AB, I share your enthusiasm but the competitive aspect of CC has blown out of proportion already. If we add #WUs as a criteria, I'm writing it here, there will be *many* donors out there who will rig a rig to fail WUs every second. I don't mind losing against them, but it will hurt the project, and drive PG (and all other who care about the project) mad, rightfully. I wish it was an ideal world where everything worked as they were designed to, but it's not...
 
I really like the idea of having multiple competition fronts and combining them. This may be much simpler, manageable and more fair than a formula.
 
I agree with Punchy that HPCS pretty much ruins all measures or factors that relate to "growth rate".

1. Taking a period of time where HPCS was not available is a dumb reference.
2. Taking a period of time where HPCS was available as reference may work but because of the uncertainty in HPCS's future availability will make this reference unacceptable for teams who are heavily using HPCS right now.

An idea would be randomizing the reference point for each team. Think about the last 30 days right before the competition start. For each team, independently, we can draw 3 days out of that 30, and take the average of those 3 days PPD as reference. This would rule out any tactics to manipulate the reference point, and it would make the HPCS factor not an issue. What do you think?
 
After that, we can have the competition run on three fronts for three weeks:

1. Raw CC-name points
2. Growth rate (a measure comparing CC-name points to the above reference)
3. Conversion (% of total team points done by the CC-name)

This way, there would essentially be nine competitions, and for each of them the first comer would receive 1 points, second runner 2 points, etc. At the end, the team with the least points wins.
 
I love this idea! No having to tune up a formula so that everything is put together at a desired effect, etc. It's simple and fair, as long as we agree on the three measures above.
 
So, what do you guys think? I'd like to hear from other teams as well.
 
 

Afterburner

^^ Little confused... Adding WU's is to much but adding two more measuring ideas is not?

Other than that I find your thinking interesting. I know I have seen this or similar thought before. Can you draw up an example of what it would look like?

 
I think you got me wrong. I didn't say adding WUs is too much (seriously, where?). I said, there will be people rigging their systems to increase the WU count, which will hurt FAH.
 

Afterburner
And why in Gods name would we get credit for a failed WU???

 
I really wish it wasn't the case, but WUs get counted by PG even if they fail, and there is no way to tell how many WUs were successfully completed. That's why WU count should not be used as a performance measure in CC.
 

Personally, I believe the discussion to be going in the wrong direction.  In the abstract many of the ideas being suggested are theoretically sound and will alleviate many of the complaints regarding the CC, however, they provide little to no direct feedback/benefit.  If there is one thing we've learned about folding on this team over the past 6 months is that direct feedback (typically in the form of points) is very important to some.  If, instead of chasing a direct measure such as CC points, their individual contribution is chasing an arbitrary improvement in a randomized drawing I believe that is going to be a hard sell. 
 
I could be wrong, but while I appreciate the theoretical beauty of the suggestions I don't get excited about such an abstract form of "competition".   I think a simple, theoretically sound, and as fair as possible system is the best bet, knowing full well going into it that no system is going to be perfect (but it will be hard to do worse than last year).  Heck, if it were perfect, we would have all teams tied for first place!  If people are interested in talking about improving the system similar to last year, I've got some thoughts and I've got some numbers.  If people prefer to go a different direction, however, that is OK too.

Well I tried. From your thoughts of this going in the wrong direction... I guess I will just pull off to the side and grab a cold drink and some chips... Because if it is anything like last year I am out. It was specifically designed to eliminate us. Period. And no. Not looking to rehash it. It is all gone. But if we cannot have a legitimate chance to compete this is all for nothing and it is time to start a different contest and move on... As long as WU's are completed as needed that is all that matters...

troy8d

Personally, I believe the discussion to be going in the wrong direction.  In the abstract many of the ideas being suggested are theoretically sound and will alleviate many of the complaints regarding the CC, however, they provide little to no direct feedback/benefit.  If there is one thing we've learned about folding on this team over the past 6 months is that direct feedback (typically in the form of points) is very important to some.  If, instead of chasing a direct measure such as CC points, their individual contribution is chasing an arbitrary improvement in a randomized drawing I believe that is going to be a hard sell. 

I could be wrong, but while I appreciate the theoretical beauty of the suggestions I don't get excited about such an abstract form of "competition".   I think a simple, theoretically sound, and as fair as possible system is the best bet, knowing full well going into it that no system is going to be perfect (but it will be hard to do worse than last year).  Heck, if it were perfect, we would have all teams tied for first place!  If people are interested in talking about improving the system similar to last year, I've got some thoughts and I've got some numbers.  If people prefer to go a different direction, however, that is OK too.

 
Troy...what new system/changes do you think offer the best chance for a "fair" competition? 

Afterburner

Well I tried. From your thoughts of this going in the wrong direction... I guess I will just pull off to the side and grab a cold drink and some chips... Because if it is anything like last year I am out. It was specifically designed to eliminate us. Period. And no. Not looking to rehash it. It is all gone. But if we cannot have a legitimate chance to compete this is all for nothing and it is time to start a different contest and move on... As long as WU's are completed as needed that is all that matters...

 
I'm not saying my way is better.  I know that realistically its far from perfect, perhaps much farther from perfect than the other ideas presented.  I observe that I am in the minority here, and the last thing I wanted to do was stifle the conversation. 
 
I suppose the thing I can add to the conversation is that we need to consider that the ultimate goal is to motivate people to action.  This is most easily accomplished by finding common ground with a measurable goal, which is the role of the contest.  Our common ground is Team EVGA.  The question we are trying to answer is the common goal, and in order for it to be meaningful it has to be simple enough for the majority of people with at 30 second attention span to easily grasp and the results have to be directly measurable.
 
Please don't bow out because I am thinking inside the box while you all are thinking outside the box.  (Also, I am exhausted over here so I could very well not be making any sense).

theGryphon

Hey Troy, did you see my suggestion above? I'm not suggesting randomizing to decide the winner, only for establishing a reference level for "before CC" points.
I'm still suggesting that raw points should be a measure, as well as growth rate (similar to handicap from last year), and conversion rate. What I'm suggesting is instead of dumping everything into an arbitrary formula, let's have a competition for each of these performance measures and reward teams on all three fronts. (You have to see that whatever formula you come up with, it will be arbitrary in the sense that it will include calculations you have no way to validate the effectiveness or fairness of. Any formula will bear questions like "why not multiply instead of add?" "why not take the square?", etc.)

 
The formula will not be arbitrary.  Zodac's formula is relatively arbitrary.  If the formula is well constructed it will be easy to answer "why not multiply instead of add?" or "why not take the square?" as there are distinct statistical implications behind the operations you suggest.  Don't get me wrong...I'm not saying it'll be close to perfect, but it will be a measurable improvement.  Is it better than the other systems being suggested?  That I cannot answer and do not want to derail or dissuade others from pursuing their own ideas.

I got a partial night's sleep last night and this morning and had the realization that what I am working on is by no means mutually exclusive with what others have been proposing.  It could very well be an improvement to a small part of the larger construct many are proposing.

the problem with that... is the last 4 months could be our best(peek and & contest) - and the last 4 could be the lowest months for another (off months - ie summer - and no contest)
 
if you look at the 9 teams some barely move their ppd over the year - some only peek during chimp
some may be seasonal...and some have jumped on HPCS- like OCN and posted their output greatly since it (and the previous months would be a lower their average)
 
the only way that works is taking best 2 week (of the year).
secondly this is almost the same formula that was used last year
which led to a small team winning(combining with another team - which was not considered in the initial data) and of course us in almost last place with the most PPD and most folding points
and it was pointed out smaller teams can convert more members to fold easier - and other like us cannot convert as many - because a good section of our team are folders from other teams here to collect EVGA buck then - may fold against us in the sontest (inflating our numbers)

 
no matter what system we use a team will figure a way to take advantage of the rules
 
which is why so many ideas are being floated.
 
mine takes out threat of cheating
the goal is to gain awareness of F@H  (PR) and growing all the teams and their membership, while trying to find cures
The goal should be very difficult to achieve and would require all teams to put in best effort to produce and grow their teams
 

Anyways they are getting twitchy on the CC forums wanting a formula soon
 
so which ideas do we take to the CC forum

So we get rid of the variables like conversion factor and EVGA bucks by not converting to new teams.

Ah well, whatever they come up with I'll still fold in it. Winning or losing the chimp challenge is the least of my concerns

Xavier Zepherious

OK here's whats up everyone

seems it's my ALL TEAM combined target idea - Posted earlier
or
Zodac's (which we have all pointed to has flaws).

That's whats proposed bringing to the teams for a vote this week

I don't know what it is... but they are so trigger happy - I was willing to delay the CC to fall if we had to

 
Can you give us a quick summary of each?

Zodac had proposed these:
---------------------------------------
As for the stats system, I still think one of these formulae would cover all our bases:

1 = ((Total Points/Handicap)*Conversion) + # of mils - THe One being offered
2 = Total Points/Handicap + # of mils*(Conversion+1)
3 = (Total Points + (# of mils*Conversion))/Handicap
 

 
and of course they want their best two months not to count in the handicap (HPCS months)- if you go look at OCN it's up quite a bit last two months 
http://folding.extremeoverclocking.com/team_summary.php?s=&t=37726 

handicap is based on past team performance(NOT CC team - the whole team)
 
 I did explain to them that is a no go - why should they get a 40% decrease (yet take our best during the CC last year)
 
I suggested Best two weeks for all teams
Conversion = CC PPD/Team PPD

 
 
My suggestion:
----------------
no race  
join everyone's point output together  
have a goal of surpassing combined team peak point output by 20 or 30% (for two week) - 
ie 14 days x peak PPD(avg of  best two weeks) of all teams combined x 1.3(for 30% increase) x conversion factor (what percent will change to chimp names)...this to be figured out by what came out last year (average conversion) 
  
you can still track regular info...so teams can see team output and rally troops to fold (and improve their own numbers) 
  
a goal line like a for a charity .every team puts in 
and see if we can meet it 
(it should be out of our reach and the goal will be to stretch each teams abilities as the goal) 
  
this forces teams that barely change during CC to radically change they way they do things - either ramp up or get new members 
  
and the other teams will have to ramp up extra just incase the other teams(that historically barely increase team output) don't make up the shortfall 
  
if we meet the goal we all get the badge 
if we don't we have to wait till next year and try again 
  
this would strengthen team spirit 
it would also foster good dialogue amongst the teams, as well as working with other teams to improve output (and would end cheating) 
  
because why would we drop wu's intentionally - since there is no winners here unless stanford gets usable data 
  
the winner is suppose to be all mankind by finding cures and science 
  
look even if no one likes it - it is a novel idea 
just thinking outside the box 
 
PS: if troy or anyone else has an proposal please submit it soon
 
 
 
 
I'll post OCAU's suggestions again - although they are not in consideration
-----------------------------------
OCAU
I think its a great idea to potentially offer/recognise the efforts of the teams in as many ways as we can.
Giving minor kudos (in the form of an E-wang.jpeg) is a way to encourage teams.
I believe we need to keep the original CC as the main attraction, but loved the idea as above of having 

best PPD - Monkey Testicles.jpeg
best growth - Monkey Brain.jpeg
best WU output - Banana.jpeg
best overall - CC Trophy.jpeg
Last Place - The Dreaded Monkey Paw.jpeg

or some variation of them.
Of course that is just a weird summation but perhaps the award for each genre could represent something to do with our research. (after all this is about awareness of our project)
for e.g - best growth (monkey brain.jpeg) category is represented by research to Alzheimers, best ppd (monkey testicles.jpeg) category is represented by research to cancer and so on an so forth.

This in itself will provide something positive for each team to go away with, rather than the thought of gathering the troops for battle and getting raped by a big black hairy monkey again  



Onto the major attraction .....
Its going to be a challenge to find a system that rewards all teams especially given the diversity of production, size etc.
There are limited ways in which case we can monitor and calculate a system.

Could we construct a system based upon:
Total CC ppd/Active clients over last 50 days (using the CC monkey name)
adding in CC point bonuses based on achieving milestones such as 1 mil points OR 1000 WU's 

Most of our teams have a decent mix of GPU/SMP/BBA so it would hopefully be a nice even spread.
This will hopefully alleviate the pressure of the big teams v's small teams so its not all about how many ppl we convert but more so about the average production of each client.
I think this helps solve our issue with merging teams too
Any new teams that enter (if we decide to open invitations) will be able to slot straight in with out having to think about qualifying periods or different noobie category.

just some random thoughts...