Global Scoring Time Window

Case number:845818-992082
Topic:Game: Social
Opened by:auntdeen
Status:Open
Type:Suggestion
Opened on:Saturday, March 10, 2012 - 23:13
Last modified:Tuesday, March 13, 2012 - 20:17

The four month window for global score calculation may have had good reason for the selected time frame when Foldit was in its early days, with few players.

With a fairly consistent 150+ folders online at any time (and I have no idea how many are truly active but client not connected), it may be time to rethink whether that time frame is maximal. We have about doubled the number logged in since September. From what I am hearing from newer players, as soon as their <15 & <150 scores drop off the window, they are (with the current recalculation) dropping severely in rank compared to newer players whose < scores are still in the time frame.

For those players - the ones who have stuck with the game, the ones we really want to keep - it's very dis-heartening to have your global ranking drop like that.

Also, the four month window with the recalculation just done is imposing a severe penalty on the hand folders for the almost 3 months that it became very difficult to hand build without the protein turning into cement. Those players (and yes, I am one of them) just lost the tail end of their good scores in the recalculation, and haven't yet had a chance to recoup.

There are many different ways to calculate global rankings, and I do like the concept that there is a window… new players can aspire to the higher ranks when there is a level playing field of sorts.

I would suggest, however, that the window be extended to 6 months. That would help mitigate the situations above. Going forward, it would also make it easier for any folder, when real life intrudes, to feel more confident that taking a break would not mean that they may not be able to recover their ranking within a reasonable time frame (which may actually provide incentive for them to return).

(Sat, 03/10/2012 - 23:13  |  44 comments)


Joined: 06/17/2010

Hmm? Bigger window? NO!
Why? Because this will make new players even harder and longer to get some ranking. It also will make longer time to get "proper" place for returning players. Every newcomer want to get top 100 fast then keep going higher and higher every puzzle.
IMO it should be even SHORTER: 3 months is enough. Recalculation should be done every puzzle close.
I`m not sure, that <15 and <150 puzzle scores are dropped when player pass that? It should NOT happen. If it is calculated this way it should be fixed. All puzzles should count. Even if player do 3 times one puzzle at start <15 <150 and normal one. Any afford need to be counted. Points need to be cumulated.
I think that rank should be changing as much as possible. This way only really addicted/dedicated folders will keep "their" rank and competition will make players more active.

Madde's picture
User offline. Last seen 10 weeks 6 days ago. Offline
Joined: 05/29/2008
Groups: Void Crushers

Just to clarify:
Global points from <15 and <150 puzzles will be dropped when they leave the 4-months time window, not when players have more than 15 or 150 points.

Madde's picture
User offline. Last seen 10 weeks 6 days ago. Offline
Joined: 05/29/2008
Groups: Void Crushers

Wow, two down-votes for stating a fact?
I didn't know March 11th was Deny Reality Day.

Joined: 06/17/2010

Hahaha... looks like tis fact is disliked by some newbs :)
If it is that way it is ok, no fix need.

spmm's picture
User offline. Last seen 32 weeks 5 days ago. Offline
Joined: 08/05/2010
Groups: Void Crushers

as far as the longer window is concerned if you take time off, as I did for six weeks to have a life, then currently you pay for it in global points for the next four months. I like maddes greatest folders list and maybe a slightly shorter running game window and a better long term hall of fame would work well together. That way the great folders who just don't get to play every single puzzle for whatever reason still show up in the lists.

brow42's picture
User offline. Last seen 1 day 3 hours ago. Offline
Joined: 09/19/2011
Groups: None

I agree with Rav3n about the shorter window. It allows new or inactive players to get into the rankings faster. It also lets bad scores from bad clients to fade faster.

Joined: 10/11/2011
Groups: None

I think the scores should be done at the end of each puzzle..its the only way to have an accurate scoring system and ranking system. Monthly at the longest.

I have left and come back many times and what madde states NEVER used to happen.

Gee... you dont need to be a rocket scientist to figure out why foldit cannot retain any new players...after 4 months they get half their score wiped..what is the point..where is the incentive to stay?
I will be brutally blunt ( altho not as blunt as I would really like).....it is the most idiotic and stupid thing I have read in the 3 years I have played this game.

Joined: 06/17/2010

? I don`t understand.
Scores CANT be accumulated all the time (global score as sum of all puzzles you ever played). If it has work this way only "old" players would be in top and newcommers would have NO ability to get some top spot.
This way as it works (scoring window) allow to any good folder get top ranking in max time of scoring window if he/she play almost every puzzle and end in some top spot.
I not see how "lose half of my points" can happen.
Every recalculation scoring window is moving for EVERYONE and EVERYONE lose points from puzzles that are left window. If you play some good some bad your total score is fluctuating much, if your condition is steady your points are steady too. If you are improving every puzzle your score will grow.
Calculation should be done only after every puzzle close to avoid some strange shifting between puzzles.

B_2's picture
User offline. Last seen 4 years 16 weeks ago. Offline
Joined: 11/29/2008
Groups: None

It appears that Tealight doesn't understand the scoring system at all.

brow42's picture
User offline. Last seen 1 day 3 hours ago. Offline
Joined: 09/19/2011
Groups: None

I don't know about losing half my points, but I know I lost half my rank. I had 80 points from my first < 15 foldit puzzle and well enough on the other two to never be in the < 150. When those points expired I dropped from #100 to worse than #200.

You can't have it both ways. A longer window makes it a six month slog from #2000 to #100, and then you stall at #100 because everybody above you does every single puzzle. A shorter window drops inactive players faster. Both require non-stop good performance to the top 100, top 50, top 25.

I personally think new players want to rank up faster than six months, so I support a shorter window. Perhaps the real problem is that the starter puzzles are worth too many points.

B_2's picture
User offline. Last seen 4 years 16 weeks ago. Offline
Joined: 11/29/2008
Groups: None

I agree that the shorter window and much less points for the starter puzzles is the answer.

Newbies get to keep big scores from all 3 versions of the same puzzles so they jet up through the ranks unfairly, then they complain that they plummet when all those easy puzzle scores fall off at the 4 month window.

spmm's picture
User offline. Last seen 32 weeks 5 days ago. Offline
Joined: 08/05/2010
Groups: Void Crushers

As far as I understand the scoring it goes like this:
There are a certain number of puzzles in any time window, so newer players are at a disadvantage until the four months is calculated, ie when they have competed in the same number of puzzles as everyone else, at that point if they have been playing well (and usually you also need to play every puzzle) their global score will go up. Whis is why it helps to have the restricted beginner puzzles.

The image shows how it works - the four players start a month after each other

Maximum number of points available at each 4 month recalculation is shown

That way if you have played the whole four months window you can be rank one.

As rav says if points were cumulative since 2008 no new people would be in the higher ranks.

Ranks are also calculated for each puzzle within the window, which means you can go up and down within the four months which keeps it interesting, your total at the end of four months is I think what you start the next four months with.

imho if <15 and <150 players are playing the same puzzles they should get the same points, it doesn't make the puzzle any easier.

Joined: 10/11/2011
Groups: None

The reality is, it isnt working like you say.

Joined: 10/11/2011
Groups: None

actually B2 I do understand it..

spmm's picture
User offline. Last seen 32 weeks 5 days ago. Offline
Joined: 08/05/2010
Groups: Void Crushers

You don't start the next four months with the previous score - blonde moment :)

brow42's picture
User offline. Last seen 1 day 3 hours ago. Offline
Joined: 09/19/2011
Groups: None

The whole issue of sudden drops can also be avoided by exponential scoring. Everyday everybody loses 1% of their global points. Any given puzzle loses half its value in 10 weeks. If you're consistent, your score levels out. There's still a max score for all players, regardless of how long they've been playing, that new players can approach in a reasonable time. There's just no sharp cutoff. If people are stressed by seeing their points disappear, it could be recalculated only when puzzles close.

Joined: 06/17/2010

Nononono, this is terrible.
It is much easier to avoid: do NOT play all three (<15, <150 and normal one) puzzles when you can :)

Madde's picture
User offline. Last seen 10 weeks 6 days ago. Offline
Joined: 05/29/2008
Groups: Void Crushers

That's a good idea *but* it would be impossible for players to catch a flawed leader-board with such a complicated system - and we all know we can't really trust the leader-board.

spmm's picture
User offline. Last seen 32 weeks 5 days ago. Offline
Joined: 08/05/2010
Groups: Void Crushers

it is not just about sudden drops it is also about sudden rises for the players who did well.

infjamc's picture
User offline. Last seen 2 years 37 weeks ago. Offline
Joined: 02/20/2009
Groups: Contenders

Since we're in the mood for brainstorming, here's another idea I would like to suggest:

Why not consider tweaking the "Points = Max(1, RoundUp( 1 - (Rank - 1)/(NumPlayers - 1) )^7 ) * X" formula instead? To see why, let's plug in a few numbers:

* 50 percentile finish: 0.5^7 = 0.0078, or 1 point on a 100-point puzzle
* 69.1 percentile finish (0.5 standard deviations above average in a normal distribution): 0.691^7 = 0.075, or 8 points on a 100-point puzzle
* 84.1 percentile finish (1 std. dev. above average): 0.841^7 = 0.298, or 30 points on a 100-point puzzle
* 93.3 percentile finish (1.5 std. devs. above average): 0.933^7 = 0.616, or 62 points on a 100-point puzzle
* 97.8 percentile finish (2 std. devs. above average): 0.978^7 = 0.851, or 86 points on a 100-point puzzle

==> Obviously, the distribution of scores are right-skewed rather than Gaussian, but the trend is clear: unless a player can consistently do very well in every puzzle, it is very likely that having a handful of top performances mixed with average performances would score better than a consistent performance that's only slightly above average. For example, a 97.8 percentile finish in one puzzle combined with a 50 percentile finish in ten puzzles would give you more global points than a 69.1 percentile finish (slightly above average) in the same eleven puzzles.
.

Why is this relevant? Because newbies are much more likely to score high in a <15 or <150 puzzle than in a "standard" puzzle available to everyone. Take Puzzle 506 as an example:

* <15 version: 1265 players, so the 90% percentile mark is at #126 (49 points). The score for this player is 10326.
* <150 version: 413 players. The same 10326 score that landed at the 90th percentile in the <15 version would be good enough for 69th place (83th percentile) and about 29 points.
* Regular version: 579 players. The 10326 score would land at 250th place (57th percentile), for 2 points.
.

So imagine the shock for the new player who graduates from the <15 and <150 puzzles:

* First puzzle: newbie plays in all three versions and picks up 49+29+2 = 80 points
* Second puzzle: the player is eligible for <150 and the regular version and picks up 29+2 = 31 points
* 80+31 = 111, so the new player continues playing in both the <150 and regular versions of a third puzzle for another 31 points (subtotal: 142). Let's say that several more CASP ROLL puzzles show up, and the new player gets 2 points in a few of those to get to 150.

* That means no more <150 puzzles. Suppose that this player's skills has improved slightly by this point so that he/she now able to consistently reach the 63th percentile, for an average of 4 points per puzzle. Since there have been about 40 puzzles in the last four months, such a progress is just enough to keep this player out of the <150 puzzles when the oldest scores expire. What kind of rank can one get with 160 points? #445, which is not that bad considering that it's out of 12317. But this might discourage the player and cause him/her to quit.
.

==> But let's see what happens if we lower the exponent in the equation used for calculating global points. For example, what happens if we use 5 instead of 7?

* 50 percentile finish: 0.5^7 = 0.031, 4 points out of 100 (vs. 1 right now)
* 69.1 percentile finish (+0.5 std dev.): 0.691^5 = 0.158, or 16 points (vs. 8 right now)
* 84.1 percentile finish (+1.0 std dev.): 0.841^5 = 0.421, or 43 points (vs. 30 right now)
* 93.3 percentile finish (+1.5 std dev.): 0.933^5 = 0.707, or 71 points (vs. 62 right now)
* 97.8 percentile finish (+2.0 std dev.): 0.798^5 = 0.894, or 90 points (vs. 86 right now)

Notice that the player would be earning more than the minimum for an average performance, and a +2.0 std. dev. performance is only worth 5.7 rather than 10.8 puzzles at the +0.5 std. dev. level. Essentially, the system is still rewarding outstanding results while the slightly above-average ones are now worth a little more. To revisit the earlier example, a 63th percentile finish would be worth 10 points per puzzle under this new system, and 10x40 would mean keeping 400 points when the <15 and <150 scores expire. That's good for #208, which looks twice as better than before.

Madde's picture
User offline. Last seen 10 weeks 6 days ago. Offline
Joined: 05/29/2008
Groups: Void Crushers

I think you forgot that in your second example (5 instead of 7) almost all players would gain more global points, so the 400 points were good for #208 in the existing system but not in your example.

infjamc's picture
User offline. Last seen 2 years 37 weeks ago. Offline
Joined: 02/20/2009
Groups: Contenders

That's a very good point... also, it has been brought to my attention that there are actually 53 puzzles in the last four months (I only counted off three months in my previous post).

Obviously, it would be difficult to estimate the new rank under the case where the exponent is 5 (since it would require recalculating everyone's scores). but I would expect that the general trend--that the gap between a slightly above average performance and an outstanding performance would be narrower--would still apply. So, the new rank will mostly likely be lower than 208, but still higher than 445.

Joined: 04/19/2009

The six month window was a first ditch attempt on my part to think of some way to soften the blow to the newer people who dropped so severely in ranks.

It was also an attempt to get the conversation started, and it's a great conversation!

Of all the ideas presented so far, I am liking the: shorter window (coupled with Maddes greatest folders list) and the exponential points (wow, nicely thought out, infjamc).

Another (perhaps hare-brained) idea to throw out here may be to keep the <15 & <150 puzzles separate from the normal puzzles for global points, period. Have a <15 global points scoreboard, a <150 one, and our usual normal one. The benefit would be that a new player can judge themselves against their peers, and be able to get their feet wet in normal (with a more realistic rank) - and do both at the same time.

That's really what the current system was designed to do, and it functioned well with fewer players - with the current larger numbers, it now fails in encouraging players who start out slowly in terms of "getting it". And yeah, I was one of those players when I started - I would have been one who dropped the 100 ranks. It took me 2 months to have one finish in the top 30... 4 months for a top 20... 8 months for a top ten (then another four before I was able to do that again)... almost a year & a half for a top 5. My progression was very slow, but it was steady, and that encouraged me. I was able to eventually gain global rank #7 at one point, so I guess that I'm not so bad at the game after all - perhaps it would have been a shame to have chased me away with a scoring system that I didn't understand when I started.

Joined: 06/17/2010

Another idea of dealing <15 <150 and normal puzzle: if player go for all 3 versions of one puzzle only best score (global points earned) counts. This way we allow to play on all 3 and this also prevent sudden rank drop if player do all of them high ranked and puzzle left scoring window.

infjamc's picture
User offline. Last seen 2 years 37 weeks ago. Offline
Joined: 02/20/2009
Groups: Contenders

Alternatively, the "regular" version of the puzzle could be hidden from those with <150 global points to prevent the case where the best Foldit score is achieved on the "regular" version of the puzzle. Also, as brow42 has already mentioned, the <15 and <150 puzzles could be weighted less-- say, 25% and 50% of that of the regular puzzle, respectively.

Joined: 06/17/2010

No, they can not be scored less, they are identical. <15 is not easier than <150

mimi's picture
User offline. Last seen 28 weeks 4 days ago. Offline
Joined: 11/17/2008
Groups: Contenders

I'm going to throw another thought into this mix while the subject is being discussed and say that while puzzle scores are a good thing for hooking players in and getting them trying to improve, they are probably ultimately a bad thing when it comes to actually producing the desired results from the game.

Score chasing tends to get us constantly digging down into local energy minima and all narrowing down to the same structure rather than experimenting and producing a diverse range of options in the solutions to a puzzle.

Problem is to find a way to reward people for puzzle results other than on the basis of the best score - which is not always an absolute indicator of the best solution anyway.

I haven't got any answers, just a feeling that it would be a good thing if we could do something of that nature in order to get better solutions for CASP and beyond.

infjamc's picture
User offline. Last seen 2 years 37 weeks ago. Offline
Joined: 02/20/2009
Groups: Contenders

I like the idea, but the tricky part is that relative contribution to science is difficult to quantify:

* For CASP puzzles, one possibility is to calculate global points based on deviation from the native in terms of GDT_TS or some other metric. The problem with this approach is that it requires manual recalculation on the part of the Foldit staff (since only the top Foldit models are submitted to CASP); plus, it obviously doesn't work for puzzles where there is no native to compare to.

* Yet another possibility is to create a new achievement type for individuals whose results represent significant contributions to a study that ultimately results in a published journal article. (Obviously, "significant" is subjective, so this will have to be awarded manually by the researchers.) For extra visibility, perhaps the player could be allowed to be adorned with a star next to their username (just like the ribbon for admins). The idea is that there would be a tangible reward for actually "doing something."

spmm's picture
User offline. Last seen 32 weeks 5 days ago. Offline
Joined: 08/05/2010
Groups: Void Crushers

the achievement idea is great infjamc, partcularly if there can also be one for coders.
My understanding is that in CASP the known structure has been found, just not released, so whilst it may take a while the deviation from the native may be calculated?

Also in the olden days (18 months ago) beginner <15 and <150 puzzles were only up for a weekish, so it was harder to do well in the <15 and then transfer that to the <150 and the unrestricted puzzles, thereby scoring well in several puzzles.
I am not suggesting that beginner puzzle times should be shorter, I don't think that is a good idea, but it may be a factor in beginners getting higher scores at first, then being disappointed when they drop.

jflat06's picture
User offline. Last seen 7 hours 56 min ago. Offline
Joined: 09/29/2010
Groups: Window Group

This issue is probably caused by the beginner puzzles being weighed too heavily in score. The intent with having the <150 and <15 puzzles available is to give new players a chance to compete against their newer peers, and not to give any long lasting advantage. At the same time, weighing them less would make it harder for these newer people to get out of the <150 and <15 categories. This may not be so much of an issue given that these puzzles are up for so long, and that often times a new player will compete any many other puzzles before their first beginner puzzle closes. The solution which I think makes the most sense is for the scores between 15/150/unlimited to be limited to your score for the version which would have given you the most points.

Joined: 10/11/2011
Groups: None

Well that may have been your intent.
Not very well thought out in that case.

The reality is, all you are doing is lifting new players up ..and then kicking them in the teeth.

Unless you have actually played and been on the "receiving" end, then one has no idea just what this phony ranking opportunity does to new players.
Of course noone really knew it was "phoney" until you just enlightened us

Not so much of an issue?..sorry I do not wish to be rude but again unless you have experienced this personally...then you really have very little idea of the effect upon players and ramifications.

Up until the other day I would also have shared many of the viewpoints shared by people who have not had the dubious pleasure of experiencing it...how fortunate that I was.( sarcasm)
Now I understand how crap... new players feel.
Now I understand how they can play their little hearts out...then sink into the abyss never to be heard of again.

Instead of reading posts by veteran players..why not go listen to a few noobs and see how they feel about it?
Some wont post here because they dont want to get "shot down" by ^^^^^^^^^^

No doubt this post will get the "natives" riled up..but seriously...not one of you has a clue as to what these "ideas" do to new players.

spmm's picture
User offline. Last seen 32 weeks 5 days ago. Offline
Joined: 08/05/2010
Groups: Void Crushers

Tea .

You have said that: ''I think the scores should be done at the end of each puzzle..its the only way to have an accurate scoring system and ranking system. Monthly at the longest.''

Perhaps you would fleshout what you mean by that - only the score of the last puzzle gives you rank? Scores are already calculated at the end of each puzzle.

Or do you want the scores to start and be totalled from when you started 3 years ago? so no new players can actually compete against the active older players even if their scores seem to go up gradually.

I'm sure everyone would like to hear from the actual new players who are unhappy, and of course the ones who are just fine with it all; as madde says no one should be voted down for providing factual information.

Returning players with the same or different names have a major advantage if they are able to include <15 and < 150 points in their totals so one would understand that they would be upset about apparently loosing those points.

Joined: 10/11/2011
Groups: None

Spmm,

I have already provided my suggestions and ideas to the people who are able to consider and implement a more equitable regime that is based on reality, as have others.

spmm's picture
User offline. Last seen 32 weeks 5 days ago. Offline
Joined: 08/05/2010
Groups: Void Crushers

also my understanding is that global scores have to be calculated manually, decreasing the frequency from 4 months to monthly, even staying with the same system, would be drawing valuable resources away from other things of higher priority. Having a longer window would probably suit fold central better but can really disadvantages players.

Joined: 10/11/2011
Groups: None

Scoring is quite capable of being automated if that is your concern

spmm's picture
User offline. Last seen 32 weeks 5 days ago. Offline
Joined: 08/05/2010
Groups: Void Crushers

well tea/vixy - not that I wish to be rude of course - if you have a fast track to the movers and shakers don't waste our time in the publicly available channel of foldit feedback, just get them to do your bidding when you throw a tantrum.

Please don't keep playing the hysterical ''the newbies'are upset'' to jflat and other developers until you can actually produce some newbies who are not returning players and are not ínstant newbies. Those comments are very rude in my opinion although the devs can look after themselves of course.

Don't worry about the rest of the players or actually providing any real or articulated solution to a problem which it seems is only of concern to you. The conversation had been quite interesting and pleasant until your hysterical comments.

Obviously anything can be automated but why would we bother when there is so much else that needs doing and you are the only person with a problem?

Joined: 09/21/2011
Groups: Void Crushers

Why not make the ranking system a weighted one?
Scores < 2 month count 100%
Scores 2-4 month count 50%
Scores 4-6 month count 25%
This way the scoredrops will not be that heavy for newbies and can be compensated if their skills increase.

B_2's picture
User offline. Last seen 4 years 16 weeks ago. Offline
Joined: 11/29/2008
Groups: None

I get it -

Vixxyn started over with a new account after her last "leaving", and just after 4 months later has had all her <15 and <150 scores fall off. This whole rant is personal, and of her own creation.

Joined: 10/11/2011
Groups: None

Actually brick I left last time due to the people like you who are total and utter jerks.

I came back after some chats with friends I made on here...only this time I wont take rubbish ..from anyone.
I guess some of you lot are not used to that...well get used to it.

I can understand your feeling threatened by a change to the system that is unfair to new players but protects you...but foldit doesnt really need the old players who contribute diddly squat except to be a totally negative force concerned only with themselves..people like you.

You help noone you offer nothing positive except to deride and attack others.

They need the new blood like the people who leave.

I left in September, I came back in October.
It had nothing to do with getting a rank.

If your inane and pathetic attack makes you feel fulfilled..go for it I am more than up for it

Joined: 10/11/2011
Groups: None

SPMM your comments directed at me are totally out of line..just because I will not engage with you.
Your past behaviour in my inbox is why I will not engage with you.

Given The crap from you lot I have recieved, regarding what is an issue for new players why on earth would I waste my energy or time discussing anything with you.

It is also this behaviour you lot are exhibiting, that has prevented some new players from participating in this chat.

Hysterical?..No. What planet are you on?

Brutally honest yes..but then as you found out in my inbox some time back..you dont like that do you.

Joined: 06/24/2008
Groups: Void Crushers

I agree with Mimi; chasing Global rank with scripting seems to discourage exploring the puzzles. Now my knowledge of biology is non-existent so maybe exploring wildly may not be productive. One of the things that happens is as soon as someone finds a productive alignment everyone in a group jumps on it. As for evolving, most people do not evolve to move a solution but to get 'evolver' global points.

As for newer players; this is a hard game. Once they get out of >15 and >150 their global rank is going to fall unless they are one of the few brilliant players. Timo's suggestion does address this.

Personally I ignore my rankings I play for my group's [which I would love to rename to TEAM] ranking.
I know for solo players this isnt possible; but I wanted to make clear how I play and think so my comments make sense.

I think Infjamc's ideas address these problems a bit.

I like Timo and Infjamc's ideas:

Timo:
Why not make the ranking system a weighted one?
Scores < 2 month count 100%
Scores 2-4 month count 50%
Scores 4-6 month count 25%
This way the scoredrops will not be that heavy for newbies and can be compensated if their skills increase.

Infjanc:
* For CASP puzzles, one possibility is to calculate global points based on deviation from the native in terms of GDT_TS or some other metric. The problem with this approach is that it requires manual recalculation on the part of the Foldit staff (since only the top Foldit models are submitted to CASP); plus, it obviously doesn't work for puzzles where there is no native to compare to.

* Yet another possibility is to create a new achievement type for individuals whose results represent significant contributions to a study that ultimately results in a published journal article. (Obviously, "significant" is subjective, so this will have to be awarded manually by the researchers.) For extra visibility, perhaps the player could be allowed to be adorned with a star next to their username (just like the ribbon for admins). The idea is that there would be a tangible reward for actually "doing something."

Joined: 04/19/2009

@ spmm & Brick

Shame on both of you for taking this to a personal level.

spmm - the words "tantrum" and "hysterical" are out of line, emotionally laden words. There was no reason to use those.
Brick - your entire comment is incendiary, unnecessary, and one of the rudest personal attacks I've ever seen here.

To both of you: please take the time to read all of the comments... This is not about tealight! Because she is a returning folder, she has been brave enough to say what happened to her and how she feels about it.

You have both missed the fact that Brow42 (a good player and an up & coming good scripter) has commented that he also was adversely affected in the recalculation in a big way. Trying to make this personal against tealight negates his voice, and certainly doesn't encourage any other newer players to engage in this conversation.

I started this feedback because I became aware, the first day after the recalculation, that there were more than a few newer players who had had a very ugly surprise. I am concerned that some of those players will leave. And no, they do NOT wish to be publicly identified - they feel bad enough about it without subjecting themselves to vets who may personally attack them.

infjamc's picture
User offline. Last seen 2 years 37 weeks ago. Offline
Joined: 02/20/2009
Groups: Contenders

I'd like to offer my perspective on this issue.

* When I first joined Foldit back in 2009, I didn't really experience a sudden drop in ranking due to the expiration of old scores myself. But that's partly because I picked up protein folding very quickly-- after achieving several top 10 finishes in a few <15 and <150 puzzles, I somehow managed to maintain an average of 47 points per puzzle over the next four months. As a result, I was in the Top 25 in no time and had no reason to complain. (In fact, I was surprised at how well I was doing.)

* But even then, I'm not immune to the issue of sudden rank drops either, as hot streaks and cold streaks can still occur from time to time. In my case, the difference could be as large as an average of 30 points per puzzle in one month and 60 points per puzzle in another-- which would extrapolate to a difference of 46 places in the current soloist rankings (#15 vs. #61) if sustained over a 4-month period.
.

==> Now, I know that I am not the best example. But I can still sympathize with newer players who are hit by ranking drops, as the line of reasoning goes like this: "If even a top-25 player like me can be surprised by being dropped 10 places in the soloist rankings from time to time, then I can only imagine how much newer players would be devastated by their <15 and <150 scores expiring."

This is exactly why I don't think that simply removing "double/triple counting" in puzzles with <15 and <150 versions is enough (even though it's a good idea). Why? Because the net effect will most likely to be minimal. I'll use myself as an example-- below are the number of points that I received from the <15, <150, and regular versions of my first six Foldit puzzles:

* 126: 4 / 1 / 4
* 128: 1 / 1 / 33 (yes, the high score was achieved on the "regular" version...)
* 129: 84 / 1 / 1
* 130: 39 / 1 / 1
* 131: NA / 66 / 1
* 132: NA / 100 / 4

In my case, counting only the highest score out of the three versions would only have dropped the total number of global points earned from these six puzzles from 342 to 326. So, if a new player manages to replicate this kind of performance in the <15 and <150 puzzles but can't find immediate success in the "regular" versions of the puzzles like a did, a rude awakening awaits them in four months:

Estimated global points after four months (53 puzzles): 326 + 4*47 = 514, rank = #179
Estimated global points after the <15 and <150 scores expire: 4*53 = 212, rank = #330
.

When the rank is already a large number, seeing that number double can be devastating. This is why I came up with the idea of lowering the exponent in the score calculations, so that the gap between a top-ten finish in the first few <15 and <150 puzzles and an average finish in the subsequent puzzles isn't as large. In addition, I think the FAQ needs to make it clear that all scores expire after 4 months.

jflat06's picture
User offline. Last seen 7 hours 56 min ago. Offline
Joined: 09/29/2010
Groups: Window Group
Status: Open » Open

This isn't an issue that people should be taking personal in any way.

I can appreciate that you are more personally affected by this than other people, and that it feels as though new players are being singled out, but that isn't the case.

While it is important for me to understand how "people feel about it", just understanding that people are upset isn't all that helpful. I'm more interested in understanding why they are upset, and what possible solutions we could implement in order to resolve the issue. That is the purpose of the discussion going on in this thread.

Now, you may feel that this setup lacked foresight on our part. However, things have been structured this way for quite a while, without people reacting this way before. The setup isn't entirely unreasonable, but I agree that it has flaws, which we will attempt to address. But we need to address it in a way that wont just end up causing more problems down the road. That's why we're having this discussion.

I feel as though this thread has become more personal than helpful, so I'm locking this feedback. I realize that some of you may have additional comments on this topic, and if you feel so inclined, you can send me a PM. Thank you for your feedback and proposed solutions - we're considering the options and should implement something soon.

Sitemap

Developed by: UW Center for Game Science, UW Institute for Protein Design, Northeastern University, Vanderbilt University Meiler Lab, UC Davis
Supported by: DARPA, NSF, NIH, HHMI, Amazon, Microsoft, Adobe, RosettaCommons