(Note: Editorials and opinion pieces will now be categoried separately)
This past season, I've typed the phrase (when recapping a women's meet), "the highest score since 2004", too many times to count. Team after team recorded new all-time high scores, or achieved their highest mark since 2004. Why is the year 2004 significant? Well, it was the season before the last time the coaches and judges collaborated to recalibrate the scoring system.
In 2004, the 12th ranked team had a Regional Qualifying Score (RQS) of 196.99. The following season, the 12th ranked team finished the season with a 195.715. This past season, the RQS of the 12th ranked team was a 196.615. This was the highest 12th place RQS since, you guessed it, 2004. In fact, this year's RQS is close to the score achieved in 2003, the year before the peak. So does this mean we are one year off from a "reset" of the scoring system? I certainly hope not, but we may be in for a record year in 2014-2015.
It is easy to say the code is too easy. However, changes in the code will be worthless unless enforcement of the code is maintained. As the Super Six Finals clearly demonstrated, scoring of the existing code is lax. Clear and obvious deductions were not taken during the Super Six finals, especially on floor. The skill level of the women competing at all levels has risen. Difficulty and composition now regularly exceeds the minimum requirements, even on teams of Rank 40 and below, and outside the ranks of the Division I teams. Yet if execution errors are not properly docked, any changes to the underlying requirements will be meaningless.
No one is advocating for a strengthening of the code to FIG levels, like in the case of the men. However, the current code is in many ways easier than the code used in 2006. The code has been static for a considerable length of time, while the ability of the athletes at all levels has risen. There is always the legitimate fear that strengthening the code will lead to more injuries. However, the emphasis on high levels of execution combined with rising skill levels makes this worry moot, as long as any changes to the code are minor. In fact, as mentioned previously, gymnasts at all levels are already going beyond the minimum in an effort to stand out among a sea of similar routines.
The rising parity in the NCAA has been aided by the static code, and further distorted by uneven and lax judging in certain parts of the country. Thus, there is a large contingent of the NCAA that would favor the status quo, to create more parity and to allow lower ranked teams a better chance to break through to Regionals and Nationals. Coaches are often measured on achievements of their teams, and higher scores and higher post-season finishes can lead to higher pay and improved recruiting prospects. People will also argue that this improves the popularity of the sport at all levels, in all parts of the country. I would also argue that it can swing too far in the extreme, to a point where athletic achievements are cheapened and subjectivity and chance rule the day. A balance is needed, and the data clearly shows we are a season or two from a required reset.
So what changes need to be made? First off, evaluation of the existing code elements needs to be implemented. Additional tools need to be provided to the judging community in terms of enhanced standards for execution and composition. By providing more structure to the NCAA Code Modifications, subjectivity and variation can be reduced. And statistical measures should be explored to provide judges (private) feedback on how often then deviate consistently from the mean score. Lax scoring is becoming a problem at the USA Gymnastics Junior Olympic level as well, so the problem is not restricted to the NCAA.
So what about the code itself? A simple solution would be to just adopt the USAG Junior Olympic L10 code. Today's NCAA code is just a simplification of the USAG JO L10 standards, with a few exceptions that may also encourage or reward certain skills. This simple solution would provide consistency between the club world and the NCAA. However, this may be too drastic of a solution for the NCAA to accept. The USAG JO L10 standards not only have different skill values, but the requirements in each event are more rigorous, creating the potential for a wider range of scores based solely on skills and composition. This type of level setting might be too drastic for coaches, athletes and fans. In addition, the NCAA code tends to de-emphasize dismounts, to reduce the risk of injury during the much longer and more rigorous NCAA season.
Here are a few smaller changes that could help strengthen the underlying code, and give the judging community a more structured and objective measure to create separation between the athletes.
Vault
On vault, the USAG JO code awards a bonus 0.1 to certain more difficult 10.0 start value vaults, effectively making them 10.1 value vaults. This is a messy solution to the vault code that simply needs to be reset to lower start values. Despite this view, I still don't think we're quite ready to lower the start values in the NCAA. Most teams in the top 50 have at least half of their lineup doing a Layout Yurchenko with a full or half twist. There is very little variety and the vault has become almost compulsary. And yet, we still don't see that many of the more difficult vaults. The added risk of training higher value vaults may result in more injuries, something that no one wants to see. We still don't seem to be at the right trigger point, where the devaluing of these vaults to 9.9 is yet warranted. For now, the answer may continue to be a stronger emphasis in judging execution, technique, form and amplitude.
One simple enhancement could help the judges separate routines. Marking the landing area with tape for distance and direction could help provide a more consistent standard by which direction and distance are measured. It would also provide fans a quick way to see why a vault was deficient in these two areas. Height above the horse, body position and other execution areas, however, would still have to be propoerly evaluated, of course. Would it emphasize height over distance? Possibly, but these deductions must also still be taken.
Bars
On the uneven bars, the NCAA code is actually easier than it was several seasons ago. The code modifications vary significantly from the USAG Level 10 standards, in particular with respect to the level of difficulty required for the highest score. The NCAA code provides only one tenth of one point to separate routines of high difficulty from those that meet a minimum standard. This minimum standard is two C releases or a D and a B plus a C dismount. To get back the remaining tenth, the so called "up to the level" compositional deduction, the athletes have a number of ways to fulfill it:
1. Through a C+C+D (any order) or D+D release combination*
2. Through a C+C+D (any order) or D+D pirouette combination*
3. Through a combination of release and pirouette, C+C+D or D+D*
4. A dismount sequence, C+C+D (any order) or D+D
5. Two D releases
6. Any two "E" skills
Only one skill in 1 to 3 (marked with an asterisk) has to contain a turn or a release to qualify. The athlete must also have a D release and a D or C+bonus dismount.
Nearly every lineup competitor in the Top 50 teams had no issue meeting this diluted minimum requirement. Thus, this "up to the level" compositional requirement has become essentially worthless, as the ability of athletes at all schools has improved. In addition, we've seen routines that have become stripped down of elements to form a minimum set of skills necessary to avoid these compositional deductions while still getting their full 0.5 of bonus. This has resulted in relatively "stock" routines being competed. There are provisions in the code to deduct for weaknesses in routine composition, but their enforcement can be spotty and the maximum deduction is just 0.1 points under the NCAA code.
Prior versions of the code provided 0.2 points to separate the gymnasts. The current USAG Level 10 Junior Oympic code uses a similar 0.2 to separate the routines based on the choice of release elements: with two Ds on one end of the scale, and a B+C on the other end. This is a broader range with a lower minimum requirement, but one that allows for more separation between the athletes. In addition, the Level 10 JO code has additional, broader compositional deductions that are intended to reward routines with a variety of skills shown.
So what can be done? The NCAA has moved to a higher minimum requirement and has focused on encouraging variety. However, instead of a single 0.1 flat deduction, the NCAA could restore a sliding scale and increase the range to 0.2 tenths, like the USAG L10 JO code. Setting the high end at two non-connected D releases could help separate the routines and add variety and excitement to the event. With a C&C at the low end (0.2 deduction), the D&C would fall to the midpoint (0.1 off). An E&C could qualify for 0.05, while two Ds would earn no deduction. In addition, a special incremental deduction of 0.05 for the lack of a single bar release would ensure the most dynamic routines are separated.
Providing a small deductions (0.05 each) for routines that have at least one single bar skills (Group III-underswings/clear hips, V-forward skills, and VI-Stalders/Toe Ons) would help encourage more variety in skill selection, while maintaining the existing 0.05 compositional deduction for "distribution of elements" could help ensure that difficult elements are placed thoughout the routine.
How would this play out in real life? If a routine otherwise had a "variety skill" and a qualifying dismount, here's how they would max out: A routine with an overshoot to handstand connected to a toe-up to high would start at a 9.85 (9.9 minus 0.05 for no single bar release). A routine with a Tkachev and a hecht from low to high would start from a 9.9. A routine with a Pak salto and a Shaposhnikova would qualify for a 9.95 (10 minus 0.05 for no single bar release).
Beam
On the beam, there has been fewer issues with separating the athletes. However, as the ability of the athletes has increased, the frequency of "no wobble" routines has increased. The teams have responded by stripping down their routines to the bare base elements, with a focus on execution of the few select elements performed. With less obvious faults, judges are separating based on pauses, rhythm, leap amplitude and position, and other faults that are less obvious to casual fans. This can result in scores that are tough for fans to understand, and can also result in increased score disparity as these execution issues are the ones most likely to not be evenly docked. The resulting routines look stock, with many of the same elements used again and again.
Inherently, however, the solution does not seem to lie in the USAG Level 10 code. That code, while favoring more difficulty, is less explicit than the NCAA code. It does, however, provide for more separation between the athletes. Some simple solutions instead could strengthen the NCAA code. Certain dance skills that are increased in value in the NCAA should be restored to their JO value. Devaluing the dance skills will not greatly contribute to injury risk, unlike, for example, requiring harder dismounts. The leading example in the NCAA code is the straddle jump 1/4 (rated a "C" in the NCAA and "B" in the JO code). This skill is often poorly executed and provides too much bonus when used in combinations with other C skills, such as the 0.2 points in bonus earned for the popular switch split to straddle 1/4. The other example is the ring leap -- a "D" in the NCAA, "C" in the JO code.
The NCAA code could be strengthened by both devaluing the back layout stepout AND require a "C" no-hand acro element in the flight series to meet "up to the level". The back layout is rated a "D" in the NCAA to encourage its use. Simply devaluing it is not enough; the flight series requirement should be strengthened to encourage its use (by requiring a C salto or any E in series) and to allow for separation of the top routines. Removing the D skill bonus will simply force the athletes to add new combinations and other elements to make up for the lost tenth.
The current NCAA code provides for a flat 0.1 deduction for "up to the level". This requires the athlete to add another D acro skill or "E" dance skill, if their acro series does not contain bonus (excepting the back handspring to layout stepout). Most gymnasts are already performing a qualifying series or have the second D salto (or "E" dance skill). This "up to the level" compositional deduction could be further strengthened by always requiring a second D acro skill or "E" dance skill in the routine, regardless of the composition of the flight series. Certain front to back flight series, such as the front aerial to back handspring, are becoming popular because they are easier for some athletes to hit and they provide a bonus that allows the athlete to leave out a risky salto or "E" dance skill later in the routine.
Floor
Floor is one event where we see a high range of scoring variation (from judging panel to judging panel) and an increasing level of athletic ability on display. However, the NCAA code is currently easier than it was in 2006. One simple solution exists: modify the "up to the level" compositional deduction.
Currently, the NCAA "up to the level" deduction provides for a flat 0.1 point penalty if the routine does not contain a "D" salto, a series with a C salto, and a dismount with a C bonus combination or "D" salto. This requirement has proven quite achievable for nearly every lineup athlete in Division I and II. When this happens, the deduction loses its utility as a tool to separate the routines. Despite guidance to the contrary, judges begin to use more subjective measures to separate the routines. Routines with higher difficulty are graded more leniently on other criteria, while routines with minimum difficult receive maximum scrutiny. The problem with this approach is that the end result increases the variation in evaluations done by different judging panels.
How can we remove some of this subjectivity and "mitigation" from the code? Taking cues from the current JO L10 Code and the 2006 NCAA code modifications, a full 0.2 sliding scale could be implemented to separate routines based on the "up to the level" compositional deduction. Using a sliding scale would allow routines to be separated based on content. At the high end, a routine that featured an E salto or D in combination (except for the Rudi), a series of at least a C in combination, and a dismount with a "D" salto or higher would get no deduction. A routine that featured at least a D salto, C bonus series and a C bonus dismount would get a 0.1. A routine with only a C dismount and two salto series, just meeting the Special Requirement, would still get docked 0.2.
As an additional compositional deduction, a routine without a D+ double flipping or double+ twisting salto could get a special, incremental compositional deduction of 0.05. Similarly, routines that achieved the "up to the level" compositional deduction in only two tumbling passes would get a deduction of 0.05. A two pass routine with a front stepout to back 2 1/2 twist and a Rudi dismount would max out at a 9.95 (10 minus 0.05).
At the high end, a routine with a front to double tuck, whip-half to front full, and 2 1/2 twist dismount would start from a 10. However a routine with a double pike, front full punch pike and Rudi dismount could score a max of 9.95. A similar routine with a double tuck, back 1 1/2 to punch layout, and a front full-punch front dismount would start at a 9.9. At the lower end, a routine with a Rudi, front full punch front, and back 1 1/2 punch pike would top out at a 9.85 (9.9 minus the additional 0.05). A routine with a double tuck and a front full punch layout (in two passes) would also max out at 9.85.
Encouraging more advanced tumbling, however, runs the risk of increasing the injury rate. However, most athletes in the top 50 teams are already going well beyond the minimum requirements. And, because the tightness of the range of final scores already provides a steep penalty for major mistakes, gymnasts are not attempting skills they have not largely mastered. Falls that occur on the floor are usually due to mistakes in concentration or timing, not a consistent reflection of marginal mastery of a skill. Providing a greater range of compositional deductions with clearly called out criteria will help reign in the unevenness in scoring and provide for better separation of routines.
Will There be Change?
Will there be change? It appears, sadly, that this is not likely. We may still have to live with the current code for another season. However, after another record year of scores we could finally be in for a reset of the code in 2006.