[Closed] 20 seconds gap on a 60 second trail

Use the compare function.

One of my favourite local descents has a certain Mr Samuel Hill as the Strava KOM.

He was 20 seconds ahead of me but I've chipped it down to something like seven or nine seconds now by riding it dozens of times on different bikes.

He probably only rode it once... on sight.

Who knows? Well, he does. Or she.

Thing I've found with speed is that some people operate outside your preconceived ideas of what's possible. You think you're pinning it through that rooty section, then you see someone else air the whole lot at 3 times the speed.

GPS errors, straight lining all the bends, dEPO or they are just properly quick.

The other possibility is that there's a hiccup on the GPS data. 10 second lag in positioning under trees coming into the section, catches up while you're in it. That's your 10 second headstart.

Probably happens a lot, but stands out a lot more when the segment time is so short.

Edit: wot he said^, but slower.

I'm really fast. Tell me where it is and I'll come down and benchmark it for you - at least then you'll know if it's humanly possible

One of my favourite local descents has a certain Mr Samuel Hill as the Strava KOM.

I know the descent (which is in a terrible state at the moment by all accounts) and I thought it had been established that it wasn't really Sam Hill?

Historical times are pretty consistent and so I think he is just quick rather than GPS error unfortunately.

Big launch of the gps down the hill.. should get you the KOM
Bubble wrap it up first though

I know the descent (which is in a terrible state at the moment by all accounts) and I thought it had been established that it wasn't really Sam Hill?

It's not the San Marino and it is Sam Hill (I've just been stalking him again now).

You're right though, the descent I'm referring to is a bit blown out and it would have been riding much faster when he set that time. Otherwise I'd have had him by now. Honest.

Thing I've found with speed is that some people operate outside your preconceived ideas of what's possible. You think you're pinning it through that rooty section, then you see someone else air the whole lot at 3 times the speed.

This. a local lad here is elite DH'er, still not quite top flight. I can keep him in sight for a couple of corners at best on an XC ride when he's cruising. Just carries so much speed.

There's a set of steps we do where my best time (all guns blazing) was something like 20 seconds. The KOM is 3 or 4 seconds..there is no way, on god's green earth, it's been done in that time..

only KOM i ever got was when i left strava on in car going home 🙁 buggers took it off me

There's a set of steps we do where my best time (all guns blazing) was something like 20 seconds. The KOM is 3 or 4 seconds..there is no way, on god's green earth, it's been done in that time..

That'll be GPS errors, anything under a minute or so is always going to be a struggle to get accurate leader board times, not every time is inaccurate, but the top places are likely going to be the fastest riders on quick runs, with added errors.

As for knocking 15-20 seconds off your time on a minute of trail, it's possible, after thinking more about technique I can now get down some trails at Swinely as fast on a rigid bike (which was probably a gap of 10 seconds or so to start with) as I did on the FS bike, and at the time I thought I was bordering on the edge of control on the FS, barely feels dangerous on the rigid so there's obviously more in there. And that's with only small changes (exiting berms earlier to help set up for the next one mostly).

stevied - Member

There's a set of steps we do where my best time (all guns blazing) was something like 20 seconds. The KOM is 3 or 4 seconds..there is no way, on god's green earth, it's been done in that time..

There's a lot like this tbh, like the Bitch at Glentress, I've seen world champions go down that at downright traumatic pace and it still took a damn sight longer than the 8 seconds the KOM took. And no wonder, when you look at that segment and see the dude's time started more than halfway down the segment and ended before the end. GPS being GPS I think.

Strava's basically not very good, let's be honest.

I'm really fast. Tell me where it is and I'll come down and benchmark it for you - at least then you'll know if it's humanly possible

I wouldn't necessarily blame Strava - it's more likely to be the accuracy (or lack thereof) of the GPS device used to record the ride,

As the saying goes - "Garbage in, garbage out"

gps error and if you or he has auto pause set for a certain speed it will auto pause giving you a boost when, if you drop below that speed.

I wouldn't necessarily blame Strava - it's more likely to be the accuracy (or lack thereof) of the GPS device used to record the ride,

Agreed.

And just for the record, Strava is AMAZING.

There are two jumps on the last bit of Cwmcarn DH, must be 5m or so apart. Or at least I thought it was two jumps, until I saw someone demonstrate that if you go fast enough it's actually a double 😯

i wouldn't necessarily blame Strava - it's more likely to be the accuracy (or lack thereof) of the GPS device used to record the ride,

An excuse only ever used by slow people...

As others have said there is probably an element of GPS error in it.
However, quick riders really are stupidly quick.
A couple of years ago I was in second place,on Strava, for the descent to Achnashellach in Torridon with a time of 5.28 and I was sure I could make up the time on a good run and get the KOM. The next month Greg Williamson's name appeared at the top of the leaderboard and he had taken a minute out of me! It was the first time he had ridden the descent so I reckon he could probably go a fair bit quicker.
Certainly puts you in your rightful place 🙂

@ Molgrips, if you've not seen the really good guys, they take off, sail way over that double, turn in the air, and land on the berm all lined up for the downhill. Some riders are just awesome.

I've recorded a time at FOD that I cannot get within 10 seconds of. Happy with 7th out of 5000. 😉

An excuse only ever used by slow people...

Cuts both ways though.

I have a KOM at Setmurthy which is suspiciously quicker than the competition, and which was recorded on my mobile.

We really need a special category on Strava for users of dedicated GPS units, eh?

We really need a special category on Strava for users of dedicated GPS units, eh?

So it's less the phone, and more where you put it (IME, although some phones might just have crap GPS)

Ignore strava and go back to being happy.

I've been lucky enough to ride with a few quick pro riders and they are truly astounding to watch. Not just quick, but blisteringly quick.

Oh, and they don't ride the same trail that we ride. Where we're bumping over the rocks, thinking we're doing just great, they take off here

...touch once here...

...and land in this corner down there and shoot out of the corner.

I rode a tricky dry riverbed crossing in Downieville - it was a tricky rocky descent into a flat dry riverbed and then a technical, steppy climb out that I nearly made, but dabbed just at the top. I reckoned I could ride it in one next time.

Then I looked behind to see Peaty and Minnaar (no really, it was the 5010 launch last year) using my 'technical climb' simply as a launchpad to air off as they laughed at each other.
Fast is simply fast.
I am neither, I have realised.

chipps - I work here

Oh, and they don't ride the same trail that we ride. Where we're bumping over the rocks, thinking we're doing just great, they take off here

...touch once here...

...and land in this corner down there and shoot out of the corner.

I remember my first race at fort william, I did a couple of practice runs the day before, got back to the b&b and thought, I'll watch some race footage, see if I can get any hints from the pros on the bits I'm struggling to find a line on. Nope. Nothing they did on those sections had any relevance to me 😆

Even top XC guys have ninja skills. If you've ever been passed by them at a race on a line that wasn't until they nipped passed you on it then it's best to think hard before attempting it yourself later on. And as for thinking "I'll close that gap on the downhill" when they go by you at the top of a hill...

I was asking this very question yesterday. Son of chainslapper (cannock) section is recored at 19 seconds ish by a lot of folk, im sat 600th odd out of 20000 at 29 seconds. I just can't see how you could be 9 seconds quicker and wanted to smash my phone like a petulant child when I saw the result.

Once saw a video dirt did with Steve Jones and Sam Hill, they'd stooped halfway down a trail to do a bit of interviewing and were setting off, both men set off at the same time, pretty much together, and within a couple of hundred feet, Hill was several bike lengths ahead of Jones. Thing is, Hill hadn't put any pedal strokes in, and Jones swears blind he didn't brake.

Marginal gains, everywhere. How many corners does this 60 second section have? If you went through each of them a second faster, how much would you improve by? Could the top man be 10-20% more powerful than you? That combine with inaccuracies in the tech ( sounds like an excuse to me but WE) could easily explain it.

So it's less the phone, and more where you put it (IME, although some phones might just have crap GPS)

^^ This, though as others have pointed out, some riders are just sooooo fast and smooth.

If you zoom right in on the Strava segment with a good GPS signal you'll see many more twists and turns whereas a segment with a crap GPS signal will show more straight lines and definite 'angles' rather than curves. On a very short segment, the GPS trace may just be one straight line

Anecdotaly, I'd say iPhones aren't very accurate for this sort of thing as I've compared my Strava segments (Android) with my riding buddies (iPhones) and mine show much better alignment to the actual trail. I'm guessing it's probably also to do with battery savings (lower sampling rate) rather than inaccurate GPS chips.

I also did an experiment with my phone in the top of my camelbak and a Garmin 520 with both GPS and Glonass enabled but the Garmin was in the camelbak side pocket (I'm not brave enough to put a $400 gadget on the bars of my mtb). The phone was waaaaaay more accurate

Simple one he's faster 😉
Plus a combination of GPS Error and the rest. Even if you are on a 1s count the point at which you enter or leave the segment can be a significant distance apart from another rider or your next go.
On one of my faster segments I'm 40s behind KOM he did it in 1:44 so 20s in a minute isn't that hard to imagine and I know I could carry more speed in a few places. In contrast there is one where I'm top 10 and the gap is 52s(KOM) - 1:04(Me) not sure where the 12s comes from in that one 😉 Thankfully thats getting replaced so the time will last a while 🙂

@wrightyson. Some of the very top times down that section will be when there was a diversion in place and people were hitting the parallel fireroad at full speed. But I'd still say the legit top times will be 23/24 seconds.

Last 12 months has seen every almost every segment in riding distance of me be annihilated.
Had a couple of WC XC pros stopping in the next town over most weeks in the spring and autumn. (Parents of one of them have recently moved into the area.)
And then during the summer a recently ex protour rider was living with a lass in another local town. They've now moved to Italy.

So those almost achievable segments are now so far out of reach it's almost funny. Even our local MTB superhero is a minute down on anything over 5 minutes, though he has improved his absolute times significantly.
And some of the local road bits have had 10+% taken off them.

I'm fat and slow these days so i don't really care that much. I was never going to get into the top 50% anyway.

Try living near Adam Brayton.....

My local trails have been Sam Hilled. Well they are his local trails as well, and he was here first but still. Brutally fast. I have absolutely no idea how.

Quite possibly just a lot faster but it's not to do with how you start the trail is it? There's one near us where we generally stop for a chat before going down but that means you're just starting off pedalling as you begin the segment, auto pause has to wake itself up etc. If you go straight into it from higher up the fireroad you go into the trail a lot faster which results in far quicker strava times. And yes i guess we could stop for a chat somewhere else but not overly fussed 🙂

My wife and her friend have the same GPS, rode the same ride together and yet recorded significantly different meters of climbing.

Probably a 20+km ride, yet they did 10m different in climbing.

I stick to racing my mates - if they are insight in front I'm happy. Its when I can no longer see them in front I get depressed.

Try living near Peaty! Him and his pro mates have KOM'd almost everything.

See them sometimes riding trails, one in front of the other, having a chat and still going far faster than I ever could.

They don't play by the same rules as us lot I'm afraid.

Probably a 20+km ride, yet they did 10m different in climbing.

cloudnine - Member
Big launch of the gps down the hill.. should get you the KOM
Bubble wrap it up first though

Liking this solution.

GPS vs phone often shows up 200m difference over a ride with 700m ascent for me.

Phone always lower and less accurate, referenced against hills that I know the actual elevation of.

belugabob - Member

I wouldn't necessarily blame Strava - it's more likely to be the accuracy (or lack thereof) of the GPS device used to record the ride,

I'd blame Strava for the lack of quality control tbh. I mentioned up the page the KOM on the Bitch at Glentress, it's a good example. If I look at my traces, you get 30-40 GPS plots from top to bottom, mostly pretty nicely in the shape of the trail. If you look at the KOM, you get 5 GPS plots and the first one is about 75% of the way down the trail. It's just not a succesful logging of the segment.

There's a level of reasonableness, Strava can't tell if you cut the corners halfway down which is worth a good few seconds, but it can tell if the GPS plots are too infrequent, or too far from the real track, or in this case nowhere near the actual start of the segment...

So yes, it's the GPS data that causes the problem but it's Strava that accepts such useless info. They seem to have faced the decision of accepting rubbish and getting bad results, or being discerning and accurate which would probably lead to unhappy punters going why didn't my time log, and a lower number of succesful segments which probably makes them look less omnipresent.

It does seem to be better than it was- when I started using it, it'd often log segments you'd never ridden just because you rode past the entrance, it now seems to look for some plots within the segment too.

None of this makes it useless, you just have to take everything with a huge pinch of salt.

Yes, vertical is pretty far out between a Garmin and iPhone. Makes a mess of all the other power/calorie-related stats too.

The KOM and runner-up on one of my regular local segments was ridiculously quick - 3mins up on some seriously fast riders (not me, I hasten to add). Was prepared to accept this until I noticed that one had titled the ride 'Argo to work'. 😀

I can put up with being 3 mins down on the leader, just not six minutes on an eight-minute segment!

If you look at the KOM, you get 5 GPS plots and the first one is about 75% of the way down the trail. It's just not a succesful logging of the segment.

Why not report it then?

I live/ride in the Tweed Valley, ever heard of the EWS?

One of the sections this year was Repeat Offender (top, middle, bottom), Greg Callaghan is sat at KOM on 3mins 57sesc (1 mile and nearly 1000ft of descending).

The fastest of the guys I ride with is at 5mins, and I'm past 7mins..., but still in the top half...

https://www.strava.com/segments/8228187

As said previously, the fast boys/girls aren't riding the same trail as us really.

Why not report it then?

Lol about the ews, https://www.strava.com/activities/497460542
JG53 seems a bit quick...

Professional cyclists in going faster than average punters shocker! 😆

Re. strava, gps etc. Its the same as any measurement, there is an error associated with it, given that strava (for most people) isnt life and death. it doesnt matter. Sort segments just compound the error. I tihnk phone GPS has a positional accuracy of approx 10m so for a 250m segment, someone could have ridden 230m and anoter 270m, thats 16% differnece in distance comapred to the actual segment. Hardly surprising times are off!!

molgrips - Member

Why not report it then?

Because it's not a one off bad result- it's one example on one segment that shows why Strava doesn't really work. I [i]could[/i] go through every segment I've ever ridden and manually report every dodgy time, but I'd not get very far before I went on a killing spree so instead I just go "Meh, strava's pretty broken isn't it" and use it within its limitations.

I just assume anyone going faster than me has a dodgy GPS device, takes EPO, or rides and e-bike !

A few years ago i took Steve Jones (dirt), Stu Thomson (cut media), Joe Barnes and Liam Moynihan (Dudes of Hazzard) round Kirroughtree and elsewhere in the area for a Dirt mag photoshoot, whilst i thought i was pretty decent and [i]fastish[/i] rider (could comfortably lap kirroughtree in under 2hrs on my SS) their riding showed me up for the wannabe mediocore mtb'r that i really was 😥 .

It wasn't so much their fitness that made the difference but more so the speed they could carry through sections and out of turns/berms, on some areas of the trail where there was a climb or a pretty much straight through section i could comfortably keep up with them but as soon as became really tight n' twisty with tricky rock or severe off camber rooty sections sections they just left me for dead despite me having "local knowledge" having lapped Kirroughtree literally hundreds n" hundreds of times (my local trail).

It was a genuine eye opener getting to ride with them for a couple of days and seeing how the genuinely fast folk can gain a second, or two, or three (or tens of seconds) 😥 on a relatively short section of trail.

Yep it's mainly about the corners, I reckon.

I usually ride with a Garmin 800 and my times on Segments are pretty consistent (slow), but I'll occasionally forget the Garmin and use the Strava app on my iphone. Whenever I do that I always seem to get PRs on random segments (especially ones with decent tree cover).

It's all just a bit of fun though.

it's one example on one segment that shows why Strava doesn't really work

It works fine as you say within its limitations.

If it makes you feel any better I'll come and ride it at least 20 seconds slower than you.

I think these are my favourite threads

On such short segments strava is massively inaccurate, the time can be greatly shortened if the first and last gps points are a good way into the segment, which can be pot luck, tree coverage or from your device settings. If you really want to have a good time on it, just finish your ride at the bottom of the segment, and crop your ride. Strava will let you crop it a surprisingly long way up the segment and you can shave a good bit off, and then feel really great about your strava result, which is the most important part of mountain biking.

But a genuinely fast rider will pump 99% of stw mincers who think they're ok, I ride with a fast guy and it's hard to get your head round it. It's good fun to watch your group on strava flyby though with one dot shooting out in front.

Our local Strava hero/ obsessive / saddo (delete as applicable) records on his Garmin and his phone, uploads both and takes the 'best' one for his KOMS, deletes the other

molgrips - Member

It works fine as you say within its limitations.

My riding works within its limitations but it still leaves me embedded in the occasional tree. Everything works fine within its limitations if you accept "doesn't really do what it's supposed to do" as a limitation. But I'm not a project manager so that doesn't really sit right with me.

bigjim - this was not about being a Strava hero, this was about how someone can be so much faster i.e. how can I get faster/better.
I think chakaping has it with corners. This trail has quite a few steep, rooty switchbacks which I possibly may not be tacking quite rightly.
My other bike is a rigid singlespeed and so Strava really is a tool for me rather than a member extension device.

Strava hero/ obsessive / saddo

the fact you know this suggests he's not the only one...

Try living near ...

Phil Pearce (World Cup XC racer) owns our local trails!

My local trails have been Sam Hilled. Well they are his local trails as well, and he was here first but still. Brutally fast. I have absolutely no idea how.

No shame in being beaten by the best.

Is Mike Jones far behind him or is he not on Strava?

If I'm more than 30-40% up on the KOM I'll stop halfway down and learn a dead language or make a couple of hundred metres of coir rope out of decency.

perusing my commute home last night showed how daft it can be, from when I switched on to about 15mins in the route was all over the shop, really badly zigzagging where I had taken a straight line, had me doing segments on the road when i was riding the canal. Later on in the ride it was really accurate, enough to show which line I took in some parts.

Like northwind said, strava seem happy to accept crap data* and while you know that's happening you really shouldn't be taking it seriously**. Look at any segment where someone has whipped you, could be a stellar rider, could be someone with similar abilities in pefect conditions putting in a PB run with a massive tailwind, or it could be a slow numpty with a crap gps unit, you'll probably never know.

*that's on top of head/tail wind, mud, trail creep, traffic,etc all making [b]big[/b] impacts before you look at timing quality
**not that you should be taking unofficial "racing" seriously anyway

I always take Strava leaderboards with a big pinch of salt. I follow a local xc endurance mtb rider who I use as my benchmark for times (read as target) if I can get anywhere close to his times I'd be very happy,

In reality I'm considerably slower but keep getting marginal improvements which is all I'm after and the main reason I use Strava.

Look at any segment where someone has whipped you, could be a stellar rider, could be someone with similar abilities in pefect conditions putting in a PB run with a massive tailwind, or it could be a slow numpty with a crap gps unit, you'll probably never know.

If you care, you should look at their other segments and rides and see if it looks legit.

this was not about being a Strava hero, this was about how someone can be so much faster i.e. how can I get faster/better.

my comments weren't aimed at you, just in general, I'm just always amazed at both how seriously people take strava and how much faith people put in it's accuracy, especially on short segments when the whole thing could be based on a few gps points.

I keep meaning to do a test of the height gain calculation from strava too, I think I can get my hands on some decent data to use so just need to work out the methodology.

I know short segments are more vulnerable to error and Strava sometimes freaks out a bit, but my experience is that if I [i]feel [/i]I've gone quicker on a sub-minute section - I usually have.

In other words, I'm constantly surprised by how accurate it appears to be.

And while I don't study everyone's profile, the gaps between myself and better riders usually seem consistent and credible.

Strava falls down when you've got adjacent fire roads or other trails, excessive shortcuts (not Strava lines, they've always occured) or excessive tree/leaf cover (which screws up the GPS). But it's pretty accurate otherwise.

Regarding faster riders being a lot faster, I think a bit chunk of it comes down to MTBs not having engines. Once you lose speed it takes a lot of time or energy to regain. If you carry speed round a corner or over a jump or drop, you end up with a virtuous circle all the way down the trail. It adds up really fast. As soon as you take a feature slower than is possible it impacts not just that feature but dozens of yards more trail (depending on how steep/fast the trail is). And if you stall, like screwing up something technical or skidding/squaring off a turn, you lose even more time.

The Strava comparison tool really shows this clearly. This was the highpoint of my weekend's riding yet I'm a similar amount per minute behind the KOM: https://www.strava.com/segments/6646217/compare/MTIxMjM1NDEwMTEsMzYzMDMxMjg2Mw==

The biggest loss in time was failing to carry speed around that right hand switchback. And despite being 40 seconds behind the KOM I was over 2 minutes quicker than the median rider (out of 4000+). I've seen what Pat CJ can ride like - he's bloody quick, in a way that I can barely understand let alone attempt to do.

As an enthusiastic and reasonably quick 30something amateur I know I can ride a bike downhill fairly quickly compared to a newbie - but there are tons of amateur DHers, enduro racers and XC racers with amazing skills and fitness who aren't much slower than the pros, and the pros are so much faster than someone of my mediocrity could achieve. I've very much enjoyed seeing how the winning riders demolish the stages on the races I've entered - they see the trail in a completely different way, carrying far more speed everywhere.

Not read the whole thread but in answer to the OP... I am guessing it was DM, if so the fact is he is fast.

I often catch people on segments when I have left huge gaps... get to the bottom and they are buzzing from their run. For balance I have been caught by guys when I have been going flat out for me.

The gaps in ability and fitness can be huge.

I tend to pick a guy who is say 10% quicker than me on Strava and then work at narrowing the gap or catching his times... it is just riding but I enjoy the challenge.