BSeevers Posted November 5, 2002 Share Posted November 5, 2002 Now I understand the comparator or factor. I see how you are figuring "relative" skills to one match. I don't remember anything from statistics so I can't vouch for the methods but my gut tells me that you cannot judge someone from one match. Poor health, mental attitude and gun problems can have a drastic effect on match performance. Sometimes you just suck. I won my class(A) at 5 major matches this year. In only one did I shoot above my class percentage against most/some of the top 10 or so GM's. I did shoot higher % in other matches against some other GM's but no offense to them(they still womped me and are outstanding shooters) but Jerry/Todd/Rob are the high HF GM's. I have yet to shoot % relative to my class wins against those three. What about Eric G?? He has whomped us twice. He is better than Jerry in about the same amount that Jerry is better than us! How would that effect standings if he had shot Nationals? I would like to see a National, moving, ranking like tennis. If you base everything on one match, well, that's fine but that's really what we do now, and kind of what your system does. An example of another concern is these 2 shooters in Area 7 ty43660 60.7 89.81 67.58713 % f37355 61.95 98.16 63.11125 % 98% to 89% would be the difference of 9th to 2nd place at most majors. Probably difference between winning M or second GM and being 4th Master. A lot more than the 67% vs 63 % would suggest? In fact if you start studying match percentages you see we are WAY off the performance of the top 10 GMs. I still question this method but really enjoy your efforts. We have got to do something til we shoot again:) Link to comment Share on other sites More sharing options...
Dowter Posted November 5, 2002 Author Share Posted November 5, 2002 "my gut tells me that you cannot judge someone from one match." I agree and that's why the ranking will be from scores from atleast 3 to 4 matches. That will give about 30 to 40 stages to judge a person. When I figure out the relative strength of a given match I don't use just one shooter - I use many as possible. My ratings are better with the more shooters I have to compare. Eric G? Eric Grauffel? The French guy? He doesn't shoot in american matches so he's not relevant. The 100% in my system is based on the best skills of people shooting in American matches. (Besides - I would like to see Grauffel beat Robbie with iron sights ) "He is better than Jerry in about the same amount that Jerry is better than us!" Whooaaa there. I think that you are greatly underestimating the burners ability. "If you base everything on one match, well, that's fine but that's really what we do now, and kind of what your system does." No, not just one match. The nationals are just the start of a system. I take the nationals and from them compare to the area matches and from there I go to the other major matches. I just used the nationals as a way to simply explain my system. It's actually more complicated. "98% to 89% would be the difference of 9th to 2nd place at most majors..." I'm not sure what you are trying to say there. The third number for the shooter represents how much easier this match was for the shooter. I take the average of these numbers then create a modifier or match handicap. For that match it was 68.65%. What I then do is multiply the match handicap (.6865) times their match percentage -(89.81*.6865=61.65% for ty43660 ) and (98.16*.6865= 67.38%) and that is what their relative skill level was on that day which is based on a prediction of what the 100% would be if the best shooters were at that match. Link to comment Share on other sites More sharing options...
BSeevers Posted November 5, 2002 Share Posted November 5, 2002 Dowter Thanks for keeping the info coming. We will need it. I see more clearly now. Jury is still out for me but I am adding it up. Well, I am as sorry as you are about US performance at World Shoot. Eric G has won it the last 2 times SOLIDLY. He could have zeroed the last stage and still won. At those guys level that is UNHEARD of. Thats like 130+ match points!! I always pick Jerry/Todd/Rob to win. They are the best. But that is the kind of beating Jerry would hand us if he came down and shot Tristate or OH state. I think Eric G could shoot a real gun also, but maybe we will never know. Link to comment Share on other sites More sharing options...
Jake Di Vita Posted November 5, 2002 Share Posted November 5, 2002 This may have already been answered, but I never saw it. Hypothetically, I shot 3 major matches this year. Now all the people that I shot against in my division at each major match had never shot a major match before, which is definetely plausible since I shoot mainly Production. How would I be ranked if you have nothing to compare my skill to? Really looking forward to this, keep up the good work. Link to comment Share on other sites More sharing options...
Dowter Posted November 5, 2002 Author Share Posted November 5, 2002 "Now all the people that I shot against in my division at each major match had never shot a major match before" If that is the case then that division at that match will have to be thrown out of the ratings. BUT the good news is that that will probably not happen to more than 1 or 2 divisions in all of the major matches this year. Link to comment Share on other sites More sharing options...
BamBam Posted November 24, 2002 Share Posted November 24, 2002 How about something easier as a starting point. I'd like to see a USPSA ranking using the current classification numbers in a single list from top to bottom for placement... throwing out any name that hasn't shot at least one major match this year to weed out the semi-retired. That would yield a top 500 in the nation for fun sake, and be a whole lot less work with no fancy math. How about that? Link to comment Share on other sites More sharing options...
Dowter Posted November 26, 2002 Author Share Posted November 26, 2002 Quote: from BamBam on 3:58 pm on Nov. 24, 2002 How about something easier as a starting point. I'd like to see a USPSA ranking using the current classification numbers in a single list from top to bottom for placement... throwing out any name that hasn't shot at least one major match this year to weed out the semi-retired. That would yield a top 500 in the nation for fun sake, and be a whole lot less work with no fancy math. How about that? This won't be a USPSA project. They're not going to be associated with it. And even if classifiers were used... there would still be problems. Please read the rest of this thread about that. The math isn't that fancy. It's a very simple computer algorithm that's no more complicated (once you understand it) as computing USPSA match scores. I guess this is as good as any other time to update on the progress of this program. I'm still hoping on getting a list of match results from Area 6 that have the USPSA numbers associated with the scores. Apparently the hang-up is a concern of privacy which I can't understand since all I'm asking for is the same information that was published with EVERY other major match. Oh well... I'm still hopeful I'll get this info. Link to comment Share on other sites More sharing options...
j1b Posted November 26, 2002 Share Posted November 26, 2002 Dowter Based on the response from this thread - a TON of interest. I have always liked the idea of rankings before going into a match. Somehow it spices things up. This sounds cool though. I'm excited to see results! JB Link to comment Share on other sites More sharing options...
Dowter Posted December 19, 2002 Author Share Posted December 19, 2002 UPDATE! OK, Area 6 is going to have to be thrown out. The reason that they don't have uspsa numbers associated with names is that they are using IPSC software and not USPSA. There is no file that has the names associated with the uspsa numbers. The good news is that that was the last hurdle and everything else will be smooth sailing. Target date for the first ranking - January 1, 2003! Link to comment Share on other sites More sharing options...
Nolan Posted December 19, 2002 Share Posted December 19, 2002 2pm Jan. 1, 2003 right? LOL! Link to comment Share on other sites More sharing options...
Dowter Posted December 23, 2002 Author Share Posted December 23, 2002 Well, funny things happen. It looks like Area 6 is back in for the statistics. Not only that but a few other major matches that I was planning on throwing out are back in too. It seems that the person who did the scoring for those matches that I was going to throw out, has the records that I was looking for and is going to send them to me. So basically the only match that will be thrown out the the Crazy Croc match. (that match had other problems...) I got these match results by the cooperation of the good people at USPSA - Charles Bond - Area 6 Director Dave Thomas - Executive Manager Linda Chico- Score Keeper of many major matches Many thanks go out to these three. Also new news. I have a website for the rankings. I just purchased http://www.topguns.info . Don't bother typing it in yet. It probably won't be running til about friday. Currently there are 2976 shooters in the database. I should get a couple hundred more when I'm sent those last matches. Almost all the software has been written to do the rankings. I see no problem for the rankings to be out by Jan.1st Link to comment Share on other sites More sharing options...
BlitzKreigBrad Posted December 26, 2002 Share Posted December 26, 2002 The idea of a ranking system outside of the classification system is terrific. One, it gives us an idea of who is active and who is not. Two, it puts our names out in an easily obtainable format that anti-gunners can use. Three, it gives pro-gunners an easily obtainable list of active shooters. Regardless, I noticed that Limited 10 shooters weren't listed. Specifically, and I deserve to catch hell for this, I wasn't listed despite having shot Area 3, Area 5, the Factory Gun Nationals, and the World Shoot. Good luck, I think it is a wonderful concept Link to comment Share on other sites More sharing options...
Dowter Posted December 26, 2002 Author Share Posted December 26, 2002 If your Brad Bolz, you're ranked 38th. Check out the lim-10 list. http://www.brianenos.com/cgi-bin/ikonboard...81&start=10 Link to comment Share on other sites More sharing options...
lynn jones Posted December 27, 2002 Share Posted December 27, 2002 Quote: from Shooter Grrl on 9:55 pm on Nov. 2, 2002 I'm interested, but only because I believe it will prove once and for all that the classification system does work :-) kath, do you really think the classificaion system works when someone is ranked #1 or #2 in the top 20, when they haven't shot in a couple of years? :] lynn Link to comment Share on other sites More sharing options...
lynn jones Posted December 27, 2002 Share Posted December 27, 2002 dowter, would you be using the classifiers shot in the current year in any of your calculations? i may have missed it in the thread. lynn jones (Edited by lynn jones at 1:36 pm on Dec. 27, 2002) Link to comment Share on other sites More sharing options...
Dowter Posted December 27, 2002 Author Share Posted December 27, 2002 Classifiers play no part in the rankings. Link to comment Share on other sites More sharing options...
Shooter Grrl Posted December 27, 2002 Share Posted December 27, 2002 Quote: from lynn jones on 1:18 pm on Dec. 27, 2002 Quote: from Shooter Grrl on 9:55 pm on Nov. 2, 2002I'm interested, but only because I believe it will prove once and for all that the classification system does work :-) Looks like my hypothesis is getting proven My national classification percent is 52, and on the top 500 list - i'm at 52%. OH yeah, my nationals % was 52 also. I'd say the classification system works just fine! IF the peope using it are honest - that's the part NO ONE can control! kath, do you really think the classificaion system works when someone is ranked #1 or #2 in the top 20, when they haven't shot in a couple of years? :] lynn Link to comment Share on other sites More sharing options...
Flexmoney Posted May 3, 2012 Share Posted May 3, 2012 Bumping this thread up too. Another thing that I was thinking of implementing is a star rating. You know how people say "I'm an A-Class shooter" They could also (or instead) say "I'm a 3 and half star shooter" The nice thing about this sytem is getting 5 stars is a Herculean task. There will be no paper 5-Star shooters. My guestimation is that there will be only 8 5-Star shooters when the list comes out. Here's how it will work. Stars Score 5 95% 4.5 87.5% 4 80% 3.5 72.5% 3 65% 2.5 55% 2 40% 1.5 25% 1 00% It might look odd at first but when the rankings come out it should give a nice distribution that makes sense. I also made it very easy not to get the lowest rating - 1 star. Why discourage new shooters. I might differ from Dowter here a little bit. Instead of some randomly (pre)decided cut off (72.5%...80.0%...87.5%), why not just run a distribution and see where things fall? Link to comment Share on other sites More sharing options...
vluc Posted May 3, 2012 Share Posted May 3, 2012 (edited) Too bad Dowter isn't around anymore to pursue this. Either he moved or no longer shoots but I've not seen hide nor hair of him for many, many years. Edited May 3, 2012 by vluc Link to comment Share on other sites More sharing options...
speedrrracer Posted May 8, 2012 Share Posted May 8, 2012 Too bad Dowter isn't around anymore to pursue this. Either he moved or no longer shoots but I've not seen hide nor hair of him for many, many years. Niftybytes probably hired him to work on Practiscore in order to squash his public efforts Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now