The Chronicle of the Horse
MagazineNewsHorse SportsHorse CareCOTH StoreVoicesThe Chronicle UntackedDirectoriesMarketplaceDates & Results
 
Page 1 of 2 12 LastLast
Results 1 to 20 of 27
  1. #1
    Join Date
    Aug. 14, 2000
    Location
    Clarksdale, MS--the golden buckle on the cotton belt
    Posts
    19,084

    Default My dressage coefficient spreadsheets on Google Docs

    I've posted them so anyone can take a look at the effect of the FEI dressage coefficient on 4* competitions in 2010 and 2011. The links are here:

    https://docs.google.com/open?id=0BzD...BhNTg2N2Q1ZDBj
    https://docs.google.com/open?id=0BzD...I2ZDA1ZjA4NjBk
    https://docs.google.com/open?id=0BzD...I1MTJjYjJkNmYy
    https://docs.google.com/open?id=0BzD...Y0MmRlM2IwY2Nh
    https://docs.google.com/open?id=0BzD...EzYjZlNDE1ODk5
    https://docs.google.com/open?id=0BzD...M3YTU0NmQ3ZDA1
    https://docs.google.com/open?id=0BzD...Y4ZDNhNmM5ZTEw
    https://docs.google.com/open?id=0BzD...E1NWY4NWJiM2Iy
    https://docs.google.com/open?id=0BzD...ExY2Q5MWUxYWYw
    https://docs.google.com/open?id=0BzD...MwMWViZWE4MjQ5
    https://docs.google.com/open?id=0BzD...A2ODg1YjI2NDFi

    They really do show that the FEI's added penalties change the results from those that would exist if only the dressage judges' ride evaluations and current SJ and XC results were decisive.

    If you have trouble accessing them, please let me know.
    "I'm a lumberjack, and I'm okay."
    Thread killer Extraordinaire



  2. #2
    Join Date
    Nov. 18, 2004
    Location
    Catonsville, MD
    Posts
    6,919

    Default

    I can see your spreadsheets, but would like a fuller explanation of what they are about? What coefficient are you taking out?
    I tolerate all kinds of animal idiosyncrasies.
    I've found that I don't tolerate people idiosyncrasies as well. - Casey09




  3. #3
    Join Date
    Aug. 14, 2000
    Location
    Clarksdale, MS--the golden buckle on the cotton belt
    Posts
    19,084

    Default

    The spreadsheets address how the dressage score is converted to penalty points. The USEA and BE (not sure about other countries) take the average percent score that the dressage judge (or judges) give and subtract that from 100. The difference is the penalty point score for the dressage phase.

    The FEI has a different way of getting to the penalty points for dressage. They take the average percent score for the ride, subtract it from 100 and then multiply the difference by 1.5. What that means is that you get half again as many penalty points in an FEI event as you would in a USEA event. Those penalty points can make a big difference at the four star in level in where a rider places. The ranking of riders in the dressage phase is the same with or without the 1.5 multiplier; the dressage score is the same. Only the number of penalty points for the dressage ride is affected.

    What my spreadsheets show is the effect of the extra FEI imposed penalty points on the final placings. In my opinion, it's very, very significant and affects many placings. I also believe that it devalues XC and SJ performance. Just as an example, Mary King went double clear in both XC and SJ at Pau this year, and if her dressage penalty points had been calculated from her dressage score without the coefficient/multiplier, she'd have finished second instead of 4th. Mark Todd would not have won Badminton this year without the extra help he got from the coefficient applied to his competitors.

    You have to remember that the coefficient seems to work geometrically. If I score 75 in dressage, my penalty points with the coefficient would be 37.5 (25+12.5). If I score 70, my penalty points are 45 (30+15) The second ride is 2.5 penalty points (or time faults in XC or SJ) worse than the dressage judges said it was. If I score 60%, my penalty points are 60 (40 + 20) or 5% more than the 10% dressage score difference.
    Last edited by vineyridge; Nov. 8, 2011 at 09:02 PM.
    "I'm a lumberjack, and I'm okay."
    Thread killer Extraordinaire



  4. #4
    Join Date
    Nov. 18, 2004
    Location
    Catonsville, MD
    Posts
    6,919

    Default

    Thank you, great explanation.

    So after reading that, a person has to ask, what the heck is the multiplier FOR?
    I tolerate all kinds of animal idiosyncrasies.
    I've found that I don't tolerate people idiosyncrasies as well. - Casey09




  5. #5
    Join Date
    Apr. 2, 2009
    Location
    North Carolina
    Posts
    5,418

    Default

    Because FEI is nutballs.



  6. #6
    Join Date
    Mar. 1, 2003
    Location
    Happily in Canada
    Posts
    5,022

    Default

    Asking out of complete ignorance: has this been brought up to the PTB in the NFs as well as the FEI? Also, wonder if there is knowledge or a consensus amongst FEI riders?

    Just wondering if all your data was provided, whether the riders would prefer the co-efficient removed, and what it would take to do so.

    (I am in favour of removing it, thereby removing some of the weight of dressage).
    Blugal

    You never know what kind of obsessive compulsive crazy person you are until another person imitates your behaviour at a three-day. --Gry2Yng



  7. #7
    Join Date
    Aug. 14, 2000
    Location
    Clarksdale, MS--the golden buckle on the cotton belt
    Posts
    19,084

    Default

    The Canadian NF mentioned it as an issue in their formal response to the FEI Eventing Committee's questionnaire about possible changes in the Rules for 2013.

    I've sent all my spreadsheets to Robert Kellerhouse, who is on the FEI Eventing Committee. And I've sent some of them to Brian Sabo and some of the USEA officials. Francis Whittington in Britain, who is the leader of their Event Riders Association has seen them.

    I just hope all of these PTBs are at least thinking about the issue.

    There might have been a reason for the coefficient in long format to increase the importance of dressage. But it hardly seems necessary in short format.
    "I'm a lumberjack, and I'm okay."
    Thread killer Extraordinaire



  8. #8
    Join Date
    Oct. 4, 2008
    Posts
    1,144

    Default

    It certainly weighs in favor of countries that have very strong dressage. I suspect this is part of the reason that we are seeing such a large amount of warmbloods in the sport now. While a TB can be very nice moving, and have suspension, it's less likely then the warmbloods. I would really like to see this changed, but think it is pretty unlikely. the FEI prolly has a vestedinterest in keeping things as they are. Frustrating as it is.



  9. #9
    Join Date
    Aug. 14, 2000
    Location
    Clarksdale, MS--the golden buckle on the cotton belt
    Posts
    19,084

    Default

    Actually the history is that the coefficient came into being in about 2000, long before short format. My understanding is that dressage used to be scored completely differently. Instead of using a percent score from the judges, they used the actual point scores from the test. That's when the FEI used a (I believe) .6 coefficient to determine the dressage score. Example: if a test has 250 points total available and the rider scores 200, (Janet, is this right?) 200 was multiplied by 1.6 (or .6).

    1.6 might be a Canadian peculiarity.
    "I'm a lumberjack, and I'm okay."
    Thread killer Extraordinaire



  10. #10
    Join Date
    Oct. 4, 2008
    Posts
    1,144

    Default

    I am REALLY confused now. If they use to multiply by 1.6 and now it's 1.5..... Or is it what they multiplied? Am very confused. Sorry.



  11. #11
    Join Date
    Aug. 14, 2000
    Location
    Clarksdale, MS--the golden buckle on the cotton belt
    Posts
    19,084

    Default

    In the old days they used to multiply the raw points; now they multiply the percent. If you got a seven on one movement and a six on another, that was 13 points and they'd use the old coefficient. Now they would average 7 and 6 and you'd get a 6.5 x 1.5 (Just an example, since the whole test is used.)
    "I'm a lumberjack, and I'm okay."
    Thread killer Extraordinaire



  12. #12
    Join Date
    Oct. 17, 2009
    Location
    Unionville
    Posts
    1,071

    Default

    Viney: Thank you so much for explaining the scoring and coefficient effects. And you have certainly done a Yoeman's job of compiling the statistics! And I'm glad to hear that you sent them to Kellerhouse....

    But I have a remaining question: why the switch to the 1.5 coefficient? Is this just another mysterious FEI decision? I just wish I knew the explanation for it. The only thing I can think of is that they want dressage to be more influential. Am I missing something?
    "We can judge the heart of a man by his treatment of animals" Immanuel Kant



  13. #13
    Join Date
    Aug. 14, 2000
    Location
    Clarksdale, MS--the golden buckle on the cotton belt
    Posts
    19,084

    Default

    From what I have gathered from Mr. Della Chiesa and others, they THOUGHT the 1.5 coefficient would give the same results when applied to the percent score that the .6 coefficient did when applied to the raw points. Another reason that's been given is that they wanted to "spread the competitors out" more, so the scores wouldn't be so tightly bunched. My impression from the spreadsheet data is that the latter hasn't happened. You have groups of horses that are tightly bunched, then a gap, then another group.

    Personally, I think that they DID want to make dressage more influential in long format. I went to Rolex in 1999 and the dressage stank. Went again in 2002, and the dressage was incredibly better. Increasing dressage influence did make sense, since in long format there were really six phases to score. Now there are only three and the scoring hasn't been modified to reflect that. To me, the dressage coefficient effects make dressage performance worth almost 49%+ (33 1/3 plus half of 33 1/3) of the total performance.

    That's why Mary King's double clears at Pau were "devalued."
    "I'm a lumberjack, and I'm okay."
    Thread killer Extraordinaire



  14. #14
    Join Date
    Feb. 4, 2004
    Posts
    2,677

    Default

    Per Wikipedia in 1998 the max dressage points were 250 and the coefficient was .6

    That would mean a 70% would be a 45 and a 60% would be a 60.

    Currently with 100 good points and a coefficient of 1.5 a 70% would also be a 45 and a 60% a 60.

    So I'm guessing it was the change in tests that motivated the coefficient.

    The short format would be a good reason to get rid of the coefficient, but I'm not sure if that was considered?

    The weighting of dressage tests in US HT has varied a great deal even since I started eventing. Back when we used USDF tests (like training, 1st level) scores were much higher--it was not unheard of to break 100 on 1st 3 @ prelim . . . ). I believe that those used .6 as the coefficient. Then the BHSA tests were much lower, closer to our current scoring system. And our current tests are the lowest yet--we see scores that are not much more than a single xc stop, or 4-5 rails, or a minute slow xc.


    Personally I hate dressage and would love to see it less influential, but I'd be curious what other national HT do? The obvious answer would be to weight our dressage more in line with the FEI, but nobody wants that!



  15. #15
    Join Date
    May. 14, 2011
    Location
    Ireland
    Posts
    29

    Default THE LADY IS RIGHT

    I have no doubt that Vineyridge raises a hugely important issue here with great clarity and hours of solid homework!

    It is a fact that the current multiplying factor does increase the influence of the dressage, therefore it has to decrease the influence of the jumping phases. I do not believe this was the original intention of introducing the current multiplying factor...it just wasn't sufficiently thought through and understood.

    However taking away this multiplying factor would bring the scores closer together which would then give a show jumping knock even greater influence...probably too great an influence. In which case the faults for hitting a fence would have to be reduced to 3 rather than the current 4 faults.

    The
    www.WilliamMicklem.com



  16. #16
    Join Date
    May. 23, 2006
    Posts
    5,487

    Default

    Quote Originally Posted by William Ireland View Post
    I have no doubt that Vineyridge raises a hugely important issue here with great clarity and hours of solid homework!

    It is a fact that the current multiplying factor does increase the influence of the dressage, therefore it has to decrease the influence of the jumping phases. I do not believe this was the original intention of introducing the current multiplying factor...it just wasn't sufficiently thought through and understood.

    However taking away this multiplying factor would bring the scores closer together which would then give a show jumping knock even greater influence...probably too great an influence. In which case the faults for hitting a fence would have to be reduced to 3 rather than the current 4 faults.

    The





  17. #17
    Join Date
    Aug. 14, 2000
    Location
    Clarksdale, MS--the golden buckle on the cotton belt
    Posts
    19,084

    Default

    I don't know how the Canadians apply their 1.6 coefficient or what they apply it to. I know BE uses the same system that the USEA uses--the dressage judges' average percent subtracted from 100.

    You have to pay 10 euros to access the Irish Eventing Rules. I don't read German, so have no idea what their dressage scoring looks like. Here is how the French do it:

    Art 9.1 - Calcul des résultats

    A - Dressage
    Le total moyen des notes est soustrait du maximum possible afin de convertir les points de bonification en points de pénalités. A ces points de pénalités est appliqué un coefficient de 0,5. C’est à ce dernier total que seront ajoutées les pénalités dues aux éventuelles erreurs. Les notes attribuées par chaque Juge sont additionnées. Le nombre de points de chaque Juge est additionné et le total est divisé par le nombre de Juges. Le résultat obtenu est retranché du maximum possible de points afin de convertir les points de bonifications en points de pénalités affectés du coefficient 0,5. Le résultat est tronqué à 2 chiffres après la virgule. C’est le résultat du test de dressage. Par ailleurs, pour des raisons pédagogiques, le résultat du dressage sera également affiché en pourcentage en suivant le mode de calcul défini ci-dessous pour les épreuves Pro 1 et plus.

    Dressage des épreuves Pro 1 et plus
    On calcule le pourcentage du concurrent en divisant le total des notes positives de chaque juge, moins les erreurs de parcours, par le maximum possible de notes et en multipliant par 100. Ce résultat est arrondi à deux décimales. La valeur obtenue est considérée comme note individuelle pour chaque juge. Le pourcentage moyen pour chaque concurrent est obtenu en additionnant le pourcentage de chaque juge et en le divisant par le nombre de juges, en arrondissant le résultat à deux décimales. Pour convertir ce pourcentage moyen en points de pénalité, ce total est soustrait de 100 et multiplié par 1,5 en arrondissant à une décimale. Le total final est le résultat en points de pénalité de la reprise.
    The second paragraph is the FEI procedure, which seems to apply to their UL National Competitions. But my French is VERY shaky, so perhaps someone could verify? And they obviously added the FEI scoring to something that they have done nationally at the lower levels. I'd be interested to know what that was, since my French can't deal with it.

    The Australians use the FEI procedure:
    EA National Eventing Rules 2011 Page 38
    The good marks from 0-10 awarded by each judge to an athlete for each numbered movement of the Dressage Test together with the collective marks are added together
    deducting any error of course or test.

    For each judge, the mark as a percentage of maximum possible good marks obtainable is then calculated.

    This percentage is obtained by dividing total good marks of the judge (minus any error of course or test) by maximum possible good marks obtainable and then multiplying by 100 and rounding the result to two decimal digits. This value is then shown as the individual mark for this judge.

    Average percentage for the athlete is obtained by adding together the percentage for each judge and dividing by the number of judges always rounding the result to two decimal digits.

    In order to convert average percentage into penalty points, this must be subtracted from 100 and multiplied by 1.5. The result, always rounded to one decimal digit, is the score in penalty points for the test Calculation of Scores
    New Zealand--this one I'm having trouble accessing.
    "I'm a lumberjack, and I'm okay."
    Thread killer Extraordinaire



  18. #18
    Join Date
    Mar. 10, 2006
    Posts
    132

    Default

    The FEI eventing committee proposed getting rid of the coefficient throughout 2005-2007; it was shelved based on NF feedback (which NFs for/against is not public info but shouldn't be that hard to find out since most of them don't even respond to the FEI anyway and there's at least one very obvious suspect in the anti camp!). The proposal was widely discussed and publicized via the normal public communications channels (Bulletin, press news, website) that existed before the information lockdown in 2008/9, after which point all this routine stuff between FEI and NFs got put in a restricted area of the site (recently opening up again via IEOC, they now have the minutes, NF feedback from last year's changes etc.)

    March 31 2005
    https://admin.fei.org/Disciplines/Ev...ax-Scoring.pdf
    Re: Proposals of new Eventing Scoring system for 2007
    ...
    2. Dressage:
    The result achieved for the dressage should be counted in total without the inclusion of a Multiplication factor. An 80% result should result into 80 plus points.
    March 16 2006
    https://admin.fei.org/Disciplines/Ev...ing letter.pdf
    In the first half of the year 2005 the FEI Eventing Committee proposed to you some considerations to change the scoring system.
    ...
    The cornerstone, which forms the basis of the scoring considerations, is an easier "sell" of our sport in public by having a simpler scoring system.

    The thoughts are to change the multiplying factor for the dressage phase from 1.5 to 1. This means that the score after dressage will be a percentage mark i.e. A good percentage of seventy percent (70%) would be a score of thirty (30.00).
    ...
    Our sport passed through a lasting change. The endurance aspect is less weighted than it was in the case of the former traditional long events.

    We would also like to ensure that the international scoring system be used nationally in the same manner. Insofar we are very interested in having a worldwide system which is synchronized hundred per cent. We feel this is a better development for all participants

    The sport has developed due to higher requirements in the dressage (implementation of the flying changes) and the requirement to ride at a more difficult level in the Jumping phase. We feel that this change in the scoring system would follow this evolution.

    The discussion we held in the last weeks and months on different levels it seems that many NF's are not sure whether the scoring system needs to be changed. If this is the case with your federation could you please discuss your argument in writing?
    FEI Bulletin 4/2006 November 1 2006
    https://admin.fei.org/Media/Publicat...letin4_000.pdf
    Scoring System
    In general, it was agreed that the current scoring was easier to understand for the public following the change of the Eventing format without steeplechase. The discussion is continuing on the coefficient for Dressage, considering that most national rules did not use a coefficient.

    The Committee would propose for the 2009 Rules revision the deletion of the Dressage coefficient and 0.5 time penalties / per second (instead of 0.4) for exceeding time on Cross Country and await /open the discussion with NFs.
    FEI Committee meeting report March 16 2007
    http://www.fei.org/disciplines/event...mittee-meeting
    The Committee considered the following changes to the rules for the 2009 revision.
    ...
    Scoring: The Committee agreed to propose for the 2009 rules revision the deletion of the dressage coefficient and 0.5 time penalties / per second (instead of 0.4) for exceeding time on Cross Country and await / open the discussion with NFs. Jumping time penalty to remain at 1 per second.
    More old meeting reports are still viewable here:
    https://admin.fei.org/Disciplines/Ev...s/Default.aspx



  19. #19
    Join Date
    Aug. 14, 2000
    Location
    Clarksdale, MS--the golden buckle on the cotton belt
    Posts
    19,084

    Default

    thanks very much for your post, Geneva. Sounds as if the FEI committee did try and reconsider the coefficient after short format came in. The history certainly helps with background.

    There is something very interesting in the press release about the 2007 meeting at Lausanne and the FEI Committee report of the same meeting. The Committee Report says that the elimination of the Dressage Coefficient would be proposed for the 2009 Rules, but the press release, which mentions all the other 2009 proposed changes in the FEI Committee Report, doesn't mention the coefficient. I have no idea if this is significant or not. Perhaps they thought it was too technical for the "public" to grasp.

    Since the FEI Committee was going to open discussion with the NFs after the 2007 Lausanne meeting, is there any way to access USEF documents on their response to this proposed change?
    Last edited by vineyridge; Nov. 8, 2011 at 09:56 PM.
    "I'm a lumberjack, and I'm okay."
    Thread killer Extraordinaire



  20. #20
    Join Date
    Mar. 10, 2006
    Posts
    132

    Default

    USEF is a closed shop and this whole issue of how the NFs go about managing their activities in the FEI just blew up in dressage in a big way and now the NFs are scrambling to cover their asses. But on something old like this I think if you asked the USEF chair you'd get at least a response and possibly an accurate picture of what happened. You also might get BS. But even BS can be useful.

    Normally there was just one FEI report sent out 3 ways, same text in the press news, website, then later the Bulletin (which was only published 5-6 times a year).
    The various versions of the 2007 mtg are here and it's the same in each as far as I can tell, so not sure what link you're looking at:
    http://www.fei.org/disciplines/event...mittee-meeting
    https://admin.fei.org/disciplines/ev...inutes_002.pdf
    https://admin.fei.org/Media/Publicat...007pdf_000.pdf p 50
    Last edited by Geneva; Nov. 9, 2011 at 12:16 AM. Reason: blond moment



Similar Threads

  1. Bedside docs, opinion on some bruising....
    By Bank of Dad in forum Off Course
    Replies: 7
    Last Post: Dec. 28, 2011, 10:04 PM
  2. Luhmuhlen after XC w/o dressage coefficient--Final
    By vineyridge in forum Eventing
    Replies: 5
    Last Post: Jun. 20, 2011, 09:50 PM
  3. Effect of the dressage coefficient Rolex 2011
    By vineyridge in forum Eventing
    Replies: 4
    Last Post: May. 2, 2011, 08:44 PM
  4. When did the FEI change the dressage coefficient?
    By vineyridge in forum Eventing
    Replies: 2
    Last Post: Mar. 24, 2011, 04:21 PM
  5. The FEI dressage coefficient--Pros and Cons?
    By vineyridge in forum Eventing
    Replies: 20
    Last Post: Dec. 19, 2010, 09:08 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •