Grading caches vs. rating them.

[+Vater_Araignee]

I like what tindling's proposed regarding a rating system:

- Everybody who found the cache, is able to rate it

A major point with which I've long disagreed. There have been plenty of times while on a hunt that I've recognized that it was pure junk and had stopped the hunt. A 1 star for sure. However, because I refuse to waste any more time on a worthless cache I wouldn't have the opportunity to present my opinion on the cache.

 

First a visit is not a find. Folks visit caches all the time and not produce a find. Sometimes it's because it's up a tree. Sometimes it's because they have to hold their nose as they drive off. I've DNFed caches that are great caches--very worthy of 5 stars and a return visit. Sometimes I don't get the chance to return.

 

One should be able to rate a cache they visit, not find. Proof of a visit could be required much like virts.

 

Allow ratings on visits and I might start logging visits on crappy caches simply to rate them the low rating they deserve.

 

Heck, TC.com allows ratings simply because you visited the cache page. A find will add weight to your vote, but you still have a say simply from looking at the cache page.

 

Yeah, I know, "how can you have an opinion if you've never visited the cache?" Well, probably with the same argument that you can look at a map or read the description in order to know it's an LPC and put it on your ignore list.

Ok, so take the idea and requote it, strike out what you think is bad and insert (in another color) what you think needs to be added, rather than shoot it down completely.

Frankly it does read as if you think one point of disagreement renders the entire concept bad.

sounds fair to me.

Found it = 100%

DNF = 50% , If there are less than 2 DNF ratings they are not counted

neither = 25%, If there are less than 4 neither ratings they are not counted

 

so GCXXXXXX recives 15 ratings

5 Found its.

5 DNF

5 neither

 

The DNFs are 4,1,2,3, and 1 star ratings for a 2.2 average.

The neither are 5,2,3,4 and 1 for a 3 average

Found its 1,3,3,5 and 5 for a 3.4 average

 

Now most people would assume that this would create a 2.9 (rounding up) star rating but we are taking weight from DNF and neither.

We do this by saying that since DNF = 50% they are averaged and then the odd person is dropped from the total averaging. IE the raters must total an even number so they are dividable by 2, not the average just the amount of voters.

The "neither" votes are the same concept but they must be divisible by 4.

Now you dont have 15 votes you have 8 that look like this

Found its 1,3,3,5 and 5

DNF 2.2 and 2.2

Neither 3

Average = 3.1

 

To me even better would be to not even count the odds at all dropping the most recent votes until they meet the required increment.

so you would have

DNF 2.5 & 2.5

Neither 3.5

Found its 1,3,3,5 and 5

Average = 3.2

 

However I'll ultimately say someone who has not made the visit should use DNF to vote.

You love/hate an LPC soooo much that you just have to vote on it without a visit? Log a DNF and vote.

I mean technically you Did Not Find it.

Now the cache is 3 average.

better yet lets split the difference between the to for DNF because we know that people will log a DNF just to give a low rating.

 

One question tho, do you want the people in your "Trusted Rater List" to judge a book by it's cover, ie lying?

I know I want a little honesty.

[CoyoteRed]

Ok, so take the idea and requote it, strike out what you think is bad and insert (in another color) what you think needs to be added, rather than shoot it down completely.

Unless I'm missing something, that's exactly what I did. I don't like that most ratings systems only allow finders to rate a cache when folks could potentially be standing there with the cache in their hands and say, "You know, this sucks so bad I'm just going to put it back without signing the log."

 

One question tho, do you want the people in your "Trusted Rater List" to judge a book by it's cover, ie lying?

I know I want a little honesty.

See my above example to see one example of having a pretty good idea how good a cache is without fulfilling the criteria of being able to claim a find.

 

EDIT TO ADD: I like your suggestion of being able to rate on a DNF.

Edited April 2, 2010 by CoyoteRed

[+amnonymous]

It seems like that main opposition to rating system is because the possibility of "rating wars". This can be avoided by positive only rating system, like the "like" in facebook.

My suggestion is to allow "likes" or "recommendations" for specific criteria of each cache, like:

• Recommended place to visit
  
• Nice hide
  
• Nice container
  
• Nice cache contents
  
• Good page description
  
• Challenging terrain
  
• Recommended for families
  
• Easy one
  
• Challenging one
  
• Nice puzzle
  
• Nice multi steps
  
• Nice letterbox stamp
  
• Nice Wherigo game
  
• Educating Earth cache
  
• etc.
  

Of course, some of the features will appear only for specific cache types. Although, it's possible to limit some features, like cache content to those who found it.

Visitors than will see information like:

17 cachers recommend this cache for families, 23 recommend it as easy one, 2 recommend it as challenging one...

Then, one will be able to search for specific criteria, e.g. for caches with at least 10 recommendations for nice hide.

[+BigFurryMonster]

Ah! Another cache rating thread. Another chance for me to point we actually do have a working system already here in the Netherlands. It's based on scoring, not on rating, and running on daughter-site gc.nl.

 

What is needed is a system that is:

• simple and straightforward
  
• easy to use (supply information) for finders of a cache
  
• ease to use (obtain information) for those who want to hunt a cache
  
• not based on some 'average'
  

The system we have is as follows:

• For every 20 founds, a cacher gets to award one 'star' to a found cache
  
• 'Stars' are accumulated per cache, and per cacher
  
• A relative ranking 'stars per find' is composed on a per-cache basis
  
• There is a minimum threshold of 60 found caches to be able to participate
  

Best of all, this system has proven it actually works. The only flaw is that gc.com hasn't implemented it yet.

 

When I go to another city and look for caches, I pick the ones with >5 stars only. The system never lets me down. As a cache owner, I'm happily accumulating stars on my caches because I put it the extra effort to make them memorable.

[+Vater_Araignee]

Ah! Another cache rating thread. Another chance for me to point we actually do have a working system already here in the Netherlands. It's based on scoring, not on rating, and running on daughter-site gc.nl.

 

What is needed is a system that is:

• simple and straightforward
  
• easy to use (supply information) for finders of a cache
  
• ease to use (obtain information) for those who want to hunt a cache
  
• not based on some 'average'
  

The system we have is as follows:

• For every 20 founds, a cacher gets to award one 'star' to a found cache
  
• 'Stars' are accumulated per cache, and per cacher
  
• A relative ranking 'stars per find' is composed on a per-cache basis
  
• There is a minimum threshold of 60 found caches to be able to participate
  

Best of all, this system has proven it actually works. The only flaw is that gc.com hasn't implemented it yet.

 

When I go to another city and look for caches, I pick the ones with >5 stars only. The system never lets me down. As a cache owner, I'm happily accumulating stars on my caches because I put it the extra effort to make them memorable.

Actually that is based on some 'average'

The average amount your allowed to say "I liked it" on, as your finds go up, the average goes down.

Starts off at 60 caches, then you average every 40 caches, then 33.3, 30, 28....

I'll add you will never be able to hit an average of every 20 caches.

 

And not to be a jerk about it just relating some facts for a point.

Netherlands will fit into my state (Michigan) 6.11 times.

Now lets assume that your country and my state have the exact percentage of cachers and we will call it 5% for simplicity.

Yeah, population density.

You would have 20 cachers per square kilometer and we would have 3.

Heck lets bump Michigan's numbers up by Making it the same size as the Netherlands but keep the population the same. We have 12.

That system would take way to long to accumulate any significant data.

What happens if you run into 20 caches that stand head and shoulders above the rest, but you only get 1 award to hand out?

 

A system that would be likened to the one you describe but would be better is...

You get a star to award for every cache you found.

You can award as many stars to an individual cache as you wish up to the amount in your inventory.

 

Here is a problem with both systems, GS will be bombarded with

"I gave it to many stars and want some back"

"I went back and it sux now, I want stars back"

"He rated me a month ago with 10 and met me yesterday and took them back because he doesn't like me personally, that's not fair"

And on and on and on.

Then you got the people that will get 5 accounts, sign each name to the log for the propose of skewing their own caches.

Creating accounts to skew results with ratings vs awards system methods only works for a short time (unless the co has nothing better to do than create accounts) and eventually they would get called out on it by the investigative portion of the community.

[+Ambrosia]

I wouldn't want to publically rate caches, especially in my home area. It would make me feel uncomfortable every time I saw someone at an event and they knew that I consistently rated their caches badly. There's a difference between respectfully worded criticism in a cache log and a cold star rating.

 

But then, on the other hand, if ratings were anonymous, they could have more of a chance of being abused.

[+Vater_Araignee]

I wouldn't want to publically rate caches, especially in my home area. It would make me feel uncomfortable every time I saw someone at an event and they knew that I consistently rated their caches badly. There's a difference between respectfully worded criticism in a cache log and a cold star rating.

 

But then, on the other hand, if ratings were anonymous, they could have more of a chance of being abused.

I would make it so that the only way you could see an individuals rating is if they where in your trusted rater list, for the fine tuning of a second set of metrics.

Now that you mention it being able to opt out of being a trusted rater is a good idea but if you do it for one you do it for all.

 

~~~edit to add~~~

depending on the rating system the difficulty to abuse goes up or down and the more people using it the harder it becomes to abuse.

Just because Billy Jim Jo Bob Brain hates me and rates all my caches with a 1 star doesn't mean it will have significant effect on on the averages unless he starts creating more accounts to give me the 1 on each.

Edited April 3, 2010 by Vater_Araignee

[+BigFurryMonster]

Actually that is based on some 'average'

 

Not sure I understand you on this point. The system works this way: at 60 Founds, you get to award 3 starts immediately. Then, one extra star for every extra 20 caches you find.

If you use your stars, you're effectively creating a personal "Top 5%" list.

 

That system would take way to long to accumulate any significant data.

 

Not necessarily. At introduction, every cacher near you does receive an inventory of stars equal to 1/20 of the number of caches he found. If he likes some of your caches, he can award stars to those (one per cache).

Also note that this system becomes useful even with 'thin' data, as opposed to an average-based rating system.

 

Or look at it this way: if all caches were equal, you'd expect one star for every 20 founds on your cache. If yours is 'better' (in any way), it'll receive more than that, making it stand out.

 

GS will be bombarded with

"I gave it to many stars and want some back"

"I went back and it sux now, I want stars back"

 

The NL system is designed so that you can move around the stars at any time (within the given constraints).

It doesn't happen very often by the way.

 

Then you got the people that will get 5 accounts, sign each name to the log for the propose of skewing their own caches.

 

I thought this would happen, but haven't seen or heard any evidence of it. This is the reason why there is a minimum of found caches.

[+BigFurryMonster]

I wouldn't want to publically rate caches, especially in my home area. It would make me feel uncomfortable every time I saw someone at an event and they knew that I consistently rated their caches badly. There's a difference between respectfully worded criticism in a cache log and a cold star rating.

 

This is a good point, and the reason why the NL system was designed to incude positive feedback (stars) only.

It is an open system (I can see who awarded stars to which cache), which has the advantage that I can see what cachers who have similar taste as I do have rated. This leads to 'people who liked cache X also liked cache Y' and advice like 'here is a list of proposed good caches for you'.

[+Vater_Araignee]

Actually that is based on some 'average'

Not sure I understand you on this point. The system works this way: at 60 Founds, you get to award 3 starts immediately. Then, one extra star for every extra 20 caches you find.

If you use your stars, you're effectively creating a personal "Top 5%" list.

OK I'll give you that part, concidering I didn't know that you got three to start.

Only in so far as that because you do mention averaging without saying it below.

 

That system would take way to long to accumulate any significant data.

Not necessarily. At introduction, every cacher near you does receive an inventory of stars equal to 1/20 of the number of caches he found. If he likes some of your caches, he can award stars to those (one per cache).

Also note that this system becomes useful even with 'thin' data, as opposed to an average-based rating system.

 

Or look at it this way: if all caches were equal, you'd expect (an average of) one star for every 20 founds on your cache. If yours is 'better' (in any way), it'll receive more than that (or a higher average), making it stand out.

GC1FECM

Assuming it is average and receives 1 award, it still took 11 months to accumulate 20 finds.

During that time, it also received 16 DNFs.

I know for a fact that some of those people hated it and some loved it even without finding it and their opinions count. (thank you CoyoteRed)

Now lets assume that most of the people who found it think it is above average and the same goes for the DNFs.

Would a "restrictive" award system bring more cachers to it? No, because there are even better caches so it isnt worthy of "GREAT CACHE".

Would a rating system? Most likely, because it is above average according to the visitors.

 

You would also see this on every one of my listings "Save your awards for when you find something better, it will happen"

Then you are going to have that subset of people who may miss out on a cache that they like because they don't believe anyone could possibly think that about their own hides and accuse me of trying to gain awards with false humility.

GS will be bombarded with

"I gave it to many stars and want some back"

"I went back and it sux now, I want stars back"

 

The NL system is designed so that you can move around the stars at any time (within the given constraints).

It doesn't happen very often by the way.

So that today's rock star is tomorrow's gutter trash.

 

Then you got the people that will get 5 accounts, sign each name to the log for the propose of skewing their own caches.

 

I thought this would happen, but haven't seen or heard any evidence of it. This is the reason why there is a minimum of found caches.

Because:

A I wouldn't tell anyone I was doing it.

B If 60 caches isn't to long to wait for my main account then it isn't to long to wait for 10 alts.