Results for FireDragon Fitness Test

[Bill Glasheen]

Fedele

I understand your impatience. However I am not going to share raw numbers here for all to peruse. To make sure more people take the test, I will maintain privacy. And I'm still working on standards and such.

The raw numbers won't change. How we combine and score stuff will depend on what we find.

Keep watching...

- Bill

[Bill Glasheen]

ALL CANDIDATES (N=23) (one candidate value rejected)

Mean ........ 51
Std Dev .... 20.73
Min .......... 14
Max ......... 124


MALES (N=17) (one candidate value rejected)

Mean ....... 55
Std Dev .... 20.23
Min .......... 28
Max ......... 124


FEMALES (N=6)

Mean ....... 40
Std Dev .... 19.02
Min .......... 14
Max ......... 70

It's worth noting that Fedele's 124 was truly an anomalous result. It's obvious when you view the histogram. Everything else indicates that the distributions are Normal (Gaussian) with the same spread. The means are just a little different for males and females.

[Bill Glasheen]

ALL CANDIDATES (N=24)

Mean ....... 7.85
Std Dev .... 6.28
Min .......... 0.08
Max ......... 21


MALES (N=18)

Mean ....... 9.52
Std Dev .... 5.96
Min .......... 0.37
Max ......... 21


FEMALES (N=6)

Mean ....... 2.85
Std Dev .... 4.52
Min .......... 0.08
Max ......... 12

If a candidate couldn't do a pull-up, they were allowed to do a one-minute hang. The percent of a minute that they could hang was made equivalent to the percent of a single pull-up. Two candidates did a full-minute hang, and got full credit for a single pull-up.

The fractional number for the males was a 6-year-old (Sterling).

Cynthia Giasetto pulled off the anomalous result for the women (12 pull-ups). I observed that performance; those were 12 honest pull-ups. Attribute that to good genes and some gymnastics training. Three of the six women were able to do one or more pull-ups.

[Bill Glasheen]

ALL (N=24)

Mean ....... 43
Std Dev .... 12.27
Min .......... 11
Max ......... 61


MALES (N=18)

Mean ....... 45
Std Dev .... 12.74
Min .......... 11
Max ......... 61


FEMALES (N=6)

Mean ....... 40
Std Dev .... 10.86
Min .......... 23
Max ......... 56


There was a smaller difference for this event between the males and the females. The little guy (6-year-old) defined the male minimum. That was a bit of a low-end outlier. But those were honest sit-ups.

[Bill Glasheen]

Remember that...
1 inch = 2.54 cm
12 inches = 1 foot
39.37 inches = 1 meter

ALL (N=24)

Mean ....... 76
Std Dev .... 15.28
Min .......... 54
Max ......... 113


MALES (N=18)

Mean ....... 80
Std Dev .... 15.28
Min .......... 54
Max ......... 113


FEMALES (N=6)

Mean ....... 63
Std Dev .... 4.59
Min .......... 58
Max ......... 70

Obviously there is a male advantage here. Average for the men is 6 foot 8 inches (2 meters). The same for the females is 5 foot 3 inches (1.6 meters). Even the little guy, small for his age, did fine w.r.t. the females.

A reality check here... There aren't many women who can dunk a basketball, and literally only a few who have done it in a regulation game.

Height did not seem to matter, in case you were wondering. One female candidate who was 4 ft. 11 inches got better than the mean result for women.

[Bill Glasheen]

ALL (N=24)

Mean ....... 53
Std Dev .... 9.78
Min .......... 33
Max ......... 73


MALE (N=18)

Mean ....... 53
Std Dev .... 9.24
Min .......... 34
Max ......... 73


FEMALE (N=6)

Mean ....... 52
Std Dev .... 12.16
Min .......... 33
Max ......... 69


There's no difference here between male and female performance.

I did see a need later on to define the squat a bit better. One person started with an extremely wide stance, and I had to change them in mid-execution. Some squats were a little more honest than others.

I am betting though that this event affected the performance on the subsequent mile run more on some than on others. And IMO, that's strictly a conditioning issue.

- Bill

[Bill Glasheen]

ALL (N=23) (one no-finish)

Mean ....... 8:57
Std Dev .... 99.56 sec
Min .......... 6:36
Max ......... 12:19


MALES (N=17) (one no-finish)

Mean ....... 8:42
Std Dev .... 78.51 sec
Min .......... 6:45
Max ......... 12:19


FEMALES (N=6)

Mean ....... 9:41
Std Dev .... 144.27 sec
Min .......... 6:36
Max ......... 12:10


Sterling was picked up 3/4 of the way through under protest by the "mercy car." Two candidates (Sterling 6 years, Chad 12 years) elected to run the event in sandals, with Chad finishing in a respectable time. (I do not recommend this, BTW.)

Vinny with an artifical hip walked the event, and was only 9 seconds slower than the next slowest time. I've convinced Vinny that he and others can practice speedwalking, and probably come up with respectable times that involve a low-impact performance.

I see a BIG spread here, meaning a big opportunity for improvement. Speed walkers should be capable of doing well compared to this field. It might even be a viable strategy for many if their thighs are very fatigued. And it certainly is a healthy exercise activity for ALL ages. Food for thought...

One candidate SWORE I called out a time of 7948, which doesn't make sense. He wrote it down on his card. But I figured it out... He was right. I called out

... 9:46 ... 9:47 ... 9:48 ... 9:49 ...

Guess when he crossed the finish line.

- Bill

[Bill Glasheen]

I have ideas about how I will combine the results. I discussed the use of Z-scores, which convert distributions to means of zero and standard deviations of one. This way I could even have a bar chart result printed out which show people (by category) if they did better or worse than the mean, and by how much (in standard deviation units). And Z-scores are easy to combine.

One could also combine results on a percentile basis. You could even cap the upper level, which I notice is what the Marines did on their test.

Coming up with a break point is an interesting thought. I'd prefer people look at the histograms as opposed to just looking at these raw descriptive statistics. But you can eyeball them a bit already.

Different categories by gender? Age? Size? I'll have to do some stats here.

Lots to think about.

I personally think that everyone who entered this event - including and especially Sterling - deserves "recognition" due to the difficulty of being the first. It's always harder to be the first.

I can keep results over time. I can accumulate means and standard deviations over time. I can help folks chart progress over time. Lots of possibilities here...

Do you give just one award if you made a threshold performance level once? Get recognition each and every year you got above a threshold value? Fitness, after all, is an ongoing affair. It is a process; we need to maintain.

Lots to think about.

The floor is open for comments.

- Bill

[gmattson]

I feel that the posted results of the test should be simpler to understand. Statistics might give the public an idea about how their scores in the various challenges relate to the best, worse and medium, but in my mind, the best way to compare is to simply show the different scores for all who took the test. That way, everyone will be able to see where they stand in each competition compared to the group.

In order to list the competitors, in order of achievement, you have to come up with some way to assign a value to each test, add up all the values and create a way to list the competitors from the top of the list to the bottom.

We might have a cutoff value. . . where people who score below what we arbitrarily assign as a "minimum" score, will not be listed. This will prevent anyone who scored badly from being embarrassed.

The competitive aspect of the test is important, since it will encourage everyone to train and try to improve their standing in the list.

We can always change the process, but it is important to publish the results in a timely manner.

[chewy]

GEM,

Surely we don't make everyone's results public to everyone on the board. I know many probably don't mind, but some may be discouraged from participating in the future if their results are made public knowledge. If people want their results couldn't they just ask Bill to PM them?

The engineer in me likes what Bill is doing with the statistics. I find it useful information, but I can appreciate that not everyone embraced mathematics in school or their careers.


cheers,

chewy

[gmattson]

Interesting Chewy. I probably don't appreciate the fact that many people might wish to take this test and not want their names and results posted. On the other hand, I know how many thousands of people read these posts and how few have the nerve to post their views.... even though they don't have to be identified!

Its like running an Olympics and having some of the participants telling the organizers that they don't want their pictures taken or names published.

I guess we will have to call everyone who make the cut, if we have permission to list their names in the FireDragon Hall of Fame. Those who don't will simply be left off the list.

[CJG]

Cynthia Giasetto pulled off the anomalous result for the women (12 pull-ups). I observed that performance; those were 12 honest pull-ups. Attribute that to good genes and some gymnastics training.

Hey friends, finally able to walk down stairs again. Actually, just referred to as one of the 'Muscle Girls' in dance, no gymnastics. Not at all surprised to hear Fedele took the lead like that. Now if we can just get him to stretch Thanks for giving us the opportunity to challenge ourselves Bill, Jimmy, and whoever else organized it. Thanks for all your work again GEM, Susan, and all presenters. Finally made it to the Port O' Call this year and somehow Roy was able to produce a glass of red wine. Great time. Madga & Bridget: meet you at the tree for Kami's belly rolls next August...

[Valkenar]

I think it's reasonable that people would not want their scores made public. We're not olympians, and nobody wants to be known as the least fit person in uechi-ryu. If someone isn't in very good shape, it's not going to motivate them to try the test if they know that people may be looking at their results and snickering behind their monitors. Or at least that's how I imagine some may see it.

What I think would be useful though is an anonymous listing of the total scores, so you can compare yourself anonymously against the rest. Just something like:

1: 104.5
2: 101.4
3: 98.9

So if you take a look at your own score you know how it compares. It's all well and good to show a mean, mode and standard deviation, but that's not intuitively obvious to much of anyone (professional statisticians aside).

Also, I think for push-ups, pull-ups, sit-ups and squats, it helps to be smaller. The broad jump and the run are better for taller people. These aren't the deciding factors by any means, but I suspect they're a good piece of it. Though if the statistics show otherwise I may eat my words.

[Bill Glasheen]

It's obvious we have a difference of opinion on posting results.

I come from the medical community. Where I come from, sharing personal data can put you in jail and clean your bank account. And that's the way it should be, damnit!

As for something like this, well I think it depends. In order to answer some questions like the below...

Also, I think for push-ups, pull-ups, sit-ups and squats, it helps to be smaller. The broad jump and the run are better for taller people. These aren't the deciding factors by any means, but I suspect they're a good piece of it. Though if the statistics show otherwise I may eat my words.

...I asked for - and got - a lot of personal information. People gave me their age, their weight, and their height. Plus we measured their percent body fat. I am very thankful for that 100% cooperation, and I will defend THOSE data to the end. That information is nobody's business but the scientist trying to benchmark the test, check what factors influence performance for possible future test modifications, and for setting standards and group categories.

Results are something else. How detailed do we get? Well I believe that's all about what we are trying to achieve here. Here is a possible list.

1) Encourage Uechika and the collateral community to make fitness a part of their overall lifetime martial arts goals.

2) Improve the performance of the overall Uechi community.

3) Give folks a venue to compete. Some love to compete on things like this, and show their stuff.

4) Assess where we are as a Uechi community. Already it's obvious to me that Uechi people are good in some areas and not so good in others. We should not be surprised that they had no problem with push-ups and sit-ups. Many dojos do these like it's the be-all, end-all of martial arts. Most define their Uechi by their thrusts and their rock-hard stomachs. But there are grappling elements to Uechi, a need for explosiveness, and a need for endurance and good movement. In these fitness results I see some opportunities with many. These data help show the naked truth to some. (And it is a credit to all who participated that they allowed themselves to be assessed in this manner.)

Yes, I think it would be a great thing to show who made it past a threshold. But think about this. When we do dan tests, do we release the numbers to everyone? No. We tell everyone who passed, and we can find out by default who did not if we do the research. The individual scores for the categories are not released.

This could be different. How much do we release? Final combined score of those who passed? That's easy enough. But we should think carefully on this.

I'd love to hear from Rich how the military handles the results on their fitness tests.

I absolutely am all for giving a beautiful report to the participants that shows their results compared to some kind of benchmark, be it an absolute scale soon to be determined, or a group mean. This kind of feedback shows people where they are, and what they should work on. And this would foster a little bit of healthy competitiveness.

I absolutely am all for competitions between teams of people in the future for the best scores.

I absolutely am all for some kind of recognition of people who achieve elite performances such as Fedele, Kevin, Bridget, and Cynthia. Let them all show us what can be done. Make those extraordinary performances something that future elite Uechika will strive to beat.

Now, back to Justin's comment. I'll have these data in SAS soon enough, Justin. I'm not sure if 24 records are enough to do the kinds of stats I'd like to do. It may take more results over time. However I have enough different kinds of measurements to see what factors drive performance.

Remember that "size" can be deceiving. The two pocket pistols you observed doing so well have bodies to die for. Fedele was a former competitive bodybuilder. In my conversations with him, I know he used to toss around the kind of weight that 200 pounders usually dream of doing. Fedele is strong as an ox IN SPITE OF his size, and not because of it. And I have seen CJ with a tank top on, and seen her performance in my sai and kata classes. Make no mistake about it; she is special. She has proportionate strength, incredible coordination, and she is shredded (to use a bodybuilding term). These two happened to be small. But they both are strong, and they both have very, very low body fat.

If some of our 200 pounders were built the same way, they would do just as well on these events.

What you guys don't realize is that there is a bit of culling that goes on in martial arts classes. The only folks the size of Fedele and CJ that survive are the ones who are phenomenal. Otherwise they get bounced off the walls. The 200 pounders and above can get by for a long time getting pounded by average people.

When you find a big person with extraordinary ability like Fedele and CJ, it's something to behold. My first Uechi instructor was Rad Smith. He was 6 foot 3, and a former sprinter and swimmer. Rad could have competed with Fedele and CJ on this; I still remember his extraordinary performances on these kinds of assessments. And his subsequent martial abilities were absolutely frightening. He hit me in the sternum so hard one time that it took about a decade before it stopped hurting. And he used to toss us around like rag dolls in sparring. GEM probably remembers his abilities. Rad had about 5% body fat. He was just a bigger version of the pocket pistols we saw at camp.

But... We will see what the data tell us, Justin.

The good thing is that - with 6 different measures - there's something in it for everyone. To be really, really good, you need to run the table. That's not easy, and it shouldn't be IMO.

More in a bit.

- Bill

[Bill Glasheen]

I feel that the posted results of the test should be simpler to understand. Statistics might give the public an idea about how their scores in the various challenges relate to the best, worse and medium, but in my mind, the best way to compare is to simply show the different scores for all who took the test. That way, everyone will be able to see where they stand in each competition compared to the group.

I am considered a national expert on the economic profiling of physicians. Physicans by nature are a HIGHLY competitive lot, and yet they get the heebie geebies if you release a lot of their detailed information for other to peruse. I am used to developing the kinds of reports that show people how they did w.r.t. others without divulging the detailed information of others. I suggest we try to follow that kind of format. This will encourage paticipation, IMO.

We can always change the process, but it is important to publish the results in a timely manner.

I am hoping everyone can be patient here.

I agree that when we get this down, releasing results in a timely manner would be a really, really good thing. But I think we should try hard to get it right the first time.

And I'm trying to share just enough information here for folks to be able to give their 2 cents before we etch it in stone.

I have no problem making changes with time. But I think we probably can get it right this time. We already have a good start, having chosen tested assessment procedures. And Rich has provided me with some examples of how the military scores these kinds of tests.

But remember... Rich ran through such a test while they were designing it. Their first step was to get empirical data to see what the standards should be. And that's what I'm doing, only I believe I'll end up delivering the finished product in a fraction of the time.

Perhaps next time it can be programmed where you can enter the results, and spit out the answer on the spot. That is possible.

- Bill