IMRA Season 2014

Dunebuggy

The action is back on the IMRA Forum :-) I was happy to see the 59 forum posts on Djouce before a huge amount before.

Scar, Glenmacnass is almost at 100 posts, its clear that between the capool and Shoe queries and ahem location queries that people are heading back to the IMRA site ? I actually can see only 2 posts on FB for the Glenmacnass.

FB does seem to attract the photos and until IMRA provides a similar service it may remain that way. The Maps posting is soon to be sorted.

Problem with FB - Is that it may not be the correct data, a bit like people picking up our calendar dates and sometimes not reflecting our changes...

IMRA website in my opinion continues to be overall the best IMRA source and as long as FB doesnt have an IMRA specific structure we can use. It should remain the place for the action.

I always liked the boards forum for the sideways rants !!! Bring back the rants !!!!

Slogger Jogger wrote: »

I'm not sure what the issue is but it appears to be difficult to get up-to-date maps etc onto IMRA events on the website. This is compounded by routes only getting agreed at the last minute. No fault here with the people concerned who are doing great work, but its the processes and hoops that have to be jumped through on permits etc from what I can tell.

I've had issues any time I tried to upload photos to IMRA site so I just gave up.
Facebook isnt ideal but it allows for easy photo uploads and chatter.
The IMRA website was great for its time but elements of it look stale now in my opinion e.g. news items on the home page can be there for months on end, so its not very dynamic and maybe thats one reason why people are finding favour with the FB IMRA group.

Peterx

Enduro wrote: »

I don't recall that it was ever the case that clicking on a runner would bring up his results history page. AFAIK it has always gone to the details page, as that's where the summary info is. Believe me, it makes no difference to me (with my webmaster hat on) which one it goes to. It may once have done, as the queries on the results page would once have taken a lot longer to execute and used a lot more server resources. That's changed in the recent massive backend re-write, so that page now flies.

And for clicking on an event... try using the site and get back to me and let me know what problems are actually there as opposed to the ones that you imagine are there. Real feedback is great to get. We don't have a bug testing team so the only way we find out issues is when someone tells us. But it would be good if you could actually ensure that it is a current issue.

No data has been lost AFAIK. There is a lot of hard work goes on to ensure that. If you think something has been lost let us know, and we might have some chance of getting back from an ancient back-up somewhere.

Golden penguin :D:D love it.

T runner has a point about the enduro points, they mean nothing and are non-intuitive. % makes sense.
In bug news the thing that is supposed to let you reset your password so you can log into myIMRA when you forget your password doesn't work. Orla and others have been trying and failing recently.
75 minutes on the new Carantoohill route, all hail King Tim!
The more race reports the better

Enduro

Peterx wrote: »

T runner has a point about the enduro points, they mean nothing and are non-intuitive. % makes sense.
In bug news the thing that is supposed to let you reset your password so you can log into myIMRA when you forget your password doesn't work. Orla and others have been trying and failing recently.
75 minutes on the new Carantoohill route, all hail King Tim!
The more race reports the better

Enduro points are meaningless in themselves. However they have very much served their purpose, which was to stop people deliberately DNFing when they were doing badly in a race so as to minimise the effect on their average percentage.

This weekend was a good illustration of the crapness of winning % as a measure of anything except your performance against one runner. Tim had such a dominant run (what a run!!) that everyone else's percentages are relatively crap, despite loads of people running superb races. So if Tim hadn't turned up, everyone else's percentages would have been radically different. It's a very poor measure of performance.

The password reset was broken for one day last week. Can you check it again for me and let me know if it's still causing a problem. (I double checked, and it's working fine for me).

Peterx

Enduro wrote: »

Enduro points are meaningless in themselves. However they have very much served their purpose, which was to stop people deliberately DNFing when they were doing badly in a race so as to minimise the effect on their average percentage.

This weekend was a good illustration of the crapness of winning % as a measure of anything except your performance against one runner. Tim had such a dominant run (what a run!!) that everyone else's percentages are relatively crap, despite loads of people running superb races. So if Tim hadn't turned up, everyone else's percentages would have been radically different. It's a very poor measure of performance.

The password reset was broken for one day last week. Can you check it again for me and let me know if it's still causing a problem. (I double checked, and it's working fine for me).

If the entire point of enduro points was to stop people DNFing just get rid of them so and give all DNF's the time of the last runner home. To be honest I never ever heard that as a reason before. And what is the problem with people DNFing as long as they let the finish line folk know?
I thought the entire point of enduro points was to produce a "better" metric.

The percentages do work and do make sense. Once again, so what if Tim O'Donohu e shows up and blasts the course record, you get your percentage of the winner, whoever he or she is.
Orla and I tried that password reset on at least 3 different days but we will try again now, thanks.

Peterx

That password reset is working perfectly. Thanks again.

Race report - Ballincollig...

Ballincollig/Kielduff, Tralee, Sunday 1st June – All that is good about mountain running..

1. It took place in Kerry. I’m completely biased when it comes to my native county. But it’s good that the most mountainous county in Ireland and the home of a World Champion and legend of the sport gets another race. For those who lifted their heads, there was some wonderful scenery to take in across North Kerry and the Shannon Estuary. And it was good to have it in this corner of the county too. The Reeks and Brandon and Mangerton may be the big draws for climbers, walkers and runners around here, but this was a great challenge in the stamping ground of John Lenihan, the place he honed his craft. And if that doesn’t inspire, nothing will.
2. I found the course extremely good, though I like the ones with variety over the straight up and down challenges. And this one had it all, a very short stint on tarred road before we moved to track, paths churned up by cattle, very steep open hillside, a lovely fast stretch across a ridge, back on to trails, and on to a second high point before a long fast descent.
3. It was very well marked. Not only with stakes and tape along the way but there must have been marshals at at least 10 different points. And even water stations.
4. The support was amazing, I don’t think I’ve ever seen so many photographers, kids out with cowbells, video cameras, people ferrying runners from the finish back to the pub afterwards...it’s was lovely to see. It seemed like half of Kielduff turned up to pitch in, clearly a sign of the very high regard in which they hold John Lenihan and the affection for him.
4. All the usual good aspects of the sport, cheap entry, lots of good food afterwards in the company of a great bunch of people. One small thing spoke volumes for me, the run was advertised and 7km but turned out to be 7 miles or so. When the news filtered through, the unanimous reaction was laughter and that “arra let’s get on with it” healthy attitude I really like.

All in all, I really hope this one is kept on the calendar and more turn up next year to sample the Stacks Mountains.

T runner

Enduro wrote: »

I don't recall that it was ever the case that clicking on a runner would bring up his results history page. AFAIK it has always gone to the details page, as that's where the summary info is. Believe me, it makes no difference to me (with my webmaster hat on) which one it goes to. It may once have done, as the queries on the results page would once have taken a lot longer to execute and used a lot more server resources. That's changed in the recent massive backend re-write, so that page now flies.

It was the case before the Golden penguin awards. 9x% of the time when a person clicks on a runner. It is to see their past results, what events they competed in, if they competed in particular events that the person is familiar with etc. It doesn't matter how fast you think the website is (slow), clicking multiples times to get where you want to go instead of clicking once, increses the slowness by a matter of thousands of percent.
This change has been flagged and pointed out many times since the awards.
It was changed to allow for the runners "stats" which no-one looks at nor cares about. What do your page usage stats tell you?. People arriving by default at the runners page will 9x% of the time click onto that runners results right?.

And for clicking on an event... try using the site and get back to me and let me know what problems are actually there as opposed to the ones that you imagine are there. Real feedback is great to get. We don't have a bug testing team so the only way we find out issues is when someone tells us. But it would be good if you could actually ensure that it is a current issue.

In fairness, You've been getting real feedback on the Ridiculous "Enduro" points since their unfortunate inception. If you were worried about "real" feedback surely you would have changed those by now?
Can you change them? We could call the percentages "Enduro percentages" if it will make the move less painful for you?


The other issue (events) seems to have been changed recently: that's good. But there has been real regular feedback on this since the retrograde change for the awards many years ago. Now bring up a runners results and a user can stay in the super results database until she/he wants to have a look at someone's Enduro points (Which is never)

T runner

Enduro wrote: »

Enduro points are meaningless in themselves.

Good at least you admit this now. Race points were calculated by multiplying base points by a value of 1.x Where x (0-9) was awarded depending depending on subjective difficulty.

This meant that for example, the person in 30th place in Carrauntoohil, got higher points than the winner of Maulin winter. Instead of dropping it, you reduced the multiplier by a factor of 10, a number to use current anglo speak, plucked directly out of your arse.
Enduro points are calculated by adding the last 8 race points toghether, in a certain timeframe giving some arbitrary big number.

You said it. Completely meaningless.

However they have very much served their purpose, which was to stop people deliberately DNFing when they were doing badly in a race so as to minimise the effect on their average percentage.

Can you demonstrate what evidence there is that there less people deliberately DNF-ing since the change?
While you're at it can you supply the evidence that there were actually people deliberately DNF-ing before the change?

This is relevant since this is now your official retrospective reason for the introduction of the system.

This weekend was a good illustration of the crapness of winning % as a measure of anything except your performance against one runner. Tim had such a dominant run (what a run!!) that everyone else's percentages are relatively crap, despite loads of people running superb races. So if Tim hadn't turned up, everyone else's percentages would have been radically different. It's a very poor measure of performance.

% points are a measure of how you performed against the race winner. Their great advantage is that they are understandable and people actually use them.

People don't understand and don't use your system. Regardless of how great you think it is: it is useless if it is not used. You haven't been able to convince people to use it in the last 4 years. It needs to be dropped.

Carrauntoohil e.g

Using % ages: A person looking through their results will see a low percentage for Carrauntoohil. They will look at the winner and see Tim's time. They will ofcourse understand the anomaly. And start looking for how they fared against runners known to them (compare their %).

They can also click into a runners results (2 page loads now due to the default stats, and enduro bull page) and quickly verify that that runner's standard by his %ages in other races. The joys of the results database!

Crucially: the % ages for Carrantoohil will also tell them how they fared against the best open mountain runner in all the land

That's important info. How you stand V's the best is something everyone wants or should want to know. Don't you agree?




Using your system. The person cannot gauge his/her performance because, well, they don't use your system. Sorry but there it is.


Since the introduction of the new points and diminishing of %ages people are finding it more difficult to gauge performance. The level of hostility shown to the change over the years is proof enough.

Lastly, a question many might have wondered: Why call it "Enduro" points? :pac:

nerraw1111

Anyone know if DNFs count towards the 7 races in the Leinster League?

Peterx

nerraw1111 wrote: »

Anyone know if DNFs count towards the 7 races in the Leinster League?

I'm hoping I can use 7 DNS's

Enduro

nerraw1111 wrote: »

Anyone know if DNFs count towards the 7 races in the Leinster League?

They count towards races for the end of league prize purposes. For that reason they get counted in the league results page on the website, with a position as if they finished after the last counting finisher for league results caclulation purposes.

Timmaay

God my brain hurts trying to understand all of this ha, the beauty of Imras is the closely knit community with loads of head to head competition against your rivals. Times, pbs and any other statistics are largely meaningless and this is a refreshing break from the likes of track and field or road racing!

Slogger Jogger

Peterx wrote: »

I'm hoping I can use 7 DNS's

or 7 race reports for races not done

Slogger Jogger

Back on topic of mountain running....

... this evenings race on Carrick was an absolute blast. In previous years the fire road when on forever. This evening there was none of it. Single track, no track, off road, severe climb, severe descent. Brilliant. And no early starts!

I think Bernard was 1st. He is flying at the moment and the route was well suited to his abilities.

Kurt Godel

Enduro wrote: »

Enduro points are meaningless in themselves.

T runner wrote: »

Good at least you admit this now.

Don't mean to put words in Enduro's mouth, but I took from what he said to mean that 432 Enduro points have as much intrinsic value as 157%. In that both need specific context to have meaning.

As regards race results, generally a score based on the 30th percentile will give a better flavour of an individuals overall "how-well-did-you-do?"-ness, as it lessens outliers and is pitched nicely in a sweet base spot between winners and mid-packers. In that respect Enduro points are better than percentages based on the winning time (you need to appreciate the ability of the winner more to comprehend what everyone else's % is based off). The 30th percentile by definition is far removed from outliers who would skew a metric.

The problem (as I see it) with Enduro points is the various multipliers, with weighting for terrain, elevation gain, distance, etc. All of these are subjective, and any formula containing them should be improved year-on-year to hone and improve the equation. This hasn't happened (as far as I can see), so they are essentially too subjective. If they aren't going to be refined, they'd be better off removed, IMO.

I've read Eoin's original proposal for an improved IMRA ranking, and it explains thins very well. Haven't seen it in the general domain though- has it been made available via the IMRA website? If not, it should be, as it would certainly inform the debate to a useful degree.

Timmaay

Slogger Jogger wrote: »

Back on topic of mountain running....

... this evenings race on Carrick was an absolute blast. In previous years the fire road when on forever. This evening there was none of it. Single track, no track, off road, severe climb, severe descent. Brilliant. And no early starts!

I think Bernard was 1st. He is flying at the moment and the route was well suited to his abilities.

Agreed. There I was thinking I'd have a nice handy trail run for my Imra comeback, couldn't have been further from that ha. But had a total blast, if only I had some hill fitness!

ThisRegard

Slogger Jogger wrote: »

In previous years the fire road when on forever.

That part of the course was horrible, I even had one guy remark that it was a road runners course one time after the race. Typical that I couldn't make it last night for when it was dropped.

Enduro

Kurt Godel wrote: »

Don't mean to put words in Enduro's mouth, but I took from what he said to mean that 432 Enduro points have as much intrinsic value as 157%. In that both need specific context to have meaning.

That's exactly what I meant, thanks!

Kurt Godel wrote: »

As regards race results, generally a score based on the 30th percentile will give a better flavour of an individuals overall "how-well-did-you-do?"-ness, as it lessens outliers and is pitched nicely in a sweet base spot between winners and mid-packers. In that respect Enduro points are better than percentages based on the winning time (you need to appreciate the ability of the winner more to comprehend what everyone else's % is based off). The 30th percentile by definition is far removed from outliers who would skew a metric.

The problem (as I see it) with Enduro points is the various multipliers, with weighting for terrain, elevation gain, distance, etc. All of these are subjective, and any formula containing them should be improved year-on-year to hone and improve the equation. This hasn't happened (as far as I can see), so they are essentially too subjective. If they aren't going to be refined, they'd be better off removed, IMO.

Yeah, you're right about that. It would be great if a few of the more statistically minded runners could crunch the data and come up with more optimal parameters (I'm all ears if you want to propose anything!). In the end it will always have some element of subjectivity to it. But then the same applies to Decathalon, for example, and most people seem to be happy with their points allocation system.

I can definitely see the merit of dropping the race-weighted points (race points) an just leaving non-weighted points (base points), and deriving the ranking points from a sum of non-weighted points. It would reduce the level of complexity quite a bit, which would hopefully help with the general understanding of how it works.

Kurt Godel wrote: »

I've read Eoin's original proposal for an improved IMRA ranking, and it explains thins very well. Haven't seen it in the general domain though- has it been made available via the IMRA website? If not, it should be, as it would certainly inform the debate to a useful degree.

An FAQ on the all the available stats, along with some good graphing to illustrate what's happening is on my list of "nice to haves" for the site. It's low priority at the moment though, as there is a massive amount of work being done on core functionality.

T runner

As regards race results, generally a score based on the 30th percentile will give a better flavour of an individuals overall "how-well-did-you-do?"-ness, as it lessens outliers and is pitched nicely in a sweet base spot between winners and mid-packers. In that respect Enduro points are better than percentages based on the winning time

Kurt. I think you're mixing Enduro points up with base points.

I believe Enduro points are the sum of the "race" points total of the last 8 races within a certain time period (don't know how long, anybody?).

Race points are calculated by multiplying base points by .x (where x is race difficulty) and adding the result to the base point total. i.e base points * 1.x

As mentioned when they were brought out initially, 30th in Carrauntoohil was equivalent to 1st in Maulin Winter in Race points.

Enduro points are then calculated based on 8 races.
Enduro points are kind of like to financial instruments during the great crises of 2008: cant understand them and are probably toxic.

(you need to appreciate the ability of the winner more to comprehend what everyone else's % is based off). The 30th percentile by definition is far removed from outliers who would skew a metric.

I've no problem with base points used as a metric. Along with % ages they do indeed tell the story of a race. Base points don't cover all anomalies though. A win in the Howth winter race generally gives the winner around 150 base points. High points for the top 10 in this race are consistant from year to year. This doesn't reflect performance either.

The problem (as I see it) with Enduro points is the various multipliers, with weighting for terrain, elevation gain, distance, etc. All of these are subjective, and any formula containing them should be improved year-on-year to hone and improve the equation. This hasn't happened (as far as I can see), so they are essentially too subjective. If they aren't going to be refined, they'd be better off removed, IMO.

The main problem is IMO that nobody actually understands it, has confidence in its accuracy, or actually uses it.

I've read Eoin's original proposal for an improved IMRA ranking, and it explains thins very well. Haven't seen it in the general domain though- has it been made available via the IMRA website? If not, it should be, as it would certainly inform the debate to a useful degree.

Will it be used?
Even forcing people to visit the Enduro stats page when clicking on a runner made no difference. As well as helping to destroy easy and fast browsing of the IMRA results database it did not influence peoples use of the grammatically incorrect Swahilli "Enduro" points.

Peterx

The percentage (ONE single metric) worked very well.
If the IMRA website could pick either % of the winner or % of the unknowable person who came 30% percentile and stick with it then it becomes useful. I personally prefer % of winner.

There are always a few people who get very exercised by this topic, myself included, but most people will take a casual look at their percentage and have their understanding of it. Unfortunately, the same cannot be said of either race or enduro points.

I tried to have a quick look at my results from 2009 and as T runner says it is a pity that my percentages are not visible on the my results page. The enduro points are meaningless. The hardest races with the best depth of field I did that year, or any year generally, were the trial races, which the enduro points don't take note of.

Ar Strae

Peterx wrote: »

as T runner says it is a pity that my percentages are not visible on the my results page.

This is the bit that gets me. I've put in a few requests for it. By all means keep the Enduro points, refine them, simplify them etc. But please can we not just have the percentages also visible on the results page. One more column beside the Base Points and Race Points. So in a quick (and admittedly not perfectly accurate way) I can glance down either my own results and pick out the anomalous races, or get a good feel for how a potential "mark" does across a season's racing.

The data is already there, I assume this would be a small change to have it displayed (as it used to be?) on the Results tab?

Obviously I still think the website is brilliant, the resources we are getting as members, online race vouchers mean that a race is only a fiver! For which we get recceing and devising of new courses, results, marking and demarking, marshalling and first aid support, fun-size bars and the chance for a bottle o' wine in the pub! And a website with all the race data collected and visible. It's insanely good...but just one more column on the Results page? :-)

Ro

Anyone know about this Rogaine?

Was just asked to enter as a team with another yesterday evening. Never really thought about it before. Would have okay navigating skills, do a bit of walking so can point and go with confidence. However not great at say more technical aspects like triangulation to work out position or pacing on different terrains.

How much navigating skill is needed?

stirwell

Anyone going to try out sorrel hill tonight?? Anyone been there before?

Ar Strae

I'll be hitting Sorrell Hill all right. I need 3 more races to hit the seven and I am only around for 4 races so it's tight! :-)

I have apparently run Sorrell Hill twice? I don't really remember much about it other than there were amazing views over Blessington.

I see that they were planning a last minute route change to cut out some of the fire road (hat tip to RD Niamh's Carrick race) but the distance (~10k) and the ascent (~400m) stay the same.

Don't forget that there is a 200 person limit imposed on this race so get there early. Just not earlier than me!

Ronan

nerraw1111

It’s a nice race. A fireroad to start with before turning off into a forest onto open mountain. Going is a bit tricky in places. It’s undulated for about 3km before hitting Sorrell for about a 1km climb. Back the way you came. Some superb views of the lakes when you do manage to glance up.

ThisRegard

Yeah, it's a nice course, weather depending it can be quite sloppy and wet in the flattish section after you climb the fence out of the forest and before you start the actual climb of Sorrel Hill itself. It's where I got one of my IMRA injuries. Stood in a bog pool but there was a rock beneath it which I slipped on and the rock then proceed to act like a cheese slicer along my shin.

Any change that cuts down on fireroad is always to be welcomed !

Peterx

nerraw1111 wrote: »

It’s a nice race. A fireroad to start with before turning off into a forest onto open mountain. Going is a bit tricky in places. It’s undulated for about 3km before hitting Sorrell for about a 1km climb. Back the way you came. Some superb views of the lakes when you do manage to glance up.

It's a good old fashioned out and back race. Very handy for the seeing where lads are as they come flying past.
There's a lake view???

axe2grind

Peterx wrote: »

There's a lake view???

Reserved for those at the back

Peterx

axe2grind wrote: »

Reserved for those at the back

I spend a lot of time in hillruns looking at backs

nerraw1111

Feels like a waste to run all that way and not sneak a look....which may explain my number of falls this year.