Discussion: The Blind Tangerine are up to speed and rolling!

The Blind Tangerine are updated, and all eligible race results for the past year are now included. That means it are up to speed and are rolling right along and are ready to officially kick off 2005. The grammar are slightly off and nobody knows why. Brian May and Erin Olafsen are the #1 Tangerines as 2004 draws to an end, and will top off things as 2005 rolls up. Additionally, the methodology are revealed (follow appropriate link). You can see it all at: http://www.geocities.com/Colosseum/Stadium/7418/ta...
Next updateable event will be the US Team Fundraiser in Alabama. Good luck to all!

We need a North American elite ranking system.

Mikell, thanks.

It is very nice to have something like this, have it be rolling, and not just have a single annual list come out in may of the following year. The Tangerine are great, and I don't even mind the grammar.

This is probably the wrong forum for it, but I always wondered why USOF used a feedback system as opposed to simply updating the ratings after each event. It seems strange that my rating for an event can be changed by someone else's results six months later.

The US Chess Federation uses a static rating system where, after a provisional period (at the end of which a "real" rating is assigned), a player's rating changes with each tournament based on the points scored and opposition faced at that tournament. Then it's done with. That's your new rating going forward.

The system isn't perfect. Every once in a while some bozo like me will come out of "retirement" and all the kids hope they get to play my rusty ass and get some free rating points. But, overall it works well and it's easy to understand (if no easier to compute). I don't ever recall meeting an active player and thinking their rating was off by more than 100 points (which translates to about 5 points on the USOF rating scale).

Go Tangerine!

I really like the continously available rolling aspect of the AttackPoint rankings list. However, as it's based on user-entered splits, it, as expected, suffers somewhat from user-bias, as (it appears to me) users in general are more likely to post their good splits and less likely to post their bad splits. I tend to post almost all of my splits, but then I don't get credit in the AttackPoint rankings for beating people who are too depressed to post their splits :) Given the (important) user-input aspect of AttackPoint, I understand this won't change, but I still like seeing all of my splits online. So now I just aim to beat people at _every_ race, to try to end up above them in the AttackPoint rankings :) And if competitor X only enters their 5 best races, that's fine. I like the splits comparison more than the AttackPoint rankings, so I'm happy to see _any_ of their splits posted, and, the AttackPoint ranking are, as they should be, quite unofficial.

From the other side, the USOF rankings are better in this aspect, because they enter all the results, and are thus largely free of the aforementioned issue. However, the fact that they only come out once a year (well, twice counting the pre-Team Trials) doesn't allow them to be nearly as continuous a motivating factor.

And now, we have the Tangerine. Thank you Swampfox.

As for North American Rankings, I suspect that Swampfox has done a bit of typing for the Tangerine, and he might be willing to share that data with someone who wants to add some Canadian results to the list, and make their own NA rankings? Any takers?

Also, Nik Weber (current USOF rankings master) is cranking through the 2004 USOF rankings. He's already done with drafts of the Red & Blue courses, and thus he could probably provide some nicely formatted results (and spreadsheet with the rankings processing built in) to anyone who wants to create rolling M/F21 North American rankings...

I hope that The Blind Tangerine will motivate to train more and run hard. Although it is very different from USOF scheme and provides different set of results - it is interesting to look at the results from this perspective. I would probably make one correction and give same 5% boost for all USA Champs (Classic, Middle, and Long), Canadain Champs, and NA Champs. Otherwise, it is as good as USOF or WRE schemas.

BTW, nothing prevents to make USOF ranking rolling. A computerized ranking system is long overdue.

Well, having looked at the methodology, J-J thinks that more than just the grammar might be a little off, and has forwarded his comments to the Swampfox.

Note that the USOF ranking system is computerized, and has been for years. The main reason it isn't updated more frequently is because of the real-world difficulties of getting the results data. A-meet officials are supposed to send results to the Rankings honcho in a specified format, in a timely manner, but it's my understanding that few do.

It would certainly be valid as Sergey suggests to give some other important events some extra weighting. And if someone else were doing this and came up with a different list of events to be weighted, I would have zero problems with it! I just chose to keep things simple by picking out the 2 US events which, almost without exception, year in and year out, have attracted very competitive fields. As a practical matter and as the numbers of US Championships have grown (with another--a sprint--surely to come), it's just not very many people that are able to make all of them in a single year, given that the championships are spread both geographically and across the calendar. But it's not too much to expect most folks who want to, to at least make the Classic Champs and the Team Trials.

me don't see night O as a ranked event. there goes all that specific training I have been doing. guess me better try chess.

Following up on some of Wyatt's comments, I must admit I find the nonchalance with which results are made available to the public after many US a-meets to be a bit frustrating. I know that volunteer staffs are stretched thin; everyone is exhausted, etc. - after all most of us have been involved in multiple A-meets in various capacities. It is a lot of work.

I just think that the hard work would be leveraged that much more by having results available on the web in a timely fashion. Most A meets have awards so the results are typically calculated by Sunday afternoon for this purpose - why not do a little more to make them public? It is certainly frustrating for the Orienteering fans (such as me) who want to know what happened. With most other sports in the US this is not a problem.

In terms of making this possible - well, I need to put in a plug for DVOA here. Our web master Kent Shaw has put together an interface that makes it very each for people to enter their results directly into the club rankings database - making everything seamless. This is done for the smallest local meets.

I just wish this would become a priority for "A-meets" which are supposed to represent the zenith of technical and administrative implementation. There is enough technical competence in USOF that someone could implement a DVOA-like system for the purposes of A-meets. I suppose this could get done if it was made a rule (unlikely) or if every club did it voluntarily (also unlikely). To me, timely results reporting should be a requirement for sanctioning, but perhaps I'm just unrealistic.

Joe, the US Night Champs were included, and will be included in the future as well. The Blind Tangerine suggest you continue doing some specific training, and come out to Colorado and defend your title next summer. The terrain at Lake Manitou should be really, really fun for Night O'. But you better be ready to run! There will surely be some other fast moving citrine fruit there who relish the thought of night challenges.

The most critical hurdle for an orienteering rankings system is that it must re-create the ranking order that everyone in the sport already expects. If it fails to do that, then it probably has an important flaw. The pecking order is pretty obvious from the results of the o-meets we all attend. On the otehr hand, if it's fun, go for it.

This is different from chess. In chess, competition is head-to-head, often against someone you don't know and have never heard of. If a chessplayer is rated 1300, say, (which is fairly low) and makes an unusual move, it is probably safe to conclude it is a mistake. If a player is rated 2300 (strong master) and makes the same move, it's probably a trick and you may be in trouble. Knowing the rating of the person you are playing is therefore very helpful.