Yes, it's finally ready to rumble! In the sacred tradition of years past, there will be a weekly MLS prediction league right here on BigSoccer. But this year it's going to be better than ever. The scoring system has been automated so that feedback will be immediate and error-free. Furthermore, there's a new scoring format that makes it impossible to win with a "borg" that doesn't know anything about the teams playing. There's no registration required and you can taunt your rivals in threads that your grandchildren will be laughing at decades from now (Or until the next server crash, whichever comes first) Details to follow.
With thanks to all those who contributed advice, here is the core system: You wager an amount between 1 and 150 on each game. If your team wins, you win your wager (if your team draws or loses you neither gain nor lose.) For every goal greater than one that your team wins by, you gain a bonus of one quarter of your wager. You can only wager each amount once per season. (There are 150 games) FAQ to follow.
dred, Lookin' fwd to it! I'm curious how you set it up so that folks won't repeat the same point value. Knowing myself, I'd be like, "that game's worth 150," about forty times during the season. Is there coding for rounding to the nearest available point value, or does the player see what point values are available? Either way, it sounds very slick! Thanks again, OGx3
Great question. Right now, the code rounds predictions to the nearest unused value. I will also be publishing tables showing who's used which wagers. At this point, I'm guessing I'll do this maybe once a month or so but it comes down to what people demand. It's a big bulky post so I don't want to spam it too often.
MLS Prediction League FAQ *What are the rules? You wager an amount between 1 and 150 on each game. If your team wins, you win your wager (if your team draws or loses you neither gain nor lose.) For every goal greater than one that your team wins by, you gain a bonus of one quarter of your wager. You can only wager each amount once per season. (There are 150 games) *Who wins the weekly cup? The weeky cup winner is determined by a Performance rating that involves the amount you bet and the amount you won. Performance takes into account the fact that you should be right more often on your big bets than your little bets. If you bet a lot on a game and lose, someone who bet only a little on the game will get a better performance score. If you bet a lot and win, you will outperform someone who only bet a little. For stat nerds, the actual performance formula is Reward - (0.4*Risk + Risk*Risk/250). Don't worry too much about the details of the Performance formula, it's just a guideline. At the end of the season, your rank in Performance will be identical to your rank in Reward. *What is the Play of the Week? The play of the week is a new feature this year. The winner is the player who's performance on one game stood out the most from the crowd. In other words, if everybody else bet a lot on Dallas to lose and I pick them to win, I have a good shot at play of the week if they win. Similarly, if everyone thinks a game is tossup but I bet 150 on it, I'm in good shape for Play of the Week if the team wins, especially if they win big. *How do I post my bets? Just place the amount first, followed by the winning team, then the losing team. An example for Week #1 might be: 20 SJE DCU 15 MET CLB 141 NE LAG 33 Chi kcw 5 Col Dal *What team names can I use? The automatic scorer recognizes most reasonable names. ie, NE, NER, Revs, and Revolution will be picked up. "Smogbreathers", "Swamp Things", "Smurfs", "Water Thieves" and "Those Pansies in Yellow" will not. Any combination of upper and lower case is fine. *Any other posting issues to consider? Use as many spaces and tabs as you want on each line, but don't use special formatting or add extra junk on the same line as any of your picks. ie, don't do: 20 NE LAG 133 Col SJE -- Who am I kidding? 35 Chi Dal or 3.14159 KCW Clb *What if I reuse the same wager amount by mistake? Do I have to keep track of all wagers I've used? The automatic scorer remembers what bets you've used. If you reuse a bet, the scorer will pick the nearest bet still available. So you can avoid keeping track of what bets you've used if that's what you want to do. Tables of used bets will be posted during the season if requested, but not too often unless there's demand for it. *What if I start late or miss a week? This league is, by its nature, very forgiving of players who miss a week. For instance missing five games is equivalent to losing your bottom five remaining bets, which probably wouldn't have produced much value anyway. In fact the scorer would automatically remove your bottom five picks for you so that you don't make the mistake of using them later and running out of games to use your higher bets on. *Why can't I pick ties? You can in a sense. If you have no idea who's going to win, just bet a very low amount on that game. *What if I want to change one or more of my picks? Just like last year, add a new post with your new pick(s). Don't edit your old post. *Why are the rules different this year? Mainly because there was demand for a system that did a better job of rewarding players for knowing about the teams than for knowing about the system. *Any tips? Don't use up all your high bets early in the season. There may be some "iron-clad locks" later in the year and some of the early games may be unpredictable. Try to keep a balance between high and low bets, stay flexible!
The week #1 thread is up! I'm going to try to make this the general thread that has links to all the weekly threads.
Another question, dred, this one would go under the heading, "Any other posting issues to consider?": Some of the prediction posts have other content, on separate lines, than just the predictions. So, extra lines are OK, just so long as there is no junk on the lines that have picks? Would a line that read, "7 reasons why SJ is going to win" in a post that had predictions be problematic for the coding? Thanks! ~OGx3
Great question OG. The scorer ignores any lines that don't have exactly 3 words in them. Furthermore, it will ignore the line unless the first word is a number and the other two words represent teams that are playing that week. So the bottom line is you can add pretty much any junk you want in your posts, as long as your picks are on separate lines. (I even put extra junk in my own prediction. )
Ok, I'm proposing a change to the scoring system, so please voice objections here if you have any. The change would be: If your team draws, you win half your wager. The idea is that it's just as bad for your team to give up the losing goal as it is to give up the tying goal. An effect of this proposed change would be that two one-goal victories would be worth the same as one 3-goal victory and one draw (or one five-goal victory and one loss). In any event, this shouldn't effect how people bet on games since you're still arranging the 150 games in order of your confidence in your favored team. It just makes the reward system a little more accurate. I'll make this change official unless the concensus objects to it.
Dred, I think this is a bad idea. If I bet 150 on Chicago at Home to wallop Dallas in the expectation of collecting a huge payday, and Dallas ekes out a 0-0 draw, I win 75 points? Doesn't seem to accomplish your stated objective, in my opinion. I think your original system is excellent, and hope you stick with it. NoSix
Either way is fine by me..I just wish there was some way to bet for a draw, but in the tow proposed scoring systems, they both sound good to me.
Unlike tree, I actually like that you can't picks ties! This way you have to make a prediction as to which team is more likely to win (yes, even if you're convinced the game will end in a tie). Anyway, I have a (slight) preference for a smaller tie bonus. Nonetheless, it's your league dred and I'm grateful to anyone that goes trouble of running a free league. Therefore, I'm quite comfortable with whatever you decide.
But if you bet 150 and "win" only 75, you are in a sense losing 75 points because most of your competition will be getting at least 150 on their 150 point bets. If you get only 75, your "performance" number will actually go down. Performance is the rating (in the FAQ) that determines whether you are doing better or worse than expected. If you bet 150 and only draw, your chances of winning the weekly cup are close to nil.
AGREED. I like the tie gets half rule. Definitely different from any other competition, but a good change.
You are missing the point. I understand that if you take a big risk (bet 150) and are right, then you will get a big reward. But in a true "high-rollers" league, if you take a big risk (bet 150) on a win and are wrong (the game ends in a draw), then I think you should pay a big penalty (get 0, not 75). If you are worried about not distinguishing between draws and losses, then assigning Asian handicaps to each match would be the proper way to eliminate that distraction, but that would be way too complicated for a game that is just for fun.
I support dred on this one. Getting half of your wager for a draw makes the most sense to me. "High rollers" notwithstanding, this is a prediction competition rather than a gambling competition, and there should be some meaningful distinction between a draw and a loss. For example, in the last week of the season I might be absolutely convinced that Columbus and Colorado are going to draw. It might also be the case that my lowest remaining wager is 112. The rational thing for me to do would be to bet 112 on one team for that game and hope my belief in a draw is wrong. Under NoSix's interpretation, if there is a draw I should get nothing at all, even though (subjectively) my prediction was correct. That makes no sense to me. At least under dred's system I get a 56 point consolation prize.
the draws equal 1/2 the wager is a good system. i support it. there are some valid points in arguing against it, but overall i don't mind not being able to pick a draw in the matches but being given 1/2 of my wager if the draw occurs. if this scoring change is adopted, i'll support it, and i'll start to wonder if wagering big or small is the best technique for games that i believe could result in draws.
Maybe that's the real reason why I don't like this idea - I'm too lazy to want to think about the implications
The goal of this format is to let players play without having to worry about the scoring system. If, for instance, you got 1000 times your wager for a victory, with no goal differential bonus, but lost 50% of your wager for a loss or a tie, that wouldn't change my picks for any of the games. I would still pick the highest numbers on games where I thought there was the biggest gap between the teams, and the lowest numbers on the games where I thought the teams were most evenly matched.
If that is your goal, the orginal system is simpler. But the order of the games by "most likely to win" is different from that by "least likely to lose", so by introducing a bonus for draws you are creating at least the theoretical possibility that an advantage could be gained by changing your betting strategy.
It turns out that last season in MLS, on average, a tie was very close to half as likely a victory by the favorite, so as a practical matter your proposed change wouldn't make a draw a desirable result relative to a victory. Objection dropped. Answer me one question, though. Your scoring system seems devised to make goal difference more important than wins, draws, or losses, but if that's your objective, why not go all the way, and make a (0 GD) draw worth 0, a 1-goal loss worth -1X, a 2-goal loss worth -1.25X, etc.?
Definitely considered that. Only reasons against it were minor ones like keeping things simple, not beating on people when they're down, and avoiding negative scores. BTW, I'm not sure what you mean by GD being more important than wins, draws, and losses. The tying and winning goals are worth twice as much as extra goals precisely because winning is more important than routing the opponent. Quite often teams dominate their opponent but opt to preserve a one goal victory, so the scoring system should reflect that.