Shadows of Nar Eurbrikka — Avenger (Imperial Headquarters)

[goatman461]

So for the betting competition does it not make the most sense to look at the past 3-4 episodes and average each player's position to see their most likely predictive outcome? I know past results don't mean that is where they will end up in the future, but this seems like the most logical path. We would then have to make individual determinations for any ties and new players.

I think that is a good place to start. Any volunteers to do the math?

Will we be trusting the Rebels not to read this forum?

[GallardoLU]

I think that is a good place to start. Any volunteers to do the math?

Will we be trusting the Rebels not to read this forum?

trust them to not read here, no we cant do that, in fact we can be assured at one Bothan Spy will find his way here the bigger question is are we as a group going to sit down and agree to not "spy" on the rebels? I for one vote no spying (un-enforceable i'm aware, but at least put it out there that we intend to play as fairly as possible)

[MstrOfPppts]

Will we be trusting the Rebels not to read this forum?

NO! I guess BEAVeR can send a "SPAM PM" to everyone since I figure most of us has left the previous group PM ...

[Kodan Black]

I think that is a good place to start. Any volunteers to do the math?

I can compile an Excel spreadsheet of the past 4 episodes and the Rebel positions. Once we do that we'll have to identify any new players and determine where they fit.

[Kodan Black]

Ok, so I went ahead and did Episodes 2-7. 2 episodes had very small participation (3 entries for Ep 5 and 5 entries for Ep 7). So some of the datapoints aren't super accurate because people entered 1 contest and got a certain ranking, but I also looked at consistency (how many episodes entered) in order to come up with the rankings. Not sure if I should post it publicly or if we should do something else. Let me know what the general consensus is.

Edit: Did Ep 2 also to improve data.

Edited March 9, 2015 by Kodan Black

[Kodan Black]

Ok, I just re-read the goals and realized we predict our own faction too. I only compiled the Rebel data. I'll have to add in the Empire data also and update the sheet.

Ok, Empire data compiled. I'm not sure how we want to handle participation as it influences results. Obviously it makes the data point less reliable as we have builders that have a first place finish and an 11th place finish. So some of the top Empire builders are ones that have few Episodes they've been in, but they also happen to be talented builders.

But let me know what you all want to do about the data. I've got it in Excel right now but I also have it in Google docs in viewable format.

Edited March 10, 2015 by Kodan Black

[VK-318]

Ok, I just re-read the goals and realized we predict our own faction too. I only compiled the Rebel data. I'll have to add in the Empire data also and update the sheet.

Ok, Empire data compiled. I'm not sure how we want to handle participation as it influences results. Obviously it makes the data point less reliable as we have builders that have a first place finish and an 11th place finish. So some of the top Empire builders are ones that have few Episodes they've been in, but they also happen to be talented builders.

But let me know what you all want to do about the data. I've got it in Excel right now but I also have it in Google docs in viewable format.

I think our best solution to the problem of participation is to use the PM to get a commitment from people to participate or not. We can't share that here in the Avenger thread, but if we can get people to make a specific commitment to build or not build, we can make those statistics much more meaningful.

[Kodan Black]

I think our best solution to the problem of participation is to use the PM to get a commitment from people to participate or not. We can't share that here in the Avenger thread, but if we can get people to make a specific commitment to build or not build, we can make those statistics much more meaningful.

Sorry, I meant past participation. For example: the top 3 ranked Empire builders have 2 past episodes, 2 past episodes, and 1 past episode respectively. So #4 having all 6 episodes I reviewed participated in makes their standing stronger as the average is from more sources.

I wonder if we need to come up with a "bias factor" to alter the standings with something like <50% participation having a penalty.

Edited March 10, 2015 by Kodan Black

[GallardoLU]

if you do add a bias factor, you may need to take into account when that person first arrived in SoNE, for example I arrived for EP6, but missed EP7, so my participation is at exactly 50% but if going back to EP2, its only 14%. so I think a good measure would be to go from first entry each person has had, and then ad a factor based on total entries (as an adder to our rankings amounting to at most 10% of the ranking criteria. (scored from 1-7 would probably suffice, 1 point per episode entered)

[James Wellington]

We need another way to communicate these "bids" and all, because if we put them here, the Rebels are most likely going to see them. I mean, all you have to do is see someone's posted a new comment, and they can click on it from the topics, and not realize that they just spied, and saw our bids.

[Kodan Black]

Interestingly when I added a bias factor it actually didn't really alter the standings. I did .1 deduction for every episode participated in beyond 1. It had no impact. At .15 deduction 1 player moved 1 spot.

I'm not sure if I can do compounding additions in Excel, I'll have to Google it. But what may help is to do .1 for the first beyond 1, .15 for the 2nd, etc.

The one thing to keep in mind with them seeing our bids (which they will be able to do eventually anyway since we have to post it) is that it still requires them to build something and have it judged. So if we put Player X as first place, in order for them to impact that they need to have Player X do worse than someone else per the judges decision. So if they game the system too much they'll lose on points for the main competition and they would need to correctly predict how the judges would rank them anyway.

[Kodan Black]

Ok, wow. Even compounding the number has virtually no impact on the rankings. Given that, I would suggest we simply use the straight rankings unless there are individual tweaks we feel are needed.

[goatman461]

Thanks for doing all of this work Kodan. Can you PM me a link to your Google doc?

[Tariq j]

Same here, I'd like to see a link to the google document, thanks for doing all this hard work.

[Kodan Black]

I'm working with Goatman to try and get the sheet sharing properly, then I can pass the link around.

[Kodan Black]

It will be interesting to see if our hours of work come to fruition. We've compiled the rankings based on past episodes, so aside from new faces we have a pretty good notion of a mathematically projected order.

[GallardoLU]

It will be interesting to see if our hours of work come to fruition. We've compiled the rankings based on past episodes, so aside from new faces we have a pretty good notion of a mathematically projected order.

this is such an Imperialistic approach to things.

[VK-318]

this is such an Imperialistic approach to things.

Yes, now that you mention it. Honestly, I think this whole effort will best even the funniest of stories told this episode.

[Kuat Drive Yards]

Hey guys! So, here's my dilemma: I really want to participate in this awesome challenge, but my LEGO selection is pretty much limited to gray and white with the odd plant. Do y'all have any advice as to I can make good dioramas without being able to make the scenery y'all have?

[goatman461]

Hey guys! So, here's my dilemma: I really want to participate in this awesome challenge, but my LEGO selection is pretty much limited to gray and white with the odd plant. Do y'all have any advice as to I can make good dioramas without being able to make the scenery y'all have?

I think the best way to stretch your LEGO is to build ships. Building architecture requires you to use the same part over and over but a ship is much less likely to repeat parts. That being said, pod racers are a bit trickier because of their symmetry.

Do you have a flickr account or anywhere that we can see your past builds? That might help us suggest something that plays to your strengths.

[Kodan Black]

I'm fairly in the same boat, my supplies of Tatooine colored parts basically fits in my hand. I'm going to attempt a very small build as a "yep he is on Tatooine" shot, and then do some interior "warehouse" type of builds. You could certainly do a shot of a bar or the like as there is no requirement that it be tan colored, bars can have all kinds of interiors!

So you could do a shot of a bar interior setting up meeting the pod racer and then if you wanted to do your engine delivery you could do that in the racer's warehouse/garage. Ideally the pod racer isn't all gray (though bare metal is that color), so hopefully you have enough color variation to at least give it some flair. You also don't have to build the entire pod racer, you could have it be in parts as it is being built; hence the need for the engine!

[Kuat Drive Yards]

Some of my creations are on MOC pages (under Mark Simmons), and I want to start uploading to this is site.

[GallardoLU]

Some of my creations are on MOC pages (under Mark Simmons), and I want to start uploading to this is site.

Feel free to post MOCs in the main star wars group, but image hosting will need to be done somewhere else, many users us Flickr (me included) and then link images for posts here on Eurobricks. (also Nice work on the Old Republic era ships, can't wait to see your newer builds!)

[Kuat Drive Yards]

Yeah, I was lazy at MOCpages so you'll have a lot of shiny new ships once I get home.

So we can't just upload pictures off our computer?

[VK-318]

Yeah, I was lazy at MOCpages so you'll have a lot of shiny new ships once I get home.

So we can't just upload pictures off our computer?

Correct. In order, presumably, to keep down overhead, Eurobricks.com keeps a tight limit on images that can be uploaded. Flickr is your best bet as an image hosting site - it's what 99% of Eurobricks users use, and the Eurobricks image-linking tutorial includes special, specific instructions on the easiest way to link Flickr pictures here. A few people (like me) use Brickshelf to host images because we already have profiles there and are (at least in my case) too lazy to learn a new system. It's possible that MOCpages would work as a hosting site for you - I'm not qualified to judge that since I now zip about MOCpages.