SoNE feedback thread

[Ninja Nin]

You got the point MstrOfPppts !

In terms of overall quality, Empire always win, so it is obviously the ( number of participants X 2 ) which makes such huge difference.

I suggest building up the two factors, Quality and Quantity, as 50% 50%.

Example:

Episode III

Average Building Score:

R: (22+21+19+17+16+6+5+3+3+3+3)/11 = 10.7272

E: (23+21+19+15+14+6+5+5+2)/9 = 12.2222

Average Score Percentage (Quality)

R: 10.72 / 25 = 0.43

E: 12.22 / 25 = 0.48

Total number of Participants - according to the player index

R : 39 and E : 37

Participants Percentage (Quantity)

R : 11 / 39 = 0.28

E : 9 / 37 = 0.24

Total Score (Quality + Quantity)

R : 0.43 + 0.28 = 0.71

E : 0.48 + 0.24 = 0.72

EMPIRE wins !

Based on this calculation, the impact from participation should become less.

Episode I

R : 0.41 + 0.41 = 0.82

E : 0.42 + 0.32 = 0.74

REBEL wins !

*from MstrOfPppts last post, the quality factor isn't making a hugh difference in this case, therefore quantity factor becomes dominant

Episode II

R : 0.42 + 0.28 = 0.70

E : 0.54 + 0.13 = 0.67

REBEL wins !

*again, the participants from Rebel side is more than double than the Empire side.

Sorry if the calculation looks too complicated.

But I think the current situation is to consider how to encourage people to build, thus not neglecting any builders with their score, hence reducing the impact due to more participants or so called "rushed entries", so balancing both quality and quantity.

Now the question is, this calculation includes all player from the index, who signed up but is idle / disappeared / banned.

Any thoughts ?

[Brickdoctor]

I don't think the scoring should take into account all the members of a faction. It brings up the issue of determining who counts as active and who doesn't, and it will also force members to consider a commitment to regular building when signing up. From my experience with Heroica, a lot of members when signing up were concerned about whether or not the inability to continuously participate might disqualify them or negatively affect their character (beyond just the loss due to inactivity of potential stat gains)

[Kaitan]

That's not a bad idea actually. It is also possible to level the playing field further by limiting the number of entries counted for the total score to the number of participants in the faction with the lowest number of entries; For example in Ep 3, 9 Imperial entries, 11 Rebel entries -> Only the first 9 rebel entries are counted.

That's a great idea! It strikes me as the only way to make things totally fair.

[BEAVeR]

Using the average of all the scores of a faction only to determine who wins, would not be totally fair.

Because the less builders you have, the greater the percentage of people who make it to the second round of judging, thus gaining more points. The faction with the fewest builders will have a higher average, assuming both factions have equal quality of builds.

MstrOfPppts' idea of just taking the average could work however, if not a fixed number of builders would make it to the second round of judging, but when the best percentage of each faction would.

I'll give the example of episode II, where there were only 5 Imperial entries. The average of the Imps was 13.6, compared to the 10.6 of the Rebs. I don't think this is an adequate representation fo the quality of the builds in that round.

I can't show any calculations to clarify what I mean (because it would require new scoring of the builds), but I hope you see what I mean. I only would like to add that this method only takes the quality of the builds into consideration, and not the quantity. Also it is likely that the judges would have more work...

So I think the best method would be to keep the system we have today, use the method of determining the average I've described above, and maintain a smaller weight for the number of builds.

[LegoFjotten]

As said before, we've always beaten Rebels in quality and here's the proof:

episode I:

R: (24+23+22+20+19+7+6+6+6+6+6+5+5+5+4+3)/16 = 10.4375

E: (23+20+20+19+18+5+5+4+4+3+3+2)/12 = 10.5 (huge win!)

That math does not hold up. Since the first 5 scores of each faction are scored differently, you cannot sum the scores and divide by number of participants to get an average. As soon as you have more than 5 entries, each new entry will decrease the average. What you think is a higher quality of builds for the Empire is just another flaw in the scoring system.

To get something closer to a mathematically correct average, it is possible to scale back the first 5 entries by dividing them by 2.5 (so all scores are 1-10 points), and then divide by the number of participants, which gives (for Ep I):

R: 9.6+9.2+8.8+8+7.6+7+6+6+6+6+6+5+5+5+4+3 = 102.2/16 = 6.3875

E: 9.2+8+8+7.6+7.2+5+5+4+4+3+3+2 = 66/12 = 5.5

But this doesn't really hold up either, so I think average score per builder is just no good as a measurement as long as two scoring systems are used.

Overall, I think the episodes have been much much closer than the scores would indicate, and I really don't care which team wins as long as we see some great mocs and have fun building!

Edit: Partially Ninja'd by Beaver, this is exactly what he is talking about.

Edited February 3, 2014 by LegoFjotten

[MstrOfPppts]

Ah LegoFjotten, now you're nitpicking!

But I agree and I've said it way back, that the quality should not be calculated simply as an average. At least not from results that are on a different scales like the top 5 entries which should first be normalized just as LegoFjotten suggested.

[MKJoshA]

I think there are some really good ideas out here right now. I like MstrOfPppts ideas about quality and I think Ninja is on to something with her new scoring ideas. I agree though that the number of signed-up verses active participants should not be a factor. However it's done, I think the most important aspect to work on right now is how to count quality over quantity.

[Bob De Quatre]

I'm in favor of deciding the winning team only by the scores of the entries that made it to the second stage of judging, and don't take into account the number of participant.

[Brickdoctor]

It is also possible to level the playing field further by limiting the number of entries counted for the total score to the number of participants in the faction with the lowest number of entries; For example in Ep 3, 9 Imperial entries, 11 Rebel entries -> Only the first 9 rebel entries are counted.

I really like this idea. (Either this one or the one with only the top five scores.) It's like the Risk battle system, my top two dies against your two dies, and we know that that works.

[MstrOfPppts]

After some thought I think that limiting it to the number of participants in the faction with lower entries might also lead to an autowin. Presuming that top builds are somewhat equal, the team with more builds would then drop out the ones with the lowest scores, meaning their average would not be lowered so much as the team's with less entries. Maybe going with the top 5 is still the fairest idea here, though this way I think that the others might feel left out ...

[LegoFjotten]

You are right MstrOfPppts, dropping the bottom entries would still not be fair.

Maybe top 5 is the most fair option after all.

[Artizan]

how about giving the top five entries a coefficent higher than 1 and the rest of the builds a coefficient of 1, than dividing the results respectively?

[BEAVeR]

Just a thought I had some minutes ago on a matter different from how to score for each episode:

how about a total score? Just a plain simple sum of the XP of all the players for each faction would give us more insight how this game is progressing on a more general note. Also, it would make freebuilds more relevant, in comparison to the present situation, in which tags, control of planets etc. are only about the episodes. I don't have any particular awards in mind for the faction that's leading, but at least it would give us something to contribute to, a general sense of where we stand.

[korpen4444]

Can our episode builds grant us XP? Oh, and when is the next episode coming?

[Brickdoctor]

Can our episode builds grant us XP?

Yes, they do; I just haven't updated the player index yet.

[korpen4444]

Yes, they do; I just haven't updated the player index yet.

Oh, so did I earn any? Because I wasn't aware.

[LegoFjotten]

Oh, so did I earn any? Because I wasn't aware.

Yes, results are in the episode thread: http://www.eurobricks.com/forum/index.php?showtopic=89320&st=25

[Pyrovisionary]

Rather late post, and I know I haven't been active here since episode 1, but I'll say what I think would draw more players in and would certainly guarantee my activity once I get back to my bricks.

- Perhaps a third faction? I dunno, bounty hunters or smugglers or something. Anyway, I think adding a third, neutral faction would make this better. I mean not everyone wants to be a rebel or Imperial soldier.

- Add more character stuff. Perhaps add a tree so that everyone starts of as a grunt but then after gaining significant Xp could become let's say a sniper or a sergeant or a pilot . Character progression, but with more depth. Even perhaps give high Xp characters certain bonuses.

- Give us players agency! Perhaps give players lives and if they lose that many challenges their character's die. Or get seriously wounded and cannot participate in the next challenge. If their high XP character gets wounded they have to make a new low Xp character just for the next challenge or so...

Edited February 15, 2014 by Pyrovisionary

[Brickdoctor]

- Give us players agency! Perhaps give players lives and if they lose that many challenges their character's die. Or get seriously wounded and cannot participate in the next challenge. If their high XP character gets wounded they have to make a new low Xp character just for the next challenge or so...

That's an interesting idea. How would we determine whether a player gets wounded? Score relative to the average score? Pitting players specifically against each other doesn't seem like it would work.

[Pyrovisionary]

That's an interesting idea. How would we determine whether a player gets wounded? Score relative to the average score? Pitting players specifically against each other doesn't seem like it would work.

Perhaps a certain chance. Something like a 1/10 chance on the losing side and a 1/20 chance on the winning side.

I still also think that you should add character trees (i.e a trooper gets 10xp and can then become a pilot or a squad leader, these will give certain small bonuses. A pilot with 20xp could become a wing commander or a bomber then). The main effect of these trees is that each role is given a more specific objective. So, using Episode 3 as an example:

Rebel Objectives:

Troopers:

Recover equipment and armaments

Destroy covert information left behind

Pilots:

Prevent imperial forces from landing on Hoth

Provide air support to Land forces

Evacuate land forces

Bombers:

Prevent imperial forces from landing on Hoth

Provide air support to Land forces

Evacuate land forces

Bomb imperial Land forces

Air-drop supplies to rebel troops

Squad leaders:

Recover equipment and armaments

Destroy covert information left behind

Rig base with explosive in burned earth tactics

Guard base from imperial forces in the previous battle's fortifications

Snipers:

Recover equipment and armaments

Destroy covert information left behind

Rig base with explosive in burned earth tactics

Guard base from imperial forces in the previous battle's fortifications

Pin down imperial forces

Assassinate key officers

These are rather poorly done examples but still....

I feel that these more complex rankings and objectives would make player characters feel more unique. Another step could be to give these rankings small different bonuses, which could range from different weapons and ships to more men under their control, to small Xp boosts.

Edited February 15, 2014 by Pyrovisionary

[Brickdoctor]

So add more 'classes'? That could work, as long as it's still based on building and affects building first and foremost (which it is in your example) — I don't want to add too many RPG elements that would take away from the fact that this is a building game.

Should we take a vote in regards to scoring? Top n Entries vs. Simple Sum of All Entries?

[Kaitan]

Perhaps a certain chance. Something like a 1/10 chance on the losing side and a 1/20 chance on the winning side.

I still also think that you should add character trees (i.e a trooper gets 10xp and can then become a pilot or a squad leader, these will give certain small bonuses. A pilot with 20xp could become a wing commander or a bomber then). The main effect of these trees is that each role is given a more specific objective. So, using Episode 3 as an example:

Rebel Objectives:

*snip*

Leaving ANYTHING to a random chance will never been seen as fair. Excluding a player because of a bad dice roll? They might never come back to the game.

I agree that the rank system could do with a bit of tweaking - purely a cosmetic thing, but just assigning actual military/navel ranks instead of "sections" to levels would make things easier to know where you stand. E.g. if another player decides to include your character in their entry, if they outrank you (actual military rank, not just more xp) then it makes sense that they can give you orders etc. It will also help players like myself know when they can start bossing non-player characters around

I think making things too complicated with regards to objectives will limit creativity and creates a load more work for the admins -who are doing loads already!

[Pyrovisionary]

So add more 'classes'? That could work, as long as it's still based on building and affects building first and foremost (which it is in your example) — I don't want to add too many RPG elements that would take away from the fact that this is a building game.

Like I said, it would not only add more variety in the builds, but everyone's character, for the most part, would be quite unique. It also gives a feeling of character progression, and once you get that next rank, it just adds to the sense of achievement.

Should we take a vote in regards to scoring? Top n Entries vs. Simple Sum of All Entries?

Definitely the former: Top n Entries

Leaving ANYTHING to a random chance will never been seen as fair. Excluding a player because of a bad dice roll? They might never come back to the game.

You are right but I can't think of anything else.

I agree that the rank system could do with a bit of tweaking - purely a cosmetic thing, but just assigning actual military/navel ranks instead of "sections" to levels would make things easier to know where you stand. E.g. if another player decides to include your character in their entry, if they outrank you (actual military rank, not just more xp) then it makes sense that they can give you orders etc. It will also help players like myself know when they can start bossing non-player characters around

I don't quite understand you.

I think making things too complicated with regards to objectives will limit creativity and creates a load more work for the admins -who are doing loads already!

That is isn't completely true. It would only limit creativity as much as it already does, it gives unique roles to separate players which is a great thing because it makes the player feel special, unique, one-of-a-kind etc. And increases the amount of objectives, therefore increasing the guidelines in which the build can take place. And seeing as you are aloud to do more things with your build creativity can be increased. But yes, it does create more work for the mods.

Edited February 15, 2014 by Pyrovisionary

[Kaitan]

I don't quite understand you.

At the moment the xp ranks are as follows -

• The possible ranks (and their required XP amounts) are:
  
  • Empire - Stormtrooper/Tie Pilot (0 XP) > Imperial Survey Corps (10 XP) > Imperial Security Bureau (25 XP) > Storm Commandos/Shadowtroopers/Dark Troopers (50 XP) > Imperial Department of Military Research (100 XP)
    
  • Rebellion - Rebel Trooper/Pilot (0 XP) > Alliance Support Services (10 XP) > Alliance Intelligence (25 XP) > SpecForce/SpecOps (50 XP) > Starfighter or Army Command (100 XP)
    

These descriptions are more like different sections/departments within the faction, rather than ranks. I'm just saying that it makes more sense to have actual military 'ranks' e.g. Lieutenant, Commander, Captain etc etc - then leave the department up to the player.

That is isn't completely true. It would only limit creativity as much as it already does, it gives unique roles to separate players which is a great thing because it makes the player feel special, unique, one-of-a-kind etc. And increases the amount of objectives, therefore increasing the guidelines in which the build can take place. And seeing as you are aloud to do more things with your build creativity can be increased. But yes, it does create more work for the mods.

Saying to a beginner "Your objective is to blow this", whilst saying to a more experienced player, "You have to blow up this, get that, take out that member of the other faction AND paint the whole planet pink" seems restrictive to the beginner and just overwhelming to the more experienced .

I know thats an extreme example, but the more general objectives, along with perhaps the occasional building restriction (e.g. size 16x16 plate etc), make for a more relaxed and enjoyable game in my opinion. Plus it makes judging a lot easier if everyone works to the same goals.

Should we take a vote in regards to scoring? Top n Entries vs. Simple Sum of All Entries?

Sorry missed this :) Yes we need to vote ! Top n Entries for me

[Pyrovisionary]

Saying to a beginner "Your objective is to blow this", whilst saying to a more experienced player, "You have to blow up this, get that, take out that member of the other faction AND paint the whole planet pink" seems restrictive to the beginner and just overwhelming to the more experienced .

I know thats an extreme example, but the more general objectives, along with perhaps the occasional building restriction (e.g. size 16x16 plate etc), make for a more relaxed and enjoyable game in my opinion. Plus it makes judging a lot easier if everyone works to the same goals.

Simple, make objectives different. In it's most simplistic form make a trooper "Do this" And then a general "Do that"

As opposed to a general having to: "Do this and that"

On judging terms, I was under the impression that builds were judged on quality, or have I missed something? Anyway, there doesn't have to be rank specific objectives. I mean there don't even have to be more complex ranks, this is a feedback thread not a make-the-game thread....

Edited February 15, 2014 by Pyrovisionary