the 100 surrounding data points. Like I said I am making dinner so this is just to make it an easy comparison between data points.

You can see for example there are 12,119 data points with a cadence of 0-9 (these are basically all data points with a missing cadence value) and the speed is 44 on average.

Another big grouping is the 180-190 cadence bucket with 18,838 data points and an average speed of 26.

Like I said this is back of the envelope (the speed is just a relative measure) but I've worked a long day and I'm making dinner and just wanted to get this out there before I'm insulted again.

Glad to have you here, thank you. Sneakers will try to show you know nothing, don't be put off by it. Everyone here wants genuine confirmation or proof of cheating.

The fact remains, as with Goodge, things could be so much easier if people didn't hide or manipulate data, and, covered every possible scenario of 'tech fail' to ensure they could be trusted.

I'm not trying to start a fight with you. You've already shown a lot more good faith and actually downloaded the data and carried through with your word to look at the data in detail. Rob Pope claimed to be a published PhD and has been accusing PJ of cheating these past few weeks everywhere and when I called him out, Rob's response was that he literally couldn't be arsed to spend any time looking at the data despite him demanding all the Garmin Connect links.

As for the cadence dropouts, that's a widely known issue caused by PJ's use of trekking poles because it interferes with the accelerometers detection of the arm swing. It is much easier to see if you calculate the stride length from the speed and cadence values and plot that into a histogram like I've done in past posts in this thread. The cadence dropouts will then show as infinity or enormously large stride length valued that you can bin.

Rob asked for the Garmin data to see if he could prove either way if this was legit or not. He asked for it to be sent privately if they wanted.

The facts remain Sneakers - why not release everything? Why hide data? Why not shut up all these idiots online once and for all.

I'll repeat again, before you go into attack mode, I'm on the fence. I can't fathom why someone doing anything like this, doesn't want full transparency in every possible way.

ibb.co

Screenshot-2024-03-21-at-19-36-28

Image Screenshot-2024-03-21-at-19-36-28 hosted in ImgBB

Link:ibb.co/HNJbzKF

Apologies to delete my last image. Small mistake and I think this view is much more digestable.

Will just point out that RP didn't claim PJ was cheating and provided pretty decent evidence of his PhD and more importantly, will point out that cadence zeroing is NOT an issue with piles - cadence reductions are. Checkmate Sneakers. You've been outwitted and now after having already been rude you're trying your usual trick of kissing up. Youre already ranked second at best of the stats crew - waiting for you to get pushed off the podium. It's a shame, really. After ALL this work...

You literally insulted me immediately for no reason. I don't see how your interactions with others are relevant to your interaction with me.

Like I said, for fear of more insults from you, I rushed out the first image but corrected it with a much better view.

Honestly I have barely looked at the table but my initial take was that IF he was riding in the van (assuming that is what we are attributing the zero cadence to) then they are going significantly slower than his running pace when he has around ~180 cadence.

Simple reason. Because Rob Pope and the other old guard accusers of PJ are not asking for the Garmin data in good faith at all as I've demonstrated with the trap I set for Rob in this thread. Just read what Rob wrote before he left in a hissy fit saying he can't be arsed to download any data, let alone plot it, or heaven forbid, analyze the data.

I will stop triple posting I promise. What I want to know is:

1. Does this view make sense to everyone? Is it helpful?

2. Do we need to bring in any other data elements?

3. Do we want to repeat this view for any other dates?

This is the data view:

ibb.co

Screenshot-2024-03-21-at-19-36-28

Image Screenshot-2024-03-21-at-19-36-28 hosted in ImgBB

Link:ibb.co/HNJbzKF

Like I've said several times, this is literally not data analysis and I'd expect an intern to be able to do this on their first day. But that's not really the point, I just want to get to the bottom of this.

Also, could someone clarify what data is available in the original Garmin files that is not in the Strava data? Or is it more that the Strava data has been through some form of manipulation by Strava itself?

That is the "micro riding" that PJ's accusers have now put forth as their main accusation of how PJ is cheating after I showed there are no enormous speed increases they previously accused him of.

The zero cadence valued can also be caused by the trekking poles.

From what I understand, the Strava data is smoothed and simplified. For example, the mobile version shows lowest cadence at ~120 and the desktop shows lowest at ~0. However, I think people are saying even the desktop data has been through smoothing/simplification and that's why releasing the Garmin source data without these Strava modifications is so important. That is why it is possibly suspicious Paul refuses to do so.

I think people have been trying to figure out/arguing about whether the cadence dropouts on Strava are statistically significant and could result from van riding every now and then, or whether they could be accounted for by usual gps glitches etc. Many say they couldn't be.

Whether the Strava data is suspicious or not I don't know but if I was innocent I would release the Garmin data immediately and have 24/7 filming so there's no room for doubt.

Or, they are asking because there are many red flags that need addressing.

You failed to answer the questions. Why not release everything? Why not shut up the trolls online and prove them wrong? It's very easy to do. The biggest FU and would surely see a lot more donations once proven.

When I first started lurking this thread, Will was doing most of the posting and keeping the tone neutral. Sneakers seems to have derailed the thread by posting nonstop and constantly putting down and arguing with others, to turn the thread about himself. This thread is about PJ, not the mysterious sneakers character who is posting from behind seven proxies so that the Brojos don’t release his IP address.

By "neutral", you mean constantly accusing PJ of cheating and grossly distorting the data to make even more accusations of PJ cheating?

Seemed like RP proves good faith for him and bad faith for you.