Box office Contest

[Jedi Master Carr]

I am going to start this mostly to work out plans to do it starting the first weekend in May. So first off should we change the way we play the contest? I say that because it seems like we had a huge drop in players the last few years and maybe a change in rules would help. I think Yoghurt had some ideas. So I will let him suggest them. If anybody else has some ideas on changes and such, just post them here.

[CMJ]

Well part of the decrease in players was the Non-SW year factor.

[Jedi Master Carr]

I think it does better in the summers too, it seems like people have no time in the Holiday period. I might not do it again this year during Christmas because of that fact. People do have more time in the Summer I hope that is part of the reason so few did it last time. We had more in the Summer as I remember. Also I am still thinking about tinkering with the rules. Yoghurt suggest using Mojo's rules but I was hoping he would post what they were.

[Zasz Grimm]

if so, it should start up again relatively soon. with all the comic book hero-esque movies. I've never played, and am now interested.

[Jedi Master Carr]

Well it will start first weekend of May, right now I am not sure with the Rules, I want Yoghurt to explain the Mojo rules for me other wise we are using the rules we had done in the past. Also there will be a prize which will be 6 months free supporter account. There might be a second prize in there we will have to see how many compete.

[Jedi Master Carr]

Here were the rules for the holiday contest, I am going to adjust them somewhat to show what they would look like fo the summer if we use them

Period of contest: 7th of May - 6th of September

1. How to play: Each weekend, predict the order of the top 5 movies, and how much each film will make. To make sure people don’t make the same predictions, post your dollar figures to the nearest ten thousand, for example "25.58M" (25 million 580 thousand).

2. Movie Order Points: For every film you place in the correct order, you get 5 points. If you get #1- 5 right, you would get 25 points.

3. Accuracy Points: The closest person to each of the five films also gets points, and they are as follows;

# 1 - 25 pts
# 2 - 20 pts
# 3 - 15 pts
# 4 - 10 pts
# 5 - 5 pts

Please notice; you only get points for a movie if you have it in the right #1-5 position.

4. Double Points Weekends. Points are doubled on the following weekends:
* Memorial Day; 28 May - 31st of May. All points are doubled.
* Harry Potter weekend; 4th-6th June. Double points for Harry Potter.
* Fourth of July weekend 30th June to 4th of July
*Labor Day Weekend 3rd of September to 6th of September

5. "The Price is Right" - bonus for extreme accuracy. If someone guess a movie correct down to the nearest 10-thousandth, the contestant will be awarded with 50 extra points, in addition to the points he won that weekend. Example: If a movie makes 25.343M and you predict 25.34M, you will win the bonus.

6. Deadline: All predictions must be in by saturday 9 am EST every weekend. On double points weekends, special deadline will apply. Please make sure your predictions are in before the deadline, however, we may look the other way if the estimates are not yet made public and/or it is obvious such results have not been seen.

7. Who can play? Everyone. It is fun! You may enter or exit the contest at any time or post predictions whenever you like, as long it is within the deadline. Of course, be aware maximum participation increases your chances of a high score at the final tally, so it is recommended to participate as often as possible (it makes your predicting look smarter! ).

8. The winner: The player with the most points at the end of contest, wins instant fame and glory (being worshipped by the all the BO forum posters) AND the winner of the contest gets his/her choice of a 6 month free "Suporter" account (6 month extension if already a supporter) or a @sw-fans.net email address.


Now we can stick with these rules or change them completely or make small adjustments. Like we could change how you get points for the top 5. In the past if you got the top film wrong you get nothing like you picked Walking Tall this weekend and Hellboy came in first. We could still give you some points for being close to that movie although that might be tough to do. It would probably be better to go to Mojo's rules if we did that because it be easier I think

Here are there rules

Derby winners are determined by their overall prediction accuracy, expressed as a percentage. Prediction accuracy for a given movie is calculated as follows: 1 - (Difference between your Prediction and Actual Box Office Gross / Actual Box Office Gross).

Example:
You predict SOME MOVIE will gross 21.5 million for the weekend. This movie actually grosses 20 million. So, we take the difference between your prediction and the actual gross and divide it by the actual gross: 1.5 / 20.0. The result, 0.075, represents the percentage your prediction was off from the actual (7.5% off). Subtracting this number from 1 (one represents being wholly accurate) tells us how close or accurate your prediction is to the actual: 1 - 0.075 = 0.925 or 92.5% accurate. Note that if your prediction is off by more than the actual gross (i.e., off by more than 100%), your accuracy will be 0% representing the fact that you were simply not accurate in your prediction (e.g., you predict 10 million and the actual is 3 million).

DETERMINING THE CHAMPIONS
Box Office Mojo will announce the results of each week's contest, as well as keep a running total of contestants over a period of time. See the Derby Champions page.

Weekly Derby Champions will be determined by their average predicting accuracy for that week.

Overall Derby Champions will be determined on a points-based system. Each week you play the contest, your average predicting accuracy will translate into points and will accumulate from week-to-week. The more you play the more points you will accumulate, so be sure to play often!

Example:
Your average predicting accuracy last week was 75.22%; you are awarded 75.22 points. This week your average is 92.50%, so you now have a total of 167.72 points.

I think it be easier on you guys but tougher on us. But I could try my best to do it if everybody wanted it.

[Master Yoghurt]

Ok, sorry for lagging behind on this :)

Anyway, what I suggest is a rule where the points are calculated entirely based on accuracy similar to how Box Office Mojos contest rules. Of course, you are probably wondering why bother with that?

Why new rules?

Its because the old contest system, made it possible to have a very good, or even the best guess on what a movie will make, and still end up with 0 points (wrong movieorder). And even if you do get the movies in the right order, you could end up with few points, because someone was lucky, and beat your estimate with 10K, thus your opponent got all the points. Thats a system based more on luck than on prediction skill IMO.

Ideally, those who make the best guesses should also make the most points. If you make a good guess, you should score lots of points. The score should gradually reflect the quality of your prediction.

So what is suggested here, is a points system based on percentage accuracy. Your guess on a movie compared to the actual result will allow us to calculate your percentage accuracy. Each week, the players % accuracy are tallied up, and we can make a list of what everyone scored, players of the week etc. Each player will have a trackrecord of his/her best results updated every week, until the contest ends and the results are tallied up to determine the winner.

* This allows you to post exactly what you honestly think the movies will make, rather than tactically picking a number with a different range of the other players to score some easy points. This also makes the tactic of posting guesses in the last minute fairly pointless *cough-cough*

* One difference from the Box Office Mojo system, our new system allows for casual players to participate and still have a chance to win (you can go away on holiday and return without having lagged behind in the race). Mathematically, its highly unlikely a winner can be determined until the last week of contest, thus making this exciting to the very end. More probably, there will be multiple possible winners at the last week of contest. With a rule of say 5 weeks (just an example) minimum participation, your 5 best results will be tallied up to a final score. Yet, those who were dedicated enough to play for 16 weeks, will be rewarded as they have a better chance to have 5 really good results among those 16 weeks.

* Such system makes it easy for those who dont know much about Box Office to play. They could pick numbers close to what the BO regulars posted and probably do fairly well. Of course, you would be unlikely to actually win that way, because if everyone just picks the same numbers, no one will gain an advantage towards each other in good results/points. The smart player chose the right moment to pick numbers wildly different to what the other players say, and scores big because he/she predicted a bomb or a surprise hit.

* Since there will be less reason for post prediction tactics, and more intencive to post what you think the movies really make, it encourages box office discussion, and posts like "haha, Spiderman II will never make 120M on opening weekend, you guys are nuts" etc etc. Could be amusing.. :lol I expect the contest threads to be just as much about discussing why a movie do well or not, as as posting the contest predictions

* If we use the same percentage calculating system as Box Office Mojo, it will be easy for people to check up their own scores. To calculate scores, we use this tool:
http://www.boxofficefactor.com/calculate/

Anyway, youre probably tired of reading my long winding post, now for the actual rules. I will try to make it simple..

[Master Yoghurt]

Rules SWFANS Box Office Contest Summer 2004 (this is a draft and subject to change)

Period of contest: 7th of May - 6th of September

1. How to play: Each weekend, there will be 10 movies for you to predict the weekend gross. Predict how much each film will make. To make sure people don’t make the same predictions, post your dollar figures to the nearest ten thousand, for example "25.58M" (25 million 580 thousand).

2. Points: Your points will be calculated based on your average percentage accuracy of your predictions compared to the actual weekend results. The percentage will be be turned into points. 75% overall accuracy makes 75 points that week. If you like to see how the percentage is calculated, or verify that your points are correct, try this tool:
http://www.boxofficefactor.com/calculate/

The formula is as follows: 1 - (Difference between your Prediction and Actual Box Office Gross / Actual Box Office Gross)

3. Weekend winners. Each week, we will post the results and percentage accuracy of every player. The player with the best result, will be determined weekend winner. There will also be a tally of the season totals updated every week.

4. Season winner. The 5 best weekend results are added up for every player into a final season score. The player with the highest season score can call him/herself SWFans Summer Box Office Champion 2004 and wins instant fame and glory (being worshipped by the all the BO forum posters) + the choice of a 6 month free "Supporter" account (6 month extension if already a supporter) or a @sw-fans.net email address.

5. Deadline: All predictions must be in by saturday 9 am EST every weekend. Certain weekends, a special deadline may apply. Please make sure your predictions are in before the deadline, however, we may look the other way if the estimates are not yet made public and/or it is obvious such results have not been seen.

6. Who can play? Anyone can play. You may enter or exit the contest at any time or post predictions whenever you like, as long it is within the deadline. However, maximum participation increases your chances of a high score at the final tally, so it is recommended to participate as often as possible. In order to contest for the Box office Champion title (and the prize), you need to participate for a minimum of 5 weekends. Failing that, you could still boast a nice season accuracy or be a weekend champion.. :)

[Master Yoghurt]

And here is an example of the calculating of points work in practice..

Lets say I predicted the movies for last weekend, and I predicted the following:

The Ladykillers - 10.23
Passion Of The Christ - 11.23
Dawn of the dead - 8.23
Jersey Girl - 7.23
Taking lives - 5.23
Starsky & Hutch - 5.23
Eternal Sunshine of the Spotless Mind - 5.23
Secret window - 5.23
Hidalgo - 5.23
Never Die Alone - 3.23
Agent Code Banks 2 - 2.23

My calculation of accuracy compared to the actual numbers would look like this:

Scooby-Doo 2: Monsters Unleashed - (25.23 / 29.40) -- Accuracy: 85.82%
The Ladykillers - (10.23 / 12.60) -- Accuracy: 81.19%
Passion Of The Christ - (11.23 / 12.60) -- Accuracy: 89.13%
Dawn of the dead - (8.23 / 10.70) -- Accuracy: 76.92%
Jersey Girl - (7.23 / 8.30) -- Accuracy: 87.11%
Taking lives - (5.23 / 6.60) -- Accuracy: 79.24%
Starsky & Hutch - (5.23 / 6.20) -- Accuracy: 84.35%
Eternal Sunshine of the Spotless Mind - (5.23 / 5.30) -- Accuracy: 98.68%
Secret window - (5.23 / 5.30) -- Accuracy: 98.68%
Hidalgo - (5.23 / 5.30) -- Accuracy: 98.68%
Never Die Alone - (3.23 / 3.10) -- Accuracy: 95.81%
Agent Code Banks 2 - (2.23 / 2.80) -- Accuracy: 79.64%

OVERALL ACTUAL ACCURACY: 87.94%

New Release: 87.48%
Still in Theater: 88.17%


************
************

The interesting number here is 87.94, which is my score for the weekend. That score happens to be a very good result, so its a good chance it would go in to my top 5 results for the season.

Now for arguments sake, lets say my top 5 results for the season are

Weekend A: 87.94
Weekend B: 93.21
Weekend C: 89.15
Weekend D: 88.49
Weekend E: 90.32

My final score for the season is:
87.94 + 93.21 + 89.15 + 88.49 + 90.32 = 449.11 points

Making an average accuracy of 89.82%

[Madmartigan]

Sounds good except for the person who has to do all the score keeping.

[Jedi Master Carr]

Yeah me and Yoghurt would be the ones to have to go through that mess :p I think we could do it, maybe be something different.

[Master Yoghurt]

The score keeping seems daunting at first, but with that calculator tool, it should not be that terrible. You just put the numbers in, click "calculate" and you can copy and paste the results. For season tally updates, you just use a normal calculator. :)

[Jedi Master Carr]

Yeah thats true, it sounds easier with it, I think the only thing extra we would have do is keep copies of the numbers so we can tally up the best weekend numbers.

[Jedi Master Carr]

Two questions, first, we pick the top 10 right? Also regardless of wether the movie made it in the top 10 you still get a percentage right?

[Master Yoghurt]

Yeah thats true, it sounds easier with it, I think the only thing extra we would have do is keep copies of the numbers so we can tally up the best weekend numbers.

Yeah, we would need to keep track of the players best weeks (thats actually the important thing to keep track of). A weekend scorepad would look something like this:

1. Player A - 93.74%
2. Player B - 89.82%
3. Player C - 87.34%

Where player A is the winner of the weekend.

Season Scorepad would look something like this:

Position - Name of player - Avg accuracy - Total points
1. Player A 90.33 - 451.64
2. Player B 89.82 - 449.11
etc etc

and the best weeks table..

Best Weeks Summary:
1. Player A: 95.67 + 91.49 + 90.12 + 89.01 + 85.35
2. Player B: 93.21 + 90.32 + 89.15 + 88.49 + 87.94
etc etc

[Master Yoghurt]

Two questions, first, we pick the top 10 right? Also regardless of wether the movie made it in the top 10 you still get a percentage right?

Question #1: Yeah, we pick 10 movies the posters will have to predict weekend numbers for. However, in order to make the calculating of points a lot easier, we could just pick the ones BOM pick. Brandon Gray picks the movies he thinks will make the top 10 for the BOM derby, and those are usually more or less correct. Furthermore, any other movies than those are usually not worth predicting, so I say we just go with that list.

Question #2: Yup, whatever the position it makes. 8 or 12, you still get points for percentage. Position does not matter in this system.

[Madmartigan]

You could use possition as bonus points though. A way to reward the people who manage to get the order right. You could also award the winner of the weekend bonus points as well. Kind of hybrid of your system.

maybe something like
1st place on Order +5% on their total
2nd place on Order +4% on their total
3rd place on Order +3% on their total
4th place on Order +2% on their total
5th place on Order +1% on their total

Best percentage of the week +5%
2nd Best percentage of the week +4%
3rd Best percentage of the week +3%
4th Best percentage of the week +2%
5th Best percentage of the week +1%

that way you award people for indivisual weekly perfromances as well as participation ("half of like is showing up") plus then by adding in your percentages it also evens on the playing field. like I said a hybrid

[Jedi Master Carr]

I wouldn't mind doing that, it would be different than mojo and would be our own thing which is good.

[ReaperFett]

I have one worry. Wouldn't making people guess double the number make it so LESS would try?

[Jedi Master Carr]

I don't see that problem, I mean it would be more up for grabs and the contest would be up in the air until the very end.