-
Discover new ways to elevate your game with the updated DGCourseReview app!
It's entirely free and enhanced with features shaped by user feedback to ensure your best experience on the course. (App Store or Google Play)
Movement in top 10
- Thread starter Mando
- Start date
BionicRib
* Ace Member *
- Joined
- Jun 7, 2011
- Messages
- 2,617
Just for argument sake lets say you have 100 reviews..............99 of them are between 3 and 5 give lets say an average of 4.2..................and then person gives a review of 1.........
You can go both ways...........100 reviews...........99 between 0 and 2.5............and one person gives it a 5
when there are only 5 reviews I agree......let them all come in, but when you have a lot of reviews with a small standard deviation and somebody pulls that you have to get rid of the outliers.
just my two cents
You can go both ways...........100 reviews...........99 between 0 and 2.5............and one person gives it a 5
when there are only 5 reviews I agree......let them all come in, but when you have a lot of reviews with a small standard deviation and somebody pulls that you have to get rid of the outliers.
just my two cents
jimi hendrix was an outlier.
i do understand what you are saying, but how does this policing happen and where does it stop? these reviews are opinions and the will vary drastically. i know some idiots come along to trash a course and drop its rating but some people may actually believe the course is a 1 and not a 5.
Cgkdisc
.:Hall of Fame Member:.
Only their numbers would get dropped, not the written reviews.
Your point about standard deviations raises an interesting idea. Only using ratings within a certain number of standard deviations from the mean would do exactly what you're suggesting. On courses with small numbers of ratings it would almost never come into play, same thing on courses with a wide spread of ratings already, but on courses with a strong consensus an outlier would be dropped.
Then again, that raises the issue (one I've personally seen, not just a hypothetical) of a traveling reviewer who plays an out of the way course that's received a bunch of high ratings from locals but doesn't stack up based on that reviewer's experience. Do we really want to drop that review?
- Joined
- Jul 29, 2009
- Messages
- 15,716
Great point!
Outliers aren't necessarily a bad thing.
... that's why I value reviews from people I know, or have reviewed courses I've played similar to me and TR's. Any of those factors gives me an idea how that reviewer sees things, hence a basis for determining if it's a course worth playing on a trip.
Last edited:
- Joined
- Jan 4, 2009
- Messages
- 3,465
Statistical analysis does not measure opinions though . . . it measures data.
This runs of the assumption that the pissy course trasher reviewers only review courses once a certain number of "good" reviewers establish some sort of norm. What if the pissy course trasher reviewer gets there first.
The thing is that outlier review score will hurt a course with 5 reviews a whole lot more than it will one with 100.
- Joined
- Jul 29, 2009
- Messages
- 15,716
Until you have a significant sample size, you really can't tell if a review is an outlier (be it from pissy course trasher or a homer pumping their local course). I say let 'em stand - over the long haul, good courses rise to the top and bad courses get revealed.
To those who have a problem with this, as has been said in numerous threads before: read the reviews. A good reviews is descriptive enough and contains the substance to support their rating... whether or not it's an outlier rating. If you're that worried about a .5 disc review damaging a top 10 rated course's rating, get over it - the couse's feeling won't be hurt.
Last edited:
gdub58
Birdie Member
Removing outliers is appropriate when you have a much wider scale or no scale at all and the response makes no sense. For example, in market research we might remove someone who says there are 50 people in their household or that they have purchased detergent 100 times in the past month.
The other rationale for removing an outlier is to avoid a specific outcome, such as excluding low rated rounds to prevent sandbagging.
Here we have neither. Now, if somehow top 10 courses got special benefits like sharing a fund for course improvements or the awarding of NT tournaments then it would be worth considering...
The other rationale for removing an outlier is to avoid a specific outcome, such as excluding low rated rounds to prevent sandbagging.
Here we have neither. Now, if somehow top 10 courses got special benefits like sharing a fund for course improvements or the awarding of NT tournaments then it would be worth considering...
Cgkdisc
.:Hall of Fame Member:.
A bigger issue than outliers in my experience is reviewers who post reviews just after a course's baskets are installed and don't go back to update their reviews once everything is in place. It's a rare course that opens with everything ready to go. In fact, as a designer I would downgrade a course that opened with cement tee pads already in place for being too eager to finalize the design before it had been tested with play. Likewise, only temp tee signs should be in place when a course opens in the event there will be tweaks. The more important signs when a course opens are directions to the next tee.
- Joined
- Jan 4, 2009
- Messages
- 3,465
That is something I thought about recently Chuck . . . I am sure a lot of designers layout a course, maybe even throw some shots on the holes and then think it is complete. What happens when the city cuts the wrong tree down, or the buckthorn gets trampled int he first few weeks of a course being open, thus opening a lane to put another tee or fairway in danger? I like the idea of temp tees initially to let the course play in a little bit.
Our first post, we want to give a 'shout out' to Foundation Park in Centralia Iliinois. Our favorite in the St. Louis area thus far. 70 miles one way but worth the voyage. Especially appreciate the logical sequence of holes, the wide open spaces and an abundance of shade, which comes in handy this time of year. Nice park.
there should be a trusted reviewer top 25
Welcome to the forums...nice first post! :hfive:
Centralia has been on my wish list for a while now. I can't wait to get down there one of these days!
I actually made a small contribution to the teepad funding efforts earlier in the year. It sounds like the $$$ was put to good use.
- Joined
- Jul 29, 2009
- Messages
- 15,716
Maybe I'm slow on the uptake, but I just noticed that Selah Ranch - Creekside topped them all, and it seems fitting that it was Martin's review that did it - his was the 20th review needed to qualify.
I noticed that this evening for the first time. I think it was only a matter of time based on what I have heard of Selah. Someday I will get to Texas and play it so I can add my review which will most likely only be 4.5 discs since I doubt it is a perfect course. I could be wrong, I have been wrong before.
- Joined
- Jul 29, 2009
- Messages
- 15,716
Given the reviews I'd read, I too, was thinking it was just a matter of time. While ratings are admittedly subjective (your 4 may be my 5, and vice-versa), I think I've read enough to safely say Selah must have a couple of killer courses, and think I'd love to see them.
I think it's interesting that you don't use the top rating or the bottom three ratings of the scale. You've compressed the 11 point rating scale here to a 7 point scale, which makes it a little harder for the rest of us to look at your ratings and really know what we're seeing. Is a 1.5 course a terrible course on your scale or have you really never played any actually bad courses?
I posted in the Selah thread that I think the wrong course is on the top 10 list. Creekside is a very cool course, I have it on the border between a 4 and 4.5 on my personal scale, but Lakeside is a much better course IMO. Fewer open shots, better water shots with some amazing landing zone shots and some nice wooded holes mixed in.
