Discussion: OUSA Rogaining Committee Seeking Rogaines

If you are part of club leadership or run an events management company in the US, please help us find rogaines and good locations for rogaines by taking this survey. If you have ideas for places that would be great for holding a rogaine, let us know in this thread!

Click here to take survey.

The OUSA (USOF) Rogaining Committee is beginning the selection process for the 2011 US Rogaine Championships. Before making our decision, we want to be certain we know of all 24-hour rogaines being planned for 2011. We would appreciate your letting us know if there is a possibility that your organization may organize a rogaine in 2011. We are hoping to gather bids for US Rogaining Championships by February 20th; however, if your organization cannot make that deadline, please let us know and we will work with you.

To be designated US Rogaining Championships, a rogaine must be sanctioned by the OUSA Rogaine committee. While we encourage all rogaines to go through the sanctioning process, this is a requirement for the US Championships. Sanctioning is intended to be helpful for organizers and to ensure that the event is safe, enjoyable, fair and competitive. Sanctioning fees are the same as for A-meets.

If you think you have an area that might be suitable for a rogaine, and would like help from experienced rogaine organizers or course-setters, please let us know and we will do what we can to make your rogaine happen. As rogainers ourselves and as the committee tasked by OUSA, our goal is to see rogaining grow.

Barb Bryant, Peg Davis & OUSA Rogaining Committee

By the way: web page with US Rogaining news

One location with great potential comes to mind: Ohiopyle state park. Easily connected by two interstate highways (I-76, I-84) and relative proximity to large metropolitan areas, rugged terrain on 20 thousand acres, area additionally extended by other public lands or lands in public trust (State Gameland 51, Bear Run Nature Reserve, the latter also location of Frank Lloyd Wright's "Fallingwater" house), and most important, park administrators that are accustomed to giving permits for similar events (adv race).

Nice! Nearest O clubs? Adventure racing companies who operate there?

Joe knows well the adventure racing organization that uses Ohiopyle: he raced there last October.

Other AR organizations: Grass Roots Racing, contact name: Frank Eyth

Morgantown Running is a shop based in Morgantown, WV, and center of many local outdoor event organizers, and open to volunteer.

Walls Are Bad is a non-profit group chartered with promoting the outdoor assets of SW Pennsylvania and facilitating communication among different outdoor clubs and organizations. They have great connections to other groups and organizations. Contact name is Ginette Walker Vinski (GVinski atsignhere sustainablepittsburgh.org)

O-clubs nearby: WPOC (1-hr drive), SVO (2-hr drive), QOC (3-hr drive)

Explorers Club of Pittsburgh is an enthusiastic group that use that park alot for their wilderness activities.

Laurel Highlands Ultra is a once-a-year running event that starts in that park, and its organizers and participants are a commited group to see the local assets used.

Ohiopyle state park. Easily connected by two interstate highways (I-76, I-84)

Actually three: I-76/70 (PA Tpk) to the north and I-68 to the south.

Google Map

Thanks!

Perhaps it would be useful to know the approximate size of an area and type of terrain used for a rogaine so that people could think about this even if they have never been in a rogaine.

The size of an area that would be appropriate for a rogaine depends on the passability and steepness of the terrain. It happened many times that an area was chosen for a 24 hour rogaine that was much smaller than would be appropriate, and the opposite has also occurred, although less frequently. Sometimes the area is so small that the winning teams clean out the course and have time to shower and eat and sometimes sleep for a few hours by the time the 24 hour limit expires. A rogaine should not be a timed race, it's not a part of the philosophy and Orienteering USA will not sanction rogaines for which the area is grossly inadequate.

Of course, whether or not a team will clean out the course also depends on the speed, endurance, and technical skills of the team. In North America, the expectation is to plan the course assuming that top North American rogainers show up. The course should be such that under ideal conditions, the top team would clean it out, so that teams who go for the win are tempted to plan a route that includes all, or nearly all, checkpoints. Deviations from this philosophy are possible, but not gross deviations. If you set a course that is twice as large as needed, you just spent all this extra time creating a challenge that won't be taken up by the participants, and also created unfairness if one team goes east and finds conditions that are very different from those who go west, and ends up scoring the same number of points/checkpoints by working harder or less hard.

Of the rogaines I have participated in, the smallest adequate area was about 75 km² (Elkhorn 2000, Sammamish OC). The largest areas were in Eastern European flatlands, where you have to go well over 250 km² to fairly challenge teams that maintan ~9 min/km throughout the 24 hours. It is safe to say that an area that is, say, 50 km² will not be adequate under any conditions. If it is so steep and/or thick that even the top people will be held down to >20 min/km, then it is too steep and/or too thick to be used at all for adequate competitor enjoyment.

Since most states rate their parks by acreage, here is the conversion, 1sq.km. = 247.1 Acres
and
50 Square Kilometers = 12355 Acres

I would recommend 2 state parks in W. Mass. Beartown and October Mt. Both over 50 sq.km. with campgrounds and easy access from most of the northeast US.

I think I've heard from PG that October Mt., at least, is one big laurel patch, though I haven't been there myself.

The course should be such that under ideal conditions, the top team would clean it out, so that teams who go for the win are tempted to plan a route that includes all, or nearly all, checkpoints. ...If you set a course that is twice as large as needed, you ...also created unfairness if one team goes east and finds conditions that are very different from those who go west, and ends up scoring the same number of points/checkpoints by working harder or less hard.

I'm curious about whether this is an opinion or whether this is stated in some official documentation on rogaining. I have always said the opposite. I believe that the best rogaine courses are designed so that no team can clean the course.

If a team is planning the best route to reach all the checkpoints, they are solving a different problem from the teams who have the arguably more difficult task of selecting the "best" subset and sequence of checkpoints to maximize their points. I think every team should be solving the same type of problem based on their own strengths and weaknesses, both physical and navigational.

Yes, there can be different conditions on different parts of the map and at different times of day, and sometimes checkpoints are worth more or less points than one would expect based on the effort to reach them. But that is part of rogaining. Even if the top one or two teams are able to avoid the "unfairness" because they visit all the controls, the vast majority of the teams have to deal with those differences and uncertainties, including the winners of most categories. Either it's part of the sport or it isn't, and if it is something that most teams have to deal with, I'd say it's part of the sport.

(BTW, I do agree that there is no point in setting a course that is twice as large as needed.)

It's the prevailing opinion (Rule C2, further Manual 6.2.3). A minority of organizers disagree, such as, for example, the organizers of the 2000 and 2010 WRCs. Loud grumbles ensue. The opposing opinion has merit, it's just that the majority of the top people or organizers don't subscribe to it.

Thanks for that, Tundra/Desert. Rule C2 matches my opinion so maybe we're actually in agreement?: "The course shall be designed so that the winning team is likely to visit most but not all checkpoints."

I can totally understand why the overall top rogainers would prefer a course they can clean. I find it easier to plan a route to visit all controls on a map than to choose the optimum subset. But I do think it's more fair if we ask all competitors to face the same type of challenge. And there are many top people in categories like SuperMasters who are rarely going to clean the course, so it seems funny to treat some categories different from others.

Well, "most but not all" is open to interpretation. If you have 60 controls, 31 is most of 60, yet most rogainers will agree that if the top team only visits 31 controls, the course is overplanned. I'd say the interpretation of "most" should be more like 90%, and it is in this interpretation that I believe the majority of top rogainers as well as the majority of organizers will agree with me.

In my original comment, note ideal conditions. There are never ideal conditions, and so a course planned to my standard above will automatically satisfy Rule C2. And it is never the same type of challenge for the top teams vs. top teams in, say, supervets, in most sports you can think of.

In rogaining, with every fewer checkpoint you have to visit, the number of choices goes up by a large factor. In the above example, say you want 90% of 60 controls, that's 60! / (54! * 6!) = 50 million choices. If you want 80% of these 60 controls, that's 1.4 trillion choices, an increase of 28 thousand—how can one say it would be the same type of challenge? This is a realistic scenario that fits your definition, and a 10% decrease in speed is typical for vets vs. supervets.

In say trail running, consider 50 km. The lightning-fast folk will never be in danger of hyponatremia, but those who take over 5 hours can and will run out of sodium if they don't consume. Or consider road marathons. A top female can be drafted/rabbited by a relatively mediocre male, but the opposite doesn't hold. I think you're striving for an ideal that is unrealistic in the overall context of life, and on top of that, mathematically impossible in rogaines.

Once you accept that the challenge will always be different for the top teams and for the top teams in other categories, the answer (for the question of how hard to make the course) is easy. Make it long enough for nobody to get all, and not any longer. This yields my original imperative.

I agree with you that "most but not all" should mean a fair bit higher than 51% and probably no more than 90%. I appreciate your analysis but still believe that removing the "subset optimization" aspect of rogaining fundamentally changes the nature of the challenge, regardless of how big the subset might be (and of course, nobody knows for sure how big their subset should be, which is part of the fun). Agreed that every event poses different challenges to different competitors but I do think that organizers should level the playing field where there are obvious discrepancies they can control.

Fortunately, the number of choices for us midpack rogainers is not quite as great as the analysis would suggest. We need to start and finish at the Hash House, having travelled some distance that we're capable of travelling in 24 hours. When we stretch our piece of string around the map with the start/finish at the Hash House, we usually end up with 3-4 realistic route choices rather than 1.4 trillion - and I am incredibly grateful for that! :)

If the top team can just barely make it in at 23:57 with all of the controls, then they still have to do subset planning in case something goes wrong and they lose a few minutes somewhere. I think that would be excellent course design.

I wouldn't complain about that scenario - although I'd hate to be the course planner attempting to achieve that level of precision! :) I'd rather see the top team get 90% of controls than the top 6 teams get all the controls. That's all I'm saying.

I agree with Bash on the next-to-last sentence.

I'm with jj. It should be a tough call as to whether you can get all the controls or not, or whether your competition can. Keep the top teams guessing. Clearly you don't want 6 teams getting them all. That just means you've under-set. Ideally you want 1 team to get them all, and crawl into the finish on their hands and knees with 10 seconds to spare.

Aim for the best team you expect to show, then if superteam1 and superteam2 arrive, they have to duke it out with speed and clever route choice, just like in any other orienteering race. I don't see any problem with that scenario. I especially hate hanging controls that no one goes to.

We design all of our rogaines exactly this way. Every decision should be as hard as possible, including "can we get them all?" "can THEY get them all?" and "we think we can get them all, but what do we do if we are behind schedule to maximize points?" It shouldn't be obvious either way for as many teams as possible.

My favorite kind of decision to leave teams is do we need that pathetic 10-point control at the far end of the map in order to win? Will Joe's team get it? If we stretch to get it will we be late and miss the 20-pointer near the hash house? I hate making those decisions when competing, and I love forcing them as a course setter :)

Even though I'm not a top contender in a well attended rogaine, I like to plan a route to get them all knowing that it is an unlikely scenario. I include in the planning process decision points where a certain subset can get dropped if behind schedule. I find this preferable to planning an abbreviated course because if you discover you're actually ahead of your plan mid-way around, you can't feasibly go back and get the ones you already opted out of. So I plan to get them all, then gradually modify the plan to stay on track as the event plays out.

And its nice to know that if all the things that can go wrong don't - that you could conceivably sweep the course - that's a mighty fine motivator at 3 in the morning...

Eddie and JJ, the scenario of the top team getting all the controls just before they run out of time is very exciting and forces them into solving the same types of problems that the midpack teams are facing. But on a 24-hour course where teams choose their own sequence of controls, where they do different parts of the course in day or night, where weather and course conditions can change before and throughout the event... no course designer can realistically "plan" it so a team will get all controls with 2-3 minutes to spare after 24 hours. Even the winning time of a fixed course middle distance race is often 2 minutes different from what the course setter thought it would be. Planning for a finish "to the wire" is how we get those rogaines where 6 teams get all the controls.

The course setter just missed. Its just like any other course setting, you aim for the target time and sometimes you are right on and sometimes over or under a little bit. It happens all the time in regular orienteering course setting. I don't see why you should compromise the course setting just to avoid going under. Putting out an excess of controls makes the rogaine problem less interesting and easier for the competitors. You should have to think about both how many points you might need, and how fast you and the others might be going, despite the weather. Choosing what to do at what times of day and night is just as much a part of the problem as any of the other aspects. Account for it in your course setting (we do). Yes, expecting to hit the target within a few minutes in a 24-hour event is unrealistic, but hitting the target within a couple of hours or a handful of controls is doable.

Accounting for the weather, course conditions and day/night is a huge part of planning how to *race* a rogaine course. For a course setter to anticipate exactly what the weather and trail conditions are going to be like on race day or what surprising choices teams might make on a 200 km2 course with 60 controls would require a better crystal ball than anyone has. There are some great rogaine course designers out there and I'm sure some of them are on this thread. But I don't believe they can purposely design a course where the top team will finish with 2-3 minutes to spare. Even a 1% error is 15 minutes plus or minus. And because policy C2 says that no team should get all the controls, the course designer is safer to err on the side of making it a little longer.

Regardless, the vast majority of participants - including the winners of most categories - will aim to visit an optimized subset of controls. Thus, that is what the sport of rogaining means to most racers. It is a tough challenge to design a course that is equally interesting for people who can go 40 km vs. 70 km vs. 110 km in 24 hours. Doing this well matters more than whatever happens to the top one or two teams.

Policy C2 makes for dull course setting. Its taking away one more variable, which makes the problem-solving less interesting. As you add more controls you reduce the dispersion in relative value between controls, making the skip decision easier. You could use fewer controls over a larger area of course. That too changes the strategy and not all areas can support this without severely compromising the navigational challenge. Also if you are running 6, 12 and 24 hr events on the same course low control density will make the shorter courses very dull indeed. Is there a difference between 6 teams getting all the controls in 23:30 vs those same 6 each missing 1 control? Sure, 1% is optimistic. How about 10-15%? I think this is easily doable. 6 teams finishing at 20:00 is obviously under-set.

I guess the key word from this discussion is "opinion" and I respect yours as I hope you respect mine. :) I'm a 3-time North American rogaining champion even though I'm not an elite orienteer who could ever hope to clean a course. Speaking of opinions, the world adventure racing champions who won the 2004 WRC in Arizona absolutely loved it even though they only got about 75-80% of the points. For them, it was a cool and different experience to win an event where they did not complete the full course; I spoke with one of them about it. So even top racers have different opinions about what constitutes the most exciting rogaine challenges.

I'm a 1-time North American rog champ myself (open, 2000), but I've been concentrating on shorter distances since then. I may start rogaining again someday, but rogaines are really hard and mostly unpleasant :).

I've talked to a number of people about this issue before and I've found that people who are primarily competitors prefer the excess number of controls, while people who are primarily setters (like myself) prefer the "just enough" approach. I suppose this is natural.

The winners of 6WRC got a lot more than 80% of both checkpoints and points. (The results are still reachable online, by the way, but require some creative Google querying; other WRCs fare less well). 3490 points out of I believe less than 4000, 55 controls out of I believe 60. I can dig up the map by this evening if nobody beats me to it.

I think there were 4000 points available for 64 controls. I'll see if I can find the exact number of points available.

PG's website (better than Wikipedia) says 4020 points. So 87% of points and 9 missed controls out of 64 for the winners.

So 86% of the checkpoints, 87% of points. That's pretty darn close to my stated guideline of 90%.

I'd rather see the top team get 90% of controls than the top 6 teams get all the controls. That's all I'm saying. (As you and I already agreed above.) It got trickier when people started talking about planning a course within seconds or a few minutes.

There's nothing tricky about it. If your target is 90% and you hit 87% you've done well. If your target was 100% and you hit 87% you've done less well. I'm simply arguing for a target of 100% instead of 90%. You're comparing the example of hitting the target (however defined) with perfect accuracy to the definition of the target itself. That was not the point.

There is a range of targets that corresponds to various conditions on the course and various issues that may happen to the top team. That is, there is a spread of outcomes. My position is that the top of the range (median outcome plus two sigma) should be 100%, which somewhat agrees with what Eddie's saying. Setting the median plus two sigma at 100% would put the median somewhere around 90% in most practical circumstances. Therefore, if you plan for median at 90%, two sigma at 100%, and you hit 87%, that's not bad. I'd say it's better than planning for median plus two sigma at 90% and hitting 87%. The latter would mean that the top team had a near-perfect race (which in this particular 2004 case, they certainly didn't).

I think the disagreement boils down to whether it is the median that should be at around 90%, or the median plus two or so sigma. I say the former, Bash the latter. Eddie needs to be a bit more specific.

I'm saying the median should be at 100% +/- 2 sigma. That is, I'm not bothered if teams sweep and come in early, provided its not more than 2 sigma too early.

So, we actually have three different opinions.

Three quotes from the manual mentioned above...
1.1.6 The Course IRF Championship Standard
"Ideally the course should be so long that it is unlikely that any team will be
capable of visiting all the checkpoints with time to spare. the latest
Australian standard is more specific - see Appendix 3 Example Rogaining
Map."

6.2.2 Length of the Course
"Ideally your course should be (more than) large enough to keep the best competitors occupied for the duration of the event."

and Appendix ARA TECHNICAL STANDARDS
Technical Standards for Organizers
"2 The course shall be designed so that the winning team is likely to visit most but not all checkpoints."

Indeed, and all of the three different opinions we have satisfy these conditions. (Eddie's opinion may be a bit outside the last one but fits the first two.) We are having an argument about further specificity.

Sorry Vlad, I incorrectly stated my opinion. I would want the *winner* to be 100%. So yeah, median of all teams at 90%. I apologize. Ignore my previous post. I'm not bothered if a 3-sigma team shows up and comes in early.

We are having an argument about further specificity.

Sounds like the first rule of Attackpoint, right?

Its hard to think in terms of the median for my club's rogaines, because there are so few teams (maybe 20) and such a wide spread in ability. I'd guess the WRC has a tighter distro. The handful of top teams should be at 100%, which may or may not be 2-sigma above the median.

I was trying to distance away from the distribution of scores for different teams. This distribution will be produced by both a spread of skills and a spread of possible conditions for any given team (weather; bilsters; twisted ankles; hyponatremia; lack of sleep two nights before because of a family emergency; etc.) In my arguments, consider only one team. The apriori top team. This team will have an expected distribution of its results. My definition of the best rogaine course is that the distribution for this particular team would have its median expected performance plus two sigma—that is, the performance under conditions that are better than 97.8% of all possible imaginable sets of conditions—covering the whole rogaine course.

:) Ah, good. This is the median I was thinking of when I posted that first opinion, but then I wasn't sure if you were talking about the median of *all* teams or the median you just described. By this definition, I'd want the median at 100%, not median+2sig at 100%. So back to 3 opinions.

Yes, in my experience the sigma for top teams is something like 5% of the median. The distribution is certainly not normal since there are quit-earlys. But for the really, truly top teams I don't think the sigma can be even as large as 10%. Looking at my team's performance, when the team is North American top-competitive, I can never see how we could have got more than 110% of what we ended up getting (unless there was early quittage).

My preference is close enough to Tundra/Desert's that I'll leave it to the two of you to sort out, keeping GHOSLO's quotes in mind. I need to get back to training for my next race! :)