Quote:
Originally Posted by MisterJ
What may be easier than a draft-style fantasy website is a "salary cap"-style website. It would also allow for as many players for each regional as wanted to play.
Users could sign up for a specific regional, then would be presented with a list of teams attending that regional. Each team would be given a "dollar value" based on their performance in previous years or previous regional, and possibly by current-season OPR. (Perennial regional champs may cost $60 while rookies may cost $10 or $20.) Each user then must pick a team of 3 robots that costs under $100. An unlimited number of users may pick the same robots.
Just a thought. Might be interesting.
|
That particular item would an interesting one to try, however, it is not going to work for Season Long. Commissioner's choice, I can't go against that. (Partly because if it's unrestricted as to how many players can have the same teams like that, I can guess how, say, Waterloo will go for every player: 1114, 2056, 188 (or 610)).
If such a website were built, ideally it would be able to work with all the types of drafts that are currently done:
1) Draft-style, 3 and 4 rounds, with multiple tiers (Season Long). Before Season Long, individual events ran this one. If only one can be done, do this one.
2) Free-for-all, pick 3 but not the same alliance as someone else.
3) Auction (which is similar to the proposed "Cap" system), where each players gets X units to build the best alliance possible.
Now, the draft-style system is actually somewhat automated already, or can be. There is a program that many of the folks who run drafts use. It takes in a list of lists, a list of players, and a list of teams (picked up from the FIRST site) and spits out a random list, the tiers and their draft times, and as a draft progresses it reads the lists and puts the proper team in. I can give the original algorithm if needed; I can tell you who to ask for the source code of the current version. I could probably automate the waiver system, but I'm lazy. I suspect some of the scoring has been semi-automated in the past; there's no way the scores came out that quickly without automation.
If I were you, I'd start with the scoring data scrapes. Use the event codes to find them quickly. Current scoring is essentially QS+elim wins for matches, seeding points, where a team ends up in the tournament, and awards won. Flag any missing teams in award slots (Dean's List and WFFA are known for not getting the teams in place). Then I'd adapt the drafting code that currently exists (and there are a couple of things I'd like to see added, including automated randoms). Implement trades next; players are always changing teams out and in for any reason they like; make sure to include trades between players in the same tier. Then throw in the missing/added team alerts for the events (I've done that in the past, and it's a real time-eater to do it manually).
Oh, and make sure that everything can be edited manually if the user has the proper permissions, just to be on the safe side.
If it sounds like a challenge, it is. Brandon Martus couldn't finish it, but I think he had work get in the way last time he tried.