Chuck Phillips updates his own Age-Graded Tables
Last April, I wrote about an Age-Graded Tables project by Chuck Phillips. Last week he gave me an update, writing: â€śWhile it still takes WMA five or so years to update their age-graded tables, I just updated mine in one week. On 19 November I downloaded 40 pages of published USATF records, entered the improved and new records in my database, made minor re-fitting adjustments to my standards, and posted them in their entirety on my tripod website on 26 November. It can be done properly in less than five years and the masters track community deserves better than it is getting in that regard.â€ť Which reminds me: WMAâ€™s official Age-Graded Tables have been available in digital form since August 2006, but a hard copy (in the form of a soft cover booklet) still has yet to be published. Whatâ€™s the deal?
Hereâ€™s an FAQ by Chuck on his new tables:
Published records were updated on November 19, 2007.
Refitted standards are current as of November 26, 2007.
What These Standards Are and What They Are Not
â€˘ Are: These standards set performance levels so that comparison of performances (whether within a single event, in separate or even multiple events) can be measured against a level of achievement common to all events. The purpose here is to produce a performance measuring system that is fair and equitable to competitors of all ages in all events.
â€˘ Are Not: These standards are neither the official performance standards nor the age graded tables of any official governing body of track and field (i.e., iaaf, usatf, wma, nmn, etc.). They are produced by the author to be used by individuals of all ages, scholastic coaches or other t&f practitioners for purposes as they see fit.
â€˘ Are Also: The standards include a documented body of running and jumping measurements useful to observation or study of the aging process for males and females ages 6 to 100 in regard to those physical activities.
What You Should Know About Performance Measuring Standards
There are three basic but significantly different kinds of performance standards applicable for track and field competitors of all ages.
â€˘ The simplest approach is to just use the existing records as the standards. The rationale here is that no one else has been able to better the record and therefore it is the standard. This presumes that no one record represents a better performance than any other record. Such â€śstandardsâ€ť are unsuitable for performance-measuring comparison purposes.
â€˘ A second approach involves curve fitting the records of all ages for a specific event to produce the standards for that event. The rationale here is that the performances of some ages produce records that are in fact better than others and the curve fitted standards will sort this out within the event. Such curve fitting is then continued until each event has its own set of performance standards. But there is a performance measuring limitation here in that the standards so generated only allow comparisons of performance within the same event. Since the events are curve-fitted individually and there is no mathematical curve-fitting relationship between separate events, noteworthy performances in one event cannot properly be compared with noteworthy performances in another event.
â€˘ The third approach involves curve fitting the records of track events in such a manner as to established standards that not only curve fit the records on the basis of all of the ages for each event distance, but also curve fit the records on the basis of all of the event distances for each age. This is of course a complicated approach involving the use of interrelated mathematical equations for the curve fitting process. This approach is actually a 3-dimentional curve fit producing a continuous surface of performance time standards over the axes of event-distance and age. The reason for using this approach is so performances between different events and different ages can be compared with the certainty that they are properly comparable and produce results having the fairness of mathematical impartiality and accuracy.
Note: This third approach is the one used for generating the performance standards here for all running and racewalk events.
Obviously any set of standards is arguable in one or more regards. Therefore, the standards here are subjected to a rigorous series of analytical evaluations to ensure their overall suitability and fairness, and the evaluated material is included with the standards as the basis for their validity.
Database Content for Standards Generation
The records database used to determine the standards initially included international records for all running, hurdle, walk and field events for men and women masters age participants. For boys, international and American records were used for running, hurdle, walk and field events. For girls, only American records have been used along with their international junior world records. USATF online sources are used now to update the records. Records were last updated November 19, 2007.
Who Can Use These Standards?
The standards are based on world record or best in world performances and as such are world record level performance standards. However the standards apply to all youth, open class and masters age performers (runs, hurdles, field events and walks; ages 6 to 100) of all ability levels through application of the performance level percentage concept.
For runners, hurdlers and walkers, the performance level percentage concept is that a runnerâ€™s performance level is the percentage obtained by dividing the event standard by the runnerâ€™s time for the event. Details in doing this are (i) the event standard used is that of the runnerâ€™s age, (ii) both the standard and the runnerâ€™s time are measured in seconds, and (iii) the decimal value of the division is multiplied by 100 to obtain the resultant performance level percentage.
For field events, the performance level percentage concept is that field event performance level is the percentage obtained by dividing the performerâ€™s distance for the event by the event standard. Details in doing this are (i) the event standard is that of the performerâ€™s age, (ii) both the standard and the performerâ€™s distance are measured in meters, and (iii) the decimal value of the division is multiplied by 100 to obtain the resultant performance level percentage.
How Are Performance Level Percentages Used?
INDIVIDUALS. Individuals can determine what event they are best at by finding the event for which they have the highest performance level percentage. They can also determine if they are basically a single event person because their other event performance level percentages are not nearly as good as that of their best event; or that they are a multi-event person because several other of their event performance level percentages are nearly as good as that of their best event.
Individuals can compare present performance level percentages with their percentages for previous years to determine whether they are improving with age or not as follows. Of course aging itself causes youth performance to improve and masters performance to decline, but the aging effect is discounted when comparing performance level percentages for an individual, and if the performance level percentage for an event is increasing for youths or masters the following year, then the performance ability for the individual is improving (due to training, increased experience, more self-confidence, perhaps improved health, etc.).
Keeping a logbook kind of record for selected performance level percentage results attained (date, event, PL%, comments of note, etc.) can be useful later when looking at the overall picture of what took place over the longer period of time. Beneficial self-evaluation may result.
COACHES. Youth group and scholastic coaches can prepare performance level percentage profiles for individual team members as described above in the â€śIndividualsâ€ť paragraphs that will benefit each team member personally and also help the coach to best distribute the talent for overall team purposes.
If performance level percentages are high for any individual, say in the 90% and above range, the performer is of noteworthy talent and has high potential for future development.
AWARDS. For purposes of determining awards or for recognition of performances of noteworthy accomplishment, members of the specified group (however large, small, inclusive or exclusive it may be) are compared by ranking them in the order of their performance level percentages â€“ regardless of age, or of event, or whether runner, hurdler, jumper or thrower.
Whatâ€™s In a Set of Standards and Age-Factors
Twelve sets of Standards and Age-Factors are available for selection, six sets for men and six sets for women. The six sets are for Track Running Events, LDR Running Events, Racewalk Events, Hurdle Events and Field Events. Each set has a table of performance standards and a table of corresponding age-factors, and also includes charts for evaluation purposes whose graphs plot various relationships demonstrating the suitability and fairness of the generated standards.
The time standard tables for running and racewalk events include standards in â€śhours-minutes-secondsâ€ť (h:m:s) format since that is the usual form, and also include the same standards in â€śsecondsâ€ť format as that is most useful when calculations are required to determine performance level percentage.
For hurdle events, the time standard tables list standards for ages and hurdle heights prescribed by the WMA/USATF hurdle specifications.
For field events, the distance standard tables list standards in meters for ages and implement weights prescribed by the WMA/USATF implement specifications.
FAQ (1): What are age-factors and how are they used? The age-factor table values for running, hurdling and racewalking events were determined by dividing the event open class time standard by the time standard for each of the other ages. Thus they directly indicate the amount of change of performance ability for youth and masters aged performers compared to open class, best-in-the-world performance ability. (This aspect is addressed under the heading of The Effect of Aging on T&F Performance.) In addition, the age-factors can be used to convert youth and masters age performances to their equivalent performance as an open class event. To convert a time run to its equivalent performance as an open class event, the time run (in seconds) is multiplied by the age-factor value listed for runners of that age for that event. The age-factor table values for field events were determined by dividing the distance standard for each of the other ages by the open class distance standard for the event. To convert a distance jumped or thrown to its equivalent performance as an open class event, the distance jumped or thrown (in meters) is divided by the age-factor value listed for competitors of that age for that event.
A caution here regarding age-factors: it is the value of the open class standard and not the open class world record that is used to determine age-factor values here. While most open class standards are the same as their open class world record some are not, and that is why the values of the open class standard rather than the world record apply when determining the equivalent open class performances that result.
In addition to the tables of standards and age-factors, backup charts are also included in each set that graph the various tables of data that are evaluated to ensure the suitability of those standards. A discussion of the graphs and their evaluation is presented later at FAQ (3).
Track Running Events
The following applies to the standards and age-factors generated for track running events for men and women, and also applies to the standards and age-factors generated for LDR running events and racewalk events.
Standards are provided for selected track running events most frequently included in sanctioned meets. While the standards reflect times run by world record level athletes, they also apply to youth and masters age runners of all ability levels through application of the performance level percentage concept. The performance level percentage concept is that a runnerâ€™s performance level is the percentage obtained by dividing the event standard by the runnerâ€™s time for the event. To do this, the standard and the runnerâ€™s time should both be expressed in seconds.
The primary purpose of performance level percentages is to facilitate comparison of performances. An individual runnerâ€™s own performances can be compared, or performances of many runners can be compared. Since the standards for all events are set at world record performance levels, comparison of performances for different events is valid. While standards are not established as such for lesser level or average runners, the â€śworld recordâ€ť level standards here are applicable to runners of all ability levels by using the performance level percentage concept.
FAQ (2): How are the standards for â€śrunningâ€ť type events generated so that all events are dependent on their neighboring events? What is not done is simply curve-fit each event separately as an independent entity â€“ the approach here curve-fits all events so that they are mathematically dependent on the records of their neighboring events. This is accomplished by using sets of equations in conjunction with each other that simultaneously curve-fit the records of all ages for each event distance and also curve-fit the records of all event distances for each age. The computer curve fitting is programmed to accomplish an iterative step procedure whereby the two different sets of curve fittings interact and fitting is continued in a back and forth manner until the same set of resulting â€śbest recordsâ€ť satisfies both sets of curve fitting equations.
The first step in developing these time standards for running events is to curve-fit the open class world records to obtain the open class time standards. This is done by curve fitting both the times of the open class world records and the running rates of the open class world records. The second step is to use the iterative curve fitting process described above to generate the standards for youth age and masters age runners.
FAQ (3): Why should I trust these computer-generated standards â€“ with computers, garbage in means garbage out? Good question with two answers. The first is that while a poor (weak, soft) record could be arbitrarily entered into the generation process, it would not survive the iterative step procedure described in faq (2) above. The second answer is that the generation process is not totally computer automated. The computer is used only as a tool that produces a set of standards based on the numbers it is given â€“ a person then â€śevaluatesâ€ť the results and accepts or rejects the computed standards so that it may take numerous re-fittings to get a suitable result.
The â€śevaluationâ€ť process mentioned above is as follows. The curve fittings are produced in both the time domain and the rate domain to ensure that specified rate requirements are met. It is difficult to qualitatively evaluate a table of time standards just by looking at the numerical progression of values therein. To better judge the quality of the standards, the running rates for all events must be graphed to facilitate determination of their suitability. Rate graphs (two kinds, one being across all ages for an event â€“ the other being across all events for one age) must have a smooth continuity for each rate line, and the rate lines must be proportionally spaced in a reasonable way for the standards to be realistic and hence considered suitable. Also age-factor graphs are produced (two kinds, one being across all ages for an event â€“ the other being across all events for one age) and evaluated in regard to their smoothness of continuity and in regard to their consistency of shape. In addition, graphs are produced for major events that plot the records and standards that resulted so that overall curve-fitting suitability can be judged for those events including a composite chart graphing ten major events that demonstrates how well the curve-fitted standards reflect the records they are based on, and how the curves of the standards are similarly shaped and proportioned for all event distances. The time scale of the chart is logarithmic (which is different from the time scale of the other records-versus-standards graphs). With this chart it can easily be seen whether some particular age range (old masters vis-Ă -vis young masters, etc.) seem to have standards that are too soft or too hard, or if some particular events (sprints, middle distances or long distances) seem to have standards that are too soft or too hard compared to other events.
Thus the standards and age-factors determined here are subjected to a rigorous series of evaluations and re-fittings before being finalized.
The evaluation graphs discussed above are included in each set of Standards and Age-Factors.
Long Distance Road Running Events
The FAQs and general remarks in regard to the generation of standards and age-factors in the section for Track Running Events also apply to the generation of standards and age-factors for LDR running events. The general remarks are not repeated here in the interest of saving space â€“ refer to the referenced section for the applicable LDR remarks.
While the track running events and LDR road running events are presented in separate sections, their standards are generated as a single curve fitting of distances ranging from 50 meters to 42,195 meters of the marathon. Thus the standards represent international levels of performance, and no distinction is made whether the performance took place on a track or a closed circuit road course or on a point-to-point road course subject to the proviso that only the ratified records of an appropriate governing body are included in the records database.
The FAQs and general remarks in regard to the generation of standards and age-factors in the section for Track Running Events also apply to the generation of standards and age-factors for racewalk events. The general remarks are not repeated here in the interest of saving space â€“ refer to the referenced section for the applicable racewalk remarks. The same curve fitting equations are used to generate the racewalk standards as are used to generate the track running event standards. The same graphs and evaluation procedure apply as described for the track running events, and the same standards re-fitting process is followed until the racewalk standards are finalized.
Standards are provided for selected hurdle events most frequently included in sanctioned meets. The general remarks in regard to the generation of standards and age-factors in the section for Track Running Events also apply to the generation of standards and age-factors for Hurdle events. The general remarks are not repeated here in the interest of saving space â€“ refer to the referenced section for the applicable hurdle event remarks.
Hurdle time standards are generated using the same time-age equation (fitting the records for all ages for one event distance) employed for runs and walks as discussed in the section for Track Running Events. The time-distance equations (fitting the records for all event distances for one age) of the runs and walks, however, do not apply for hurdle events.
Hurdle event time standards are uniquely indexed to running event time standards of the same distance in order to determine hurdle factor values. Hurdle factor values are used in generating standards for a family of hurdle events (different hurdle heights but same event distance), and for ensuring that no hurdle time standard results that is faster than the standard of its flat distance. The time-age equation curve fitting process is carried out separately for each family of hurdle events. This allows hurdle factor values to be analyzed and readjusted as may be required for each family. Thus individual events are dependently related within a family of hurdle events, and as a result the standards are not generated on a stand-alone basis for each separate hurdle event.
It is difficult to qualitatively evaluate a table of time standards just by looking at the numerical progression of values therein. To better judge the quality of the standards, the hurdle running rates for all hurdle events must be graphed to facilitate determination of their suitability. Rate graphs must have a smooth continuity for each rate line, and the rate lines must be proportionally spaced in a reasonable way for the standards to be realistic and hence considered suitable. Also age-factor graphs are produced and evaluated in regard to their smoothness of continuity and in regard to the suitability of their shape. In addition, graphs are produced for each family of hurdle events that plot the records and standards that resulted so that overall curve-fitting suitability can be judged for those events. The graphs described above are evaluated and changes are re-fitted until suitable standards are obtained.
Standards are provided for selected field events most frequently included in sanctioned meets. The standards reflect distances jumped or thrown by world record level competitors. While the standards reflect distances jumped and thrown by world record level athletes, they also apply to youth and masters age field eventers of all ability levels through application of the performance level percentage concept. The performance level percentage concept is that a competitorâ€™s performance level is the percentage obtained by dividing the competitorâ€™s distance for the event by the event standard. To do this, the standard and the competitorâ€™s distance should both be measured in meters.
The primary purpose of performance level percentages is to facilitate performance comparisons. An individualâ€™s performances can be compared, or performances of many competitors can be compared. Since the standards for all events are set at a world record level, comparison of performances for different events is valid. While standards are not established as such for lesser level or average field event competitors, the â€śworld recordâ€ť level standards here are applicable to field event competitors of all ability levels by using the performance level percentage concept.
The age-factor values for field events were determined by dividing the distance standard for each of the other ages by the open class distance standard for the event. To convert a distance jumped or thrown to its equivalent performance as an open class event, the distance jumped or thrown (in meters) is divided by the age-factor value listed for competitors of that age for that event.
The distance standards for the jump events are generated by curve fitting the records for the pole vault, high jump, long jump and triple jump as separate stand-alone events
The five throwing events (shot put, hammer, discus, javelin and weight throw) are curve-fitted by the same equation used to curve-fit the jumps while applying the dynamics theory of masses to â€śproportionâ€ť records for all implement weights of a throw event family. The standards curve for each implement (e.g., 16lb, 6kg, 5kg, 4kg and 3kg shot put) within the family can then be evaluated and readjusted as may be required. The curve fitting process is carried out separately for each of the five families of throw events.
It is difficult to qualitatively evaluate a family of distance standards just by looking at the tabulated values that result. To better judge the acceptability of the curve fitting, graphs of the plotted standards for all events within the family must be evaluated using some basis of comparison. â€śComparisonâ€ť graphs are produced that plot the distance standards for each implement of the family against age. Each standards curve must have a smooth continuity and the curves must be proportionally spaced in a reasonable way for the standards to be realistic and hence considered suitable.
Also age-factor graphs are produced and evaluated in regard to their smoothness of continuity and in regard to their consistency of shape. In addition, graphs are produced that plot the resulting records and standards so that overall curve-fitting suitability can be judged for those events. Thus the field event standards and age-factors determined here are subjected to a reasonable series of evaluations and re-fittings before being finalized.
The evaluation graphs discussed above are included in the set of Standards and Age-Factors
Some evaluation graphs plotting records-versus-standards for throw events show numerous records that plot as being better than the standards. See note below for explanation.
Note: A special kind of masters age record problem studied by WAVA in 1994 was that of the sustained â€śtoo goodâ€ť performance which occurs primarily in the throwing field events for younger masters age men and women. Many competitors complained over the years that some menâ€™s and womenâ€™s field event throw records were just too good to be acceptable, and it was contended that performance enhancing substances were probably being used by those competitors, and that it was unfair to accept those performances as standards against which the â€ścleanâ€ť competitors were measured.
The probably â€śtoo goodâ€ť records were subsequently identified â€“- however those performances are listed as the existing records in the current record books and so they are in the database for the field event standards generated here. While those records were not used to generate the standards here they are nonetheless plotted on the records-versus-standards graphs to show just how â€śtoo goodâ€ť they are. Most of these â€śtoo goodâ€ť records are now over 20 years old and are still in the NMN record books as nobody has been able to exceed them â€“ yet another indication of their â€śtoo goodness.â€ť
Ultra Distance Running Events
Note: Ultra distance event â€śstandardsâ€ť and their age-factors are included here only to demonstrate their feasibility as potentially useful standards. These demonstration standards are for illustrative purposes only as they have in no way been proven to be suitable or reliable.
The general remarks in regard to the generation of standards and age-factors in the section for Track Running Events would also apply to the generation of standards and age-factors for ultra distance running events. The general remarks are not repeated here in the interest of saving space â€“ refer to the referenced section for remarks applicable to ultra distance running events.
These demonstration ultra distance standards were generated using records for event distances ranging from 42,195 meters to 200,000 meters, and used the same equations and algorithms as do the track and road running events.
The effort to generate performance standards for ultra distance events is an unfinished undertaking here and thus the results are at best only exploratory standards. While fewer records than desired were available for curve-fitting purposes, the exploratory standards are â€ścautiously reliableâ€ť but are certainly not well founded to a suitable statistical level of confidence. More records will be required to establish a suitable database of records. The â€ścautiously reliableâ€ť standards are reasonably indicative of what reliable results would look like, however changes would be expected before stabilized and fully reliable standards could be attained.
The Effect of Aging on T&F Performance
The original purpose of the authorâ€™s effort here commenced as a personal curiosity in 1976 to investigate the effect of aging on oneâ€™s ability to run; generating standards was an unintended outgrowth that occurred later on. The effect of aging is still an important part of the standards package and is briefly discussed here.
In each of the Event sections (Track Running, LDR Running, Racewalk and Field Events) figures 1, 2 and occasionally 3 present graphs plotting age-versus-records for individual events, with the youth and masters age spread being nominally 6 to 100. These graphs each represent an aging profile for their event. The importance of these figures is two-fold, first in verifying how well the standards fit the records data, and then by being a picture of what the effect of aging is in regard to those events.
In each Event section there is also a table of age-factors and graphs of age-factors plotted for selected events. By definition an age-factor determines the effect of aging on performance (for that age for that event) as a fractional value ranging from zero to one. For example if the age-factor value is 0.75 that means the worlds best performer for that age can achieve only three-quarters of the performance of the worlds best open class performer in that event, and if the age-factor value is 0.25 the worlds best performer for that age can achieve only one-quarter of the performance of the open class level for that event.
Thus the table of age-factors and/or the performance standards can be analyzed to determine a variety of things such as (1) is sprinting ability or endurance ability more effected by aging, (2) is the aging effect the same for both males and females, (3) is the effect of aging the same for running events, racewalking events and the jump events, (4) is the ratio of female to male performance standards the same over the entire age range, and (5) is the effect of aging the same for youths as it is for masters.
It was determined that aging occurs in a logarithmic manner for the activities of running, racewalking and jumping â€“ the youth and masters age-versus-records aging profile for each event are all quite symmetric when plotted on a logarithmic age scale.
It was also determined that several relationships of significance are apparent when comparing the standards for men and women to one another for running, race walking and field jumping events. The table summarizes the relationships obtained by the ratio of menâ€™s to womenâ€™s time standards for runs and walks and the ratio of womenâ€™s to menâ€™s distance standards for the jumps. If a ratio is for example 0.85, this means that women run or walk only 85 percent as fast as do men of the same age, or that women jump only 85 percent as far or as high as men of the same age do.
Performance of Women vis-Ă -vis Men
Age 5 10 15 20 30 40 50 60 70 80 90 100
Runs* 1.00 0.93 0.90 0.90 0.90 0.90 0.90 0.89 0.86 0.78 0.49
Walks* 0.91 0.91 0.91 0.90 0.89 0.87 0.85 0.80
Jumps* 0.85 0.85 0.85 0.85 0.84 0.82 0.81 0.79 0.76 0.72 0.67
*Runs: averaged results of 13 events from 100-Marathon. *Walks: averaged results of 13 events from 1500-50K. *Jumps: averaged results of HJ, LJ and TJ.
Two observations here: the first about youths, and the second about masters. At age 5, boys and girls exhibit about the same ability to run. Then boys gain more ability than girls and by age 10 girls run only about 93 percent as fast as boys, and then from ages 15 through 50 women continue to run about 90 percent as fast as men of the same age.
Then looking at the age range of 20 through 50, women walk about 91 percent as fast as men, and women jump about 85 percent as far or as high as men. For ages 50 and on for the older masters, women walkers maintain their ability vis-Ă -vis men significantly better than do women runners (though neither venture much into the 90-100 age range), while women jumpers exhibit a gradual but orderly loss of ability vis-Ă -vis men and continue to jump through age 100.
These annually updated standards are not the published standards of any official governing body of track and field. The standards are prepared by C. A. Phillips (an individual not affiliated with any track and field organization) and represent his personal determination of what the most representative standards are.
Permission to Use Standards
Individuals or organizations choosing to use these standards for their own purposes may do so, and in return, referencing this website (http://phillipstrackstds.tripod.com) as the source of the material used would be appreciated. For questions or comments email â€śDr.email@example.com.â€ť
How the Phillips Standards Came About
These standards for track and field events are the end result of investigating the effect of aging on oneâ€™s ability to run commenced as a personal curiosity in 1976. The initial investigation was made possible because hand-held programmable-calculators became available in the marketplace providing the means to more easily solve mathematical problems that were previously left untouched because of their complexity or the onerous repetition involved.
Using the simple pocket calculators of that time, the effort began by fitting time-versus-age curves for masters age group running records in an attempt to understand what happens to running performance as the result of aging. An evolutionary process was soon taking place â€“ the investigator began to learn more about the equations involved in fitting performance curves for runners, increased participation in competitive masters age running worldwide produced records for ages where no records existed and improved many records which did exist, and affordable calculators/computers of ever-increasing capability flooded the marketplace. Initial efforts dealt only with menâ€™s running events.
Hurdles, field events and racewalking were added along the way for both men and women including running events for youths. Obtaining new or improved records while also using more capable computers made it necessary to go back frequently and re-do everything previously done because better fitting curves were then available to produce improved results.
What started out as a sports hobby endeavor evolved into a serious work guided by the principles of the scientific method.
The effort commenced in an exploratory development fashion, it then went through an initial development period where additional information on the nature of the subject matter was still being discovered, and it then passed through the stage of examining numerous hypotheses concerning why the standards didnâ€™t take into account some other consideration.
Comprehensive analytical investigations and evaluations were conducted as every new hypothesis emerged. The work is academically creditable; its standards and age factors now fully reflect international records for men and women track and field competitors of all ages.