Discussion: "Adversarial" sprint course setting?

I'm splitting this off into a separate thread since, as an orienteer who actually very much enjoys urban sprints, I'm interested in the topic.

EricW posted a mini-rant here about "adversarial course setting", and I'd like to hear more details about what Eric (or any other interested parties) mean by this. In particular, what is the difference between an "adversarial" course, and a course that is a challenging but fair test of one's ability to read a ISSOM map at speed?

I have several pet peeves about sprint course "traps" that personally make me unhappy when I encounter them, but I'm interested in hearing the point of view of others first.

I think there is a very very fine line between a "trap" and a good route choice problem. Here are some of the things I think about... not a definitive list, but certainly long.

- Clearly legible pathways. You want to test whether the runner has read ahead and considered all the reasonable options. But the runner shouldn't have to squint to see a gap. If a route choice is important, it must be clear. (That is where ISSOM 728, 525, and to a lesser extent, 506 come into play, and should be used as guidelines for legibility).

- The same is true of obstacles. If an obstacle is important, it should be distinct. Thin bands of olive green are not distinct. Minimum thicknesses of 0.4 mm ought to be observed, and distinct boundary lines should be included.

-For the sake of clarity to beginners, I try not to use symbols 421 and 528 on the same map. Which usually means 421 is scratched.

- The response of a runner to mistakes should be considered. If a competitor runs into the trap, what are their possible responses? A course setter must avoid trapping people in situations where the runner will be tempted to cheat or commit a dangerous act to recover from their mistake.

-Uncrossable features should be visually distinct in the terrain too, and should have clearly defined edges. You really want to avoid having beginners ask themselves "did I just cheat?" or "is that cheating?"

On a related notes, course-note statements like "don't cross x feature" are scary, because runners don't necessarily know the boundaries, or how they will identify the features, or how well the feature will be mapped.

A beginner runner might see a path going through olive green and not know if it's legal to take. Heck, a lot of long-time orienteers probably don't know the answer to that question.

-If obstacles are going to play a very important role in your sprint course, it's probably best practice to build an example into the model event. Or think about some other way of communicating this to competitors. (Things to address: What the important feature will look like. How it can be identified in terrain. How it will be identified on the map. Is there anything in terrain that might look similar.)

-(Debatable?) The line of approach should be deducible from the map, without needing to read the control description... (at least for beginner courses)

- Avoid "shoehorning" control descriptions (i.e. describing a wall as a cliff).

A good list of generalities and details.

The point about "course-note statements" critical to the course is one of my long term pet issues, and applies to all formats. Almost invariably this indicates the course setter didn't do his/her job.

I'll add one point- any situation involving the double dot line underpass, 518.1, is a potential problem for the general O public. This symbol is simply too weak or confusing given its importance as a viable route. I recognize that hardcore competitors and course setters love this feature, and has it been critical in at least two WOC Sprints. I don't have a solution, other than "not for beginners" and "handle with care".

Adversarial course setting-
For clarity, I am talking about the "course setter versus competitor" attitude, which I believe inevitably leads to bad things in the long term.

I believe the course setter should first SERVE the competitors as a group, and to secondary degrees the fellow event workers, and sport in general. All too often the course setter is thinking about themselves, and "my course" and "gotcha", rather than trying to make/find the best course possible for the most people.

I do not intend this term to be exclusive to the Sprint format, although iit seems to rear its head more often in urban Sprint events.

I can agree that the difference in the finished leg/course is sometimes a fine line and open to legitimate debate, but not always. Some competitors will love the same leg/course that causes others to never return. I would encourage "toning it down", rather than risk turning people away at a general event, but for a select group the riskier, more interesting approach might be appropriate. A "select" group might apply to WOC/ WCup events. It would certainly applyl to a designated advanced course at a non-sanctioned recreational event.

To a large degree this issue is terrain related. In the other ISSOM symbol threard, Tundra/Desert introduces a useful term, obstacle and non-obstacle terrain, which covers much of the problematic situation.

While this attitude is not always evident on the printed course, it can also be witnessed in micro control placements and the course setter's own words and chuckles.

Agreed, we are all entitled to free speech, and in many settings this attitude has little effect. Ultimately we don't know the content of somebody's heart, but in the long term I think this attitude is not good for the sport, and should be discouraged.

double dot underpass...

In many, many situations when the course goes under the underpass there is no plausible route over the top. It means the "underpass" is the "main running level", in which case it should, IMO, be mapped using the canopy symbol (which is not weak).
And it shouldn't have uncrossable walls at each end.

People have aesthetic dislike of roads going from brown to gray, but to me clarity is more important - if you aren't running down the road, why would you care how it looks on the map?

IMO

Traps...

are good. A leg where the "obvious" route leads up a dead end is fair game. It's a bit worrying that a single leg can be "game-over" in a sprint but the benefit of knowing its a possibility keep runners on their toes throughout, even legs which aren't traps.

Tricks...

however are bad. Course setter's chuckles about micro placements: "Ha ha, you went there and didn't see it - how rubbish are you?"

Similarly, "Ha ha - can't you understand the clue-heiroglyph for
'in a disused lavatory with a sign on the door saying Beware of the Leopard' "

Double dot underpass. If such an obscure feature is to be so important - there was one in WMOC qualifier in Pecs - use purple overprint symbol on it. Its a decision making exercise after all - not an eyesight test.

The real difference between good and bad is in the competitor response.

You want "boy, what a fun challenge!"
You don't want "what the @%#$ was that?"

And one must always keep in mind that what frustrates a neophyte is very different from a seasoned orienteer.

We regulars have internalized so much over the years... we only tend to notice changes from the norm. If some challenge doesn't exactly match with our hard won skillset, we can get pretty annoyed pretty quick.

Neophytes have no norm to compare with, and no vested interest in a certain type of game. They will accept practically whatever challenge you give them, as long as it seems achievable, meaningful, and well defined.

But if a neophyte feels like they are just walking around some campus with a map, instead of exploring and searching and seeing beautiful sites.... or if the neophyte keeps getting stuck on the wrong side a wall with no idea why... well then their perception can pretty quickly turn from meaningful challenge to pointless waste of time.

In many, many situations when the course goes under the underpass there is no plausible route over the top.

If the main running level is the same on both sides, canopy is the correct choice.

The problem arises when you have buildings built into the side of a hill, and the running level is different depending on which side of the building you're on. If you draw a canopy connecting to the uphill side then the runner could reasonably expect to see a passage at the main running level on that side, and then get horribly confused.

The latest post on the RouteGadget FaceBook page is an eye-opener on how many of the legibility issues with urban sprint orienteering can be reduced by letting go of ISOM colour conventions in the ISSOM. It is really worth a look.

I would agree with the sentiment that traps are ok, with the caveat that it should be obvious during the race that you've fallen for the trap, so that you can go back and still punch the control without DQing. This is in contrast to the "trap" where the course setter deliberately puts nearby controls on similar features to check whether competitors are paying attention, and if not it's a mispunch.

Also, if the trap involves "uncrossable" features which are, in reality, able to be crossed, then some kind of caution tape should be put up or marshals posted to redirect competitors. Otherwise some people will forget/be unaware of the rules.

How is it a trap to have nearby controls on similar features? (Provided they are far enough apart per the rules.) It is obvious during the race if that has caused a mistake - it's called checking the control code. That's something I learned to do on my very first day of orienteering.

As a course setter, I prefer to avoid situations in which competitors don't get a result. I feel no satisfaction from "testing" people like that. It's not a question of whether it's legal or not, it's a question of having the best possible outcome. In my opinion, the best possible outcome includes having no mispunches.

I think a lot of the perception of adversarial-ness comes from the fact that obstacle courses (as in T/D, not as in mudders) are less forgiving to sloppy map reading in general, and reactive navigation style in particular. Someone like myself could get away with being mostly reactive most of the time, especially in North America where route-choice legs are infrequent and never decisive, and still do pretty well... until Sprints came around. A lot of the offended-ness comes from the realization, conscious or not, that one's skills were not up to the challenge—to a greater degree than anticipated.

I am all for meritocracy, but indeed if meritocracy is going to upset peace in the family, perhaps we the course setters should tone it down. Note: if your whole course is heavily obstacle-loaded, like the multilevel urban terrain we have in the Bay Area and in Vancouver, experience shows that even the top navigators will make plenty of mistakes, so the overall public feeling afterwards won't be as bad since most people will be about as far back as they expected to... the most problematic Sprint courses are ones with a small number of decisive obstacle-driven legs. It's these ones that course setters should be wary about.

At the Scottish Sprint champs a couple of years back there were two controls near the end of the courses, one on each side of a high cypress hedge. The problem I think was that the control on the near side was visible to all approaching runners, so the chances of MP if you should have been heading to the far side of the hedge was relatively high. An unintended trap I am sure but a trap all the same and it caught a number of people out.

I like graeme's idea of mapping underpasses as canopies, but only if the road itself is OOB or completely off route, otherwise its going to confuse anyone coming down that way.

I also feel that there is a difference between what we call sprint (12-15min EWT) and Urban (45min +EWT) in the UK when it comes to "traps". In longer Urban races you have much longer to work out routes and generally more dead running to think ahead. If I get caught out staring up at a control on a bridge above me because i didnt read my control description in an Urban race then I say "Well done!" to the planner and try and figure out how to correct my own stupid mistake.

When I first started orienteering over 20 years ago, I became aware of a number of course setters who thought it was perfectly OK to set controls in areas of thick green, or to set bingo controls in bland areas. Their logic being that "orienteering should be a navigational sport, and it's not fair that the fastest runners win all the time". So by slowing everyone down to the same speed in heavy fight, or by making the time it will take you to find a control essentially random, the slower runners are given more of a chance to win.

I would hate to see this same ridiculous philosophy of course setting carry over into the sprint discipline. I think it's equally poor course setting to bait fast runners into entering areas of olive green, or to force runners to quickly and correctly interpret an ambiguously mapped, multilevel area of canopies and underpasses. Is there such a thing as a "bingo" control in sprint orienteering? I think there certainly is, and as I read these threads, I see that others clearly agree with me.

Besides all of the technical considerations that go in to course planning, I strongly believe that the best single thing a course planner should do (whether for a forest or sprint course) is to objectively ask themselves (and honestly answer): How would I feel as a competitor running this leg/looking for this control; would I feel that it was legit, fair, enjoyable?

Richard Feynman talked about creativity within the constraints ("handcuffs") of the laws and equations of science. I've always felt that the course setter must be both adversarial and empathetic. Empathetic in the sense that orienteering is not an objective race---there are lots of unfair things going on during the running of a course, and the course setter's job is to get the runner through the unfair stuff as smoothly as possible. Avoidance of excessive green and climb and bingo controls and the shortcomings of the map is the job of the course setter looking out for the competitor. But *within the confines of fairness and the accepted rules of the game*, certainly the course setter is adversarial. The best criteria I know for when the course setter has screwed up is, "I know it when I see it, more or less." For sprints, the biggest problems I've seen are legs set where the runner has to see a tiny feature (a 0.35mm gap in a wall, for example) to execute the best route.

I really enjoyed the IU Sprint. I was going really slow, but it was still fun, with a lot of variety. My course's first leg was the opposite of the 0.35mm problem I mentioned above---My first read was "go around to the right", but then I saw this enormous gap that saved a few steps---and I saw other people who didn't see that big gap. The olive green control was a big area of olive green, not a thin strip or a small patch. While I wish we had a posted an ISSOM reminder poster in the start area for the Interscholastic competitors, my experience on that leg was "Is that third green or the pukey-green out-of-bounds?---Then I looked for some real third green, found a tiny patch of it, and aimed for the legal sidewalk route. But admittedly, I'm an ISSOM geek---I probably read it twice a year, mostly because of threads on attackpoint.)

On the Yellowwood Long, I didn't run that day, but I did do a small amount of vetting a few weeks earlier. I was only concerned about the climb and difficulty on the Brown course---A lot of time went into trying to improve the Brown course, and while I don't think anyone was completely happy with it, it was the best we could do within the limits of time left to print maps and the difficulty of driving hours to do further checking in the park (adding controls). I thought *all* the courses were "tough but reasonable". It was the setter's first time setting an A-meet, and I was impressed with the scale and vision and effort involved.

best criteria I know for when the course setter has screwed up is
When I check to see if I screwed up, here's what I look for.

Lots of MPs.
Lots of dnf 's.
People going OOB.
Controls being found by groups of people, not individuals.
Legs on winsplits where the "red/error" is evenly spread across runners rather than concentrated at the bottom.
Target times missed by large percentages.

I might even talk to people and see if they had a good time, but everyone around here is so nice they always say they did. (Or maybe its that big stick I carry?)

Although sometimes those things aren't indicative of bad planning, usually they are.

Those are all good things to look at, but they're fairly broad indicators. On a trap leg, I think it's much more subjective and less likely to be found with those indicators. For example, what if 90% of the runners "went around" and 10% took the 0.35mm gap? (I'm assuming the gap is the best route.) That is unlikely to be seen as red in the results. No MPs. It might hurt the target time.

One of the original criteria listed by IOF was "ability to solve the leg at speed." I've always disagreed with that---I've felt the tempo change from standing still for a few seconds is a desireable "sprint feeling" goal. But at some point as the legs become more complex, most people aren't going to solve the leg in a few seconds, and that is where I start to wonder if it's a bad leg.

The obvious counter example is that the best map readers are probably going to see the solution sooner than others, and that is a desireable outcome. I'm not sure how fine a line it is, but at some point in map complexity the trap solution becomes just too arcane and too hard to see.