BEST-mix.net - DDR Simfile Tournament

I wanted to make this topic FEEDBACK2BACK.

Anyway: I need feedback about the following!

1) Tournament in general, rules, schedule
2) Bonus Votes from round 1 scores (good/bad?)
3) Community Pack
4) Popular Votes
5) Anything else you can think of!!!


Be as honest and blunt as you'd like. I don't get offended unless you say you didn't have fun; in which case I get sad because I only run these tournaments to make it as fun as possible for everyone.

I do not foresee this being my last tournament, and I whenever I decide to end my run doing tournaments, I'll go out with a big splash (and I'll let you all know so we can throw a big party!). haha So let's make the next one even better (like always).

community pack was cool would do again

I think we need some way to score a challenge chart beyond just a extra, because I saw some files this time where the challenge was better than the heavy and it kinda sucks not to give them some more points for it. Just scoring them both and splitting the heavy points will scare people away from doing in, and just taking the better of the two basically gives you a mulligan. I was thinking something along the lines of 2/3 of the score of the better one and 1/3 of the score of the worse one.

I like the bonus vote idea. I wonder if we can't extend the tiebreaker idea further and go 1.75, 1.5, 0.25 and 0 bonus votes, although even with that it would be possible to tie across brackets.

I miss the wildcard. Even if it's limited to one or two files, I think we should have a wildcard round.

The community pack is good. I wouldn't mind seeing it as a "free division" with separate judging, or at least run by some of the judges for more feedback.

Community pack needs to happen more often. I had more fun making a sim for that.

-BM needs more exposure. Everything worked out smoothly, but I'd definitely like to see more people getting involved with this.

-Popular vote categories must be up from day 1 of the competition. I didn't had enough time to vote for every category fairly.

-Bring back a separate public choice, a standard poll where people freely chooses their favorites without any influence of the judges, just to see what people liked. Back in the tournamix days, it was fun too see how different (or how similar) judges votes and public votes were.

I think that's it.

I'd say that there needs to be some sort of "sway" category. Not everything means as much in every file. Sometimes graphics and vids add more to a particular file than others. Sometimes, a challenge and beginner set add more than a light. Sometimes, the quality of a standard set far outweighs the quality of a light step. I'd like to be able to reward that.

For me, BM3 was almost perfect. The only thing I'd have liked to have seen was if you did what you did for BM2 and advertised it on BMS and such. Bonus votes were a good idea. Some of the differences of .5 between positions in the final shows they are significant, and gives an incentive to place top in the brackets.

TheBaldOneMpls wrote:I'd say that there needs to be some sort of "sway" category. Not everything means as much in every file. Sometimes graphics and vids add more to a particular file than others. Sometimes, a challenge and beginner set add more than a light. Sometimes, the quality of a standard set far outweighs the quality of a light step. I'd like to be able to reward that.

Actually, make the Presentation part of the sway category, with a -5 to +5 thing.

I also believe that the contest went almost perfectly and that you shouldn't change things too much. Nonetheless, I still have to wonder whether what happened to Barbie Girl (5th to 12th, ouch) and Sonic Speed Riders (1st to 8th? really?) should happen again.

Another thing I didn't address during the contest because I felt it inappropriate to do so is the ruleset on the late entries (that appears to have been quickly improvised). Well, first of all let me state that you did it once, so people might anticipate you doing it again, which would be an entirely different situation from this time. The -2 late penalty would've knocked out any of the finalists, so in order to make the finals, a late file would have to get nearly perfect scores from almost everyone. If you end up doing it again, I'd suggest keeping the penalty so that people don't anticipate you allowing late entries again and unfairly make use of spare time to perfect their files. However, I would also make it very explicit that chances of making the finals late are very low, and that the main advantage of participating late would be to get feedback. If you don't, you may run into drama in the future (And to be clear, I would announce allowing late files in the same manner as you did, that is, as a surprise thing due to a lower number of entrants.)

The community pack was a great idea. I have a feeling that part of its success was the novelty of it, though, and that it might not work as well the next time for some reason. That doesn't mean you shouldn't do it again; you absolutely should. Furthermore, in this case too, I think the fact that it wasn't announced earlier was good: it kept interest in the contest later on, because people were making files again, or at least completing them. Make sure you don't make the deadline for the pack the same as the deadline for the main entries - that would probably be a mistake. I unfortunately disagree with jammitch; I think it's better as a simple pack without judge involvement.

I think we should leave challenge steps as is. You should do them because you want to as a fun extra. On a personal note, I have no interest in judging peoples' random keyboard/ITG steps. I'm not really qualified (:P) and beside that they were just fun extra steps in DDR much of the time anyway.

I'd say that there needs to be some sort of "sway" category. Not everything means as much in every file. Sometimes graphics and vids add more to a particular file than others. Sometimes, a challenge and beginner set add more than a light. Sometimes, the quality of a standard set far outweighs the quality of a light step. I'd like to be able to reward that.

In contests I've been a part of in the past, judges used sway pretty unfairly so I'm not sure I like "tilt" categories.

Juub005 wrote:I still have to wonder whether what happened to Barbie Girl (5th to 12th, ouch) and Sonic Speed Riders (1st to 8th? really?) should happen again.

Not sure how to fix that with still having judges...

The -2 late penalty would've knocked out any of the finalists

I would also make it very explicit that chances of making the finals late are very low

,
So in other words keep it at -2? or slightly less (1.5?)

Make sure you don't make the deadline for the pack the same as the deadline for the main entries - that would probably be a mistake.

Definitely! The two will not have same deadlines, ever!


And regarding sway, I've had it; not had it, and most of the time people have seemed not to like it, because it's been used by at least someone each time as a "I feel this song deserves a better score, even though parts of it really sucked" which isn't fair compared to good files that had to work on getting the points.
I'd like arguments for/against it, because I don't mind it; but it never seems to be used for non-personal reasons.

BLueSS wrote:

Juub005 wrote:I still have to wonder whether what happened to Barbie Girl (5th to 12th, ouch) and Sonic Speed Riders (1st to 8th? really?) should happen again.

Not sure how to fix that with still having judges...

WILDCARD ROUND

May I revert you back to the situation in BEST-mix #1? DJ Potatoe somehow gathered enough hate for SHE BLINDED ME WITH SCIENCE to haul it's total score down from 1st (totaled from the other 2 judges) to 12th in its bracket. Luckily enough, there was a Wildcard Round of approval voting, giving it the chance that if enough people really liked the file and wanted it to move on, it would advance. But even though it got 6 votes just like the last few files that made that cut-off for the next round, based on judges scores and keeping to top 15 only, it didn't advance. At least the file was given a chance, unlike files this time around like SONIC SPEED RACERS, SON OF SUN, etc.

I think that just like old OSC/BEST-mix times, we should really bring back the whole "top 3 files from each bracket automatically advance to the finals, with the rest of the top half of the files moving on the the next round" sort of thing, or with BEST-mix #1, "top half files get to be marked yellow and 'advance' even though the worst file in the bracket can still move on if it somehow recieves enough votes". It really prevents situations like SONIC SPEED RACERS dropping from 1st to 8th and not advancing from happening again, since based on old rules it finished right under half of the total files (8th of 15, 8th of 17 including late entries), which in that case it would still move on to "the next round" since 8 files from Bracket 1 would've moved on, so 8 should move on as well from Bracket 2.

Back to the other things, I like how the "judges scores count in the finals too" thing was taken in again, since after BEST-mix #2's 3-way tie for first definitely called for something like this. However, I do think it could be improved. Why are the differences in bonus votes at a constant .5 points? Take the difference between RELEASE ME and MY BOY based on Bracket 1 scores, for instance. Should a difference in just .01 point decide which file gets the extra boost in final votes? I mean, they're virtually the same score. A score change by just .25, the very least fraction of a point allowed to give out during the public review/judging, in one area out of the 2 areas in a public review with just Heavy and Presentation graded, in only one person's public review with just Heavy and Presentation graded would have put MY BOY on top of RELEASE ME by less than .01 of a point, thus giving MY BOY the .5 bonus vote rather than RELEASE ME. I honestly can't stress enough how ridiculous of a difference that is, and it goes to show how the .5 bonus vote system did not justify extremely close finishes.

I think next time, if this same system is used, bonus votes should be determined in a more mathematical scale, making it just a little more fair for everyone who gets that advantage. I think there should be a pool of, I don't know, let's say X amount of extra votes in the beginning, and to those who place better than others after the first round results, they won't only get a higher bonus vote than the person who finished behind them, but it will be either a bonus vote value that is comfortably much higher, or a value that is pretty much equal to the other file's bonus vote total.

To show how this would work, I'll take Bracket 1's results as an example. The difference between 1st place and 4th place was by only 1.26 points (based on the Scores.pdf Final Score totals, and does not contain actual point values that exist in the thousandth of a point and beyond). Here's how the differences would look like if MY BOY represented 0, being the last place finisher in the top 4, and Back 2 Back represented the difference of 1.26.

Back 2 Back: 23.72 - 22.46 = 1.26
COME BACK TO ME: 22.53 - 22.46 = 0.07
RELEASE ME: 22.47 - 22.46 = 0.01
MY BOY: 22.46 - 22.46 = 0.00

So if a pool of 3 bonus points was used (which it technically was (1.5 for first, plus 1 for second, plus .5, plus 0, totaling at 3 points)), here's how the bonus votes would be calculated.

BONUS VOTE POOL: 3
TOTAL DIFFERENCE: 1.26
BONUS POINT FOR EVERY .01 DIFFERENCE: 3 / 1.26 = 2.38095238 (~ 2.38)

Back 2 Back: 1.26 x 2.38 = 2.9988 (3) Bonus Votes
COME BACK TO ME: 0.07 x 2.38 = .1666 (.17) Bonus Votes
RELEASE ME: 0.01 x 2.38 = .0238 (.02) Bonus Votes
MY BOY: 0.00 x 2.38 = 0 Bonus Votes

Okay, so the bonus point for every .01 difference math was wrong (since I forget how you divvy up values using a number pool, but I bet Juub knows), but I guess this version would still be acceptable to use. Just cut down the total Bonus Votes for each file in half and you get 1.5 as the highest possible bonus value a file can get, just like it was in this contest. Except it wouldn't be a pool. It would represent the most possible points a file can recieve as an added bonus in the finals, and then taking the highest and lowest values and everything in between as a scale on how many points should be rewarded for each entry.

This may seem a little... time consuming and unnecessary, and after doing all that math, believe me, I feel the same way too :sweatdrop:, but if we're seriously going to determine differences in which files get bonus points or not based on first round scores that are microscopic in differences (even smaller than a .01 difference!! seriously?), then we really should use a system where the final scores are equally as accurate in total differences.

This whole shpiel is just my proposal of how things could be improved. It's feedback just like everyone else's feedback, but it's also a suggestion for the future. I'm not asking for this same exact system to be used in the next contest, but I am suggesting something similar to this as a consideration, where if bonus votes are given out, they should be given out fairly using a mathematical scale of some sort like the one I gave as an example.

And now I'm finished.

Math opinion:

If we're using a pool of 3 votes, perhaps we should normalize it so that a total of 3 bonus votes are given out per bracket, rather than awarding the top song 3 and filling in below that. To do that, we'd want to divide by (1.26 + 0.07 + 0.01) = 1.34. This means that the bonus factor is 2.23880597... and the bonuses are:

Back 2 Back: 2.820
COME BACK TO ME: 0.157
RELEASE ME: 0.024
MY BOY: 0

Due to rounding, we're slightly over, but I'm obviously not going to complain about this result :p

Each way has some pros and cons - with your way, each bracket will award a different total number of bonus votes, but with this alternative, a top file with no competition in its bracket would receive an advantage over a top file in a tight bracket, even if the quality is similar. Either way, nothing we do can really normalize scores between brackets.

Oooh, so that's how you would do that.

I just got stuck in all that math I was doing. D: But that makes a bit more sense now.

My only complaint is that the file size is too much of a challenge.

I would like to start seeing some really high quality MP3 without having to worry about file size. I personally want some 600kbps songs. I want to hear the full potential of my stereo system!

With an increased file size, this also means the videos could be HD without having to worry about file size. Every video in best-mix 3 I was troubled by the quality. {Even "My Boy"}

Other than that, the contest was fine. I would of liked it if the 3 judges were consistant & the 3 judges are not allowed to compete.

So in general, bigger file size, and consistant judges.

Chrisketchum1 wrote:My only complaint is that the file size is too much of a challenge.

I would like to start seeing some really high quality MP3 without having to worry about file size. I personally want some 600kbps songs. I want to hear the full potential of my stereo system!

Are you insane?
You are playing a dancing game. These aren't meant for high-quality listening pleasure. If you want that you can buy the songs you like or find a rip of it elsewhere. Besides, many of us don't even have that high quality for our own listening pleasure.

You also have to consider hard drive space. If we did higher quality music ON TOP OF HD video (which would only work well in SM4 by the way) each bracket could go upwards of 200-400mb each.

Also to the full extent of your stereo system? I have $70 headphones here with a killer bass and every song that is 192kbps or greater sounds perfectly fine.