by Ultraviolent4 and Paul Selick
Posted: September 9, 2018
What follows is an email exchange between Paul Selick and me about the idea of a International Crawl Association.
With Paul's permission, I'm publishing this with the hope of inspiring those open to inspiration.
From Paul to UV4:
Love your channel and web site.
I am writing to present you with my idea for an International Crawl
If you think it is a good idea, I thought that you and your friends might have enough influence in the Crawl community to make it a reality.
The major sports/games have their world associations: the recently completed World Cup is organized by FIFA; the Olympics has the IOC; figure skating has its ISU; chess has FIDE; contract bridge has the World Bridge Federation.
I think Crawl has the intrinsic qualities to be worthy of such an organization. So I propose and World Crawl Association, governed by an elected board of directors, whose mandate would be to spread the growth of Crawl.
Note: I do not suggest an organization encompassing rogue-likes as a whole as most do not have the quality of the game Crawl and spreading the recruitment effort too thin will dilute it.
Below I give some of the things I envision such an organization might do.
In some cases I give my idea of how it might be implemented, but the exact details are not of concern.
1) Award titles of Grandmaster and Master for elite players good enough to satisfy the stringent criteria decided upon by the board.
2) Organize a World Championship.
Yermak has the best record in recent tournaments, but there are many (no
doubt including yourself) who would welcome the opportunity to take him on
in a match.
Existence of official titles of Grandmaster and World Champion might even rekindle the interest of some legendary players who rarely play now. This championship is intended to be a more serious event than the fun tournaments coming with each new version.
The organization should have a YouTube channel and record the games of the match on the channel for the fans to watch.
The organization would have a "Competition Committee" consisting of grandmasters who would decide the details of the match conditions.
There could also be a "15 runes World Champ"; a "turncount World Champ"; and a speedrunning World Champ (establishing rules regarding what is or is not allowed in terms of macros, etc. is likely to be a challenge for the Competition Committee, but better that some official organization should make such decisions than the present situation where there is no consensus on exactly what is permitted in a speedrun).
My idea: A match would be a specified number of games, say 25.
For this, 25 race/background combos would be decided upon and each player would play the same combos: one game per combo with a point for a win resulting in a score out of 25.
Each player would have a generous amount of time (say 3 or 4 months) to play the games so that no one would be penalized by having outside commitments during a specified time, as sometimes happens during the 2.5 week tournaments for the new versions.
The combos could be chosen by some combination of: random; selection by the Competition Committee; HORSE (i.e. each player gets to select some of the combos for both to play).
3) A rating system.
While the above may interest the best players, the rank and file will not be
in contention for the prestigious master/grandmaster titles.
But in most competitive games they like to see a numerical rating system, similar to the ELO rating system in Chess, through which they can measure their improvement.
My idea: A player shall have the option before starting a game to declare
it to be a "rated game".
The Competition Committee assigns difficulty levels to each combo.
Winning a rated game causes the player's rating to go up; losing causes it to go down with the amount of gain/loss calibrated according to some scale based on the difficulty of the combo.
I suggest that the not all games need be rated so as not to discourage experimenting or just playing for fun when not feeling in the mood to take it too seriously.
4) Instructional Articles
The organization could have a web site including instructional articles.
One of the difficulties for new players is determining what weight to assign to the sometimes ill-advised suggestions on the wiki or the contradictory viewpoints offered on the tavern.
It would be useful to have a place where articles have been approved by an Editorial Board and the reputation of the author is visible through the rating system.
5) A Hall of Fame
As in many of the major sports, each year some legendary figures would be
voted into the Hall of Fame recognizing their lifetime Crawl achievements.
As in other Halls of Fame, I suppose that there would be different categories to which someone might be elected: player; developer; author; promoter; ...
These are a sample of what the Organization might achieve.
I hope you like the idea.
From UV4 to Paul:
Thank you for your email. I do like your ideas and they really got me thinking. I sort of floated them around to some other members of the Crawl community as well and, if nothing else, you've got people talking and imagining.
I'm not sure there's a will to run an official World Crawl Association. The people who would need to be involved are either going to be like me (trying to use their time to play the game) or developers who are trying to use their time to develop the game. Speaking for myself, I struggle to find enough time to record my YouTube videos and any extra time I have, I mostly put into answering noob questions or trying to write guides.
For some of your suggestions, I do already aspire to achieve them in an unofficial capacity or I think they could probably be set up with some initial work to run mostly automatically in the future. I'll cover each in turn:
1) I think the most straightforward way to assign titles to players would be related to a rating system so I'll cover this in 4).
2) There are currently talks going on among the devs to reform the whole tournament system so that it doesn't encourage playing as many hours per day as possible for 16 days straight to place highly.
Your tournament idea sounds to me much like the "Crawl Sudden Death Challenge" tournament. Here's a Tavern thread from an old season: https://crawl.develz.org/tavern/viewtopic.php?f=23&t=17860
There are concrete plans happening right now to resurrect CSDC. I've never played it before but maybe it could form the basis of the sort of thing you're thinking of.
If I were to express my ultimate wish, I would love for there to be a way to generate seeded games of Crawl (https://en.wikipedia.org/wiki/Map_seed). I think people have undertaken the task before but it's a huge undertaking because the whole game needs to be remade from the bottom up. Seeded runs would not only be a fantastic learning tool (as a newer player you could make attempts at a map and then see how a "pro" approached that exact same run) but it would allow for more direct competition. With seeds you could have realtime or turncount "races" with fundamentally the same dungeon and they wouldn't be decided just because one player was luckier and found better items, generated faster branches, etc.
3) This idea of a rating system is extremely interesting to me. I've had some preliminary talks with some other players and I think this is something that could be introduced with a bit of work. At present, the only basis to compare players is with winrate but not every player plays every game to win and some players undertake difficult challenges such a low-turncount wins. It's possible to look at longest winstreak but some people are content to win 40 + Deep Dwarf Fighter games in a row. With a rating system in place, you could easily compare players and assign ranks like "master" or "grand master" in a somewhat objective way.
The way I imagine it working is that you examine all tournaments in history to get an idea of the relative difficulties of species and backgrounds. Things like Minotaur and Berserker are won the most frequently so they are probably the easiest whereas things like Mummy are won the least so they're probably more difficult. You then assign each player a starting ranking such as 1500. At the beginning of each online game, players would be given the option of whether they want the game to be ranked or not. If it's ranked, the result would contribute to their streaks and rating. If you win the ranked game, your rating goes up and if you lose, your rating goes down depending on your current rating and how difficult the combination played was. For instance, if you have a high rating and you play an easy combination such as MiBe, you might only get +2 points for winning but -50 points for losing because the system expects someone of your calibre to succeed. Perhaps you would only be allowed to win each combination once to prevent players from grinding out 200 MiBe games in a row.
If nothing else, I really think Crawl should have a means of marking games as being serious or not. As things stand, if you're in the middle of a winstreak, any other losing game you play on the same account will break your streak. It would be nice to be able to play non-serious games that didn't count in streaks - I wouldn't have had to play in the tournament on a different account because Ultraviolent4 had an active 5-game winstreak, for example. This would also allow players to keep track of a meaningful winrate without having to maintain separate, winrate accounts.
4) Your suggestion is essentially what I'm trying to do with my website by having experienced players contribute well-written guides. I've been doing my best to get more players to contribute. Some have agreed to do so but then discovered that writing a guide is no easy task!
5) This is another idea I really like. There was a Reddit series of writeups called DCSS Classics which I enjoyed immensely. Here's one: https://www.reddit.com/r/roguelikes/comments/499ysw/dcss_classics_2_rob_the_almost_pacifist/
I probably wouldn't call it a Hall of Fame but the thought of amassing the stories of famous players and devs and showing what they've done/contributed to the game is one which appeals to me. I'm going to pencil it in as a long-term project I'd like to work on. In fact, I've just discussed the idea with gammafunk and he thinks the knowledge more or less exists out there. I'd like to sit down with mikee (a Crawl legend) to see what his thoughts on this project are.
So thank you once again for your thoughts and suggestions. While I doubt that an official body will ever exist for Crawl, you've certainly succeeded in getting some minds turning about what might be possible for the future. You're more than welcome to send me any new ideas or suggestions you come up with.