Forums >Racing>Sub 1:30 Half Marathon in 2020
Cant resist a measurement discussion. I havent run a single course that is certified where my GPS showed up short. Not a single one. Maybe you guys run some goofy courses with a ton of L turns and turnaround pints where that could happen, but otherwise, it means the course isn't accurate (unless its just one or two people's watches). Even on relatively straight courses for a person (me) who is really about the SPR, it's hard to measure short. The only time I get right on the money is during these time trials, when I'm by myself and can fully ignore weaving around anybody.
Cant resist a measurement discussion.
I havent run a single course that is certified where my GPS showed up short. Not a single one. Maybe you guys run some goofy courses with a ton of L turns and turnaround pints where that could happen, but otherwise, it means the course isn't accurate (unless its just one or two people's watches). Even on relatively straight courses for a person (me) who is really about the SPR, it's hard to measure short. The only time I get right on the money is during these time trials, when I'm by myself and can fully ignore weaving around anybody.
The closest I've got on a certified course is 21.12km @ Coatesville HM. It has basically three u-turns but it's fairly straight road running with few corners other than that. A little bit of tree cover in places but probably not enough to make any real difference.
Taupo's explanation on course measurement is:
"Our courses are measured as accurately as we can. Our course crew get out on each course every year measuring the distances by driving, running, and/or riding on their bikes. We measure using two separate GPS devices and by vehicle and/or bike odometers where possible."
I'm not sure how anyone could think a car odometer and GPS watch could replace a proper measuring wheel, but anyway.
Piwi - if you look closely, they've clearly changed the Taupo course slightly since the last time I ran it (2018). The u-turn around the 10km mark is gone, and there's now a turn slightly further down that road that takes you on to the lakefront (I assume they extended the lakefront pathway). Just from eyeballing it, it's clearly longer than it used to be.
3,000m: 9:07.7 (Nov-21) | 5,000m: 15:39 (Dec-19) | 10,000m: 32:34 (Mar-20)
10km: 33:15 (Sep-19) | HM: 1:09:41 (May-21)* | FM: 2:41:41 (Oct-20)
* Net downhill course
Last race: Waterfront HM, 7 Apr, 1:15:48
Up next: Runway5, 4 May
"CONSISTENCY IS KING"
Just gone back to Strava to look at my 21.05km in 2016, which is my PR.
Thankfully I've found 9 others on Strava who ran that year. And they're all longer, 21.15 km to 21.29 km.
PRs: 5km 18:43 (Dec 2015), 10km 39:59 (Sep 2020), half 1:26:16 (Sep 2016), full 3:09:28 (Jun 2015)
40+ PRs: 5km 19:31 (Oct 2020), 10km 39:59 (Sep 2020), half 1:29:39 (Jun 2018), full 3:13:55 (Sep 2022)
2023 PRs (hope to beat in 2024): 5km 20:34, 10km 41:37, half 1:32:32, full 3:21:05
2024 PRs: 5km 20:25
What confuses me is that on the Taupo site they give the explanation for getting a short reading which is the gps drawing a straight line over corners. Why does gps then measure long on a track ? 5000m for me always comes up 100m short of where it should be.
We definitely put too much trust in our gps to be 100 %
55+ PBs 5k 18:36 June 3rd TT
" If you don't use it you lose it, but if you use it, it wears out.
Somewhere in between is about right "
What confuses me is that on the Taupo site they give the explanation for getting a short reading which is the gps drawing a straight line over corners. Why does gps then measure long on a track ? 5000m for me always comes up 100m short of where it should be. We definitely put too much trust in our gps to be 100 %
Yeah and it seems like an odd argument to run when the course seems to have come up around 21.3km for most runners I could see? (i.e. if anything the course is long, not short!). In my experience GPS almost always over-measures where corners are involved.
I checked and it was 21.03km for me both times I ran it. If I had to guess, I'd say it was slightly short under the old course, and maybe the course change this year slightly over-corrected for that.
RIP Milkman
DRIVING A CAR??? Wow. I can't believe they actually use that to justify how well their course is measured! Better to use a damn GPS watch than a car!
5K: 16:37 (11/20) | 10K: 34:49 (10/19) | HM: 1:14:57 (5/22) | FM: 2:36:31 (12/19)
I forgot - good call by Steve to skip the race.
Measuring a race with a car and a couple watches? What a joke.
Steve, prudent but unfortunate call.
DW - you're coming back really fast. Congrats
Rest of you solid weeks.
My week was rather pedestrian
M - 8 miles easy with 5x10sec hill sprints in the middle. 12 mi bike PM
T - 4x[300fast, 100 walk, 100 sprint, 800 jg]. RIP quads. Somehow hit a :50 first 300 which I didn't know I was capable of. Rest of them were 55-56, all the 100s were :15/16.
W - 7.5 easy on trails at the hottest part of the day 🥵
Th - 8 w 8minutes on/ off. Still sore from Tuesday, on paces were 6:20-40 which is a little disappointing but at least I stayed steady and didnt blow up.
F - off
Sat - 20 miles ultra style. Walk the uphills, > 100feet gain per mile. Slowest 20 I've ever done but it was fun times w friends.
Sun - 8.4 easy on one of my favorite Trail loops. Surprisingly strong run given the time on feet yesterday.
60 miles
1600 - 5:23 (2018), 5k - 19:33 (2018), 10k - 41:20 (2021), half - 1:38:57 (2018), Marathon - 3:37:17 (2018)
Resident Historian
Unless they've got some GPS that samples every 10 seconds or so, that explanation is bogus. GPS will almost always show a longer course unless there are places where it loses signal. For a marathon, I know people who set their Auto-Lap to 1.01 miles because they know they'll run 26.22 + at least 0.15. GPS recordings of 26.4x are not unusual. I prefer to let the GPS record, but have a watch on the other arm to hit the button at each mile (or Km.) marker.
1. The track is measured at the inside of Lane 1. Nobody can run exactly that line. Middle of lane 1 (~1m wide) gives you an extra 3+ m./ lap.
2. For the track again, which arm is your GPS on? If it's the right arm and you're going the normal counter-clockwise direction, your turns have a little more than 1/2 wider radius. That's 3+ meters/lap.(BTW, I've found that the GPS laps on a track come in closer to actual measurement if you set Data Recording to every 1 second)3. Any course: GPS accuracy is generally 3-4m, which means one data point can be on the actual course run, the next point to the left, the third the right. On a straight course, that may almost cancel out, as the errors tend to drift to one side of the course or other, then back, but there will some added distance because the GPS thinks you ran a slightly zig-zag course. On turns, the tendency of accuracy errors is to add distance, more like overshooting a curve than "drawing a straight line over corners".Oh, and I agree with Jmac "DRIVING A CAR???"
Neil
--------------------------------------------------------------------------------------------------“Some people will tell you that slow is good – but I'm here to tell you that fast is better. I've always believed this, in spite of the trouble it's caused me. - Hunter S. Thompson
Incidentally, for those who remembered the infamous Strava battle for Bomb Point about 2 months ago... Johan never showed up to Parkrun on Saturday. No real surprise there I guess.
Of course, what is the point to race with old slow dudes, who cannot even "run" on a bicycle?
paces PRs - 5K - 5:48 / 10K - 6:05 / HM - 6:14 / FM - 6:26 per mile
Hot Weather Complainer
5km: 18:34 11/23 │ 10km: 39:10 8/23 │ HM: 1:26:48 9/23 │ M: 3:34:49 6/23
2024 Races:
Motorway Half Marathon February 25, 2024 1:29:55
Christchurch Half-Marathon April 21, 2024 1:27:34
Selwyn Marathon June 2, 2024
Dunedin Half Marathon September 15, 2024
Zebano that is pretty much how most races are measured over here. Very few are certified or wheel measured. Most events are run by event companies rather than runners etc.
Longboat thanks that's a thorough explanation. What is this data point time adjustment?
Mmmm Bop
2. For the track again, which arm is your GPS on? If it's the right arm and you're going the normal counter-clockwise direction, your turns have a little more than 1/2 wider radius. That's 3+ meters/lap.
Ah, so that’s another contributing factor why my watch showed 28.00 miles for the track marathon I ran last year. 😳
5k - 17:53 (4/19) 10k - 37:53 (11/18) Half - 1:23:18 (4/19) Full - 2:50:43 (4/19)
Zebano that is pretty much how most races are measured over here. Very few are certified or wheel measured. Most events are run by event companies rather than runners etc. Longboat thanks that's a thorough explanation. What is this data point time adjustment?
That's how often the GPS checks it's position. Let's see... on my FR 235, Menu, Settings, System, Data Recording, then the choice is "Smart" or "Every Second". Smart Recording logs key points when the fitness device detects changes in direction, speed, heart rate or elevation. Every Second uses battery and recording space faster, but it seems to be more accurate on a clearly measured course. (I'm not sure why, because if Smart Recording checks every second for changes, it should be recording every second on a track corner.)
Not a lot, about 2/10 of a mile of the difference if the watch was on the outside arm all the way.
I never thought of the fact that my watch on the right hand on the track adds distance.