Saturday, March 29, 2014

No-Holds Barred Interview with Carl Valle

No-Holds Barred Interview with Carl Valle



It is always fun and insightful to have a no-holds barred talk with Carl Valle. Hate him or love him, whatever you ask him he always provide no BS answer and for that he needs to be respected.

I took the chance to ask him 5 nasty questions and he really took his time and energy to answer them in great detail. All I can say is enjoy the interview and the insights of Carl Valle.


Mladen: Everybody seems to big into monitoring and analytics lately. What I wonder is how the pro clubs solve the frequent human nature issues, which I like to call (1) “self fulfilling prophecy” and (2) “bad performance alibi”. What happens if an athlete sees bad score – he might expect bad performance and injury (“self fulfilling prophecy”) and/or use that metric as an excuse for a bad performance or a game (“bad performance alibi”). Assuming that athletes’ buy-in into monitoring is there, along with accountability culture, what is your opinion on the mentioned human nature issues and how to solve them?


  
Carl: I am fully aware of the problems of merging monitoring or any feedback to athletes and have made more mistakes here than anyone and will make more in the future. Most of my work is track and field, but do work with some team sport athletes. I experiment and fail more than I succeed because if you are not winning every race, you are not perfect.  It’s very surprising how much technology I use that people don’t realize because I focus on fun and the human side and try to hide the sport science as much as possible. Passive aggregation of data is key, and good monitoring is efficient and minimal. Data mining has some hype and I want to encourage people to get more out of their data though since we tend to want to add more monitoring and testing and not do what we know is helpful.

Back to your question, how do we solve the interaction of objective feedback of monitoring to a human athlete? Human problems need human solutions, so it’s a combination of sport psychology, education, salesmanship, and trust. I said in an interview in 2010 it’s not what you know it’s what you can get your athletes to do. Many coaches are smart but when it’s applied it’s not effective.

  •      Sport Psychology- The “soft science” is the strongest variable because it’s about dealing with the human element of emotion and behavior. One can argue it’s about connecting but I think it’s about placing the athlete in a mind state of doing what it takes to succeed with conscious decisions. Coaches should be aware of the three Ps, personalities, perspectives, and what has worked in the past. Motivation is also a part here, and tapping into the athlete’s goals is about bridging behavior to belief.


  •  Education- Another three letter approach using the three Vs in monitoring or athlete education in general. Explaining just enough information, be it verbal (dialogue) or visual (infographics) is extremely helpful to get the athlete to value the data. Talking about parasympathetic changes or wattage in jumps is not the language of athletes (although they are evolving fast thanks to the internet) but the use of creative analogies and layman’s terms is helpful here. Numbers are ok as well, since solid objective feedback of the clock, the bar, and the tape all ensure the athletes know what is going on.


  • Salesmanship- Having athletes buy in requires coaching marketing and campaigning a cultural reset. Sales is easy when you love what you are doing. If monitoring is a pain to the coach, no way is the athlete going to leave feeling positive either. If you don’t use the data or don’t demonstrate


  •       Trust- A combination of experience, history of results, knowledge, and the coaching relationship determines everything. Athletes want to know your motives because caring is hard acquire in large group situations. You can claim to care when you barely know names or spend time with someone. Sharing your motivation for helping them gives the athlete a perspective to how much effort you plan to invest in and what your limits are. Is the coach in it for the passion and art or financial reward and ego?


At the end of the day athletes are not horses and even horses are not just four legged athletes who don’t talk back. Make sure what you get from them isn’t a pain to those involved and the value of the data is beyond actionable but reinforces the difference of not doing it and the difference between other options. Your specific example of the feedback of poor or even great physiological or testing readings being a possible blow or creating overconfidence is something one must manage. This is especially helpful during tapers and playoffs when the stakes are higher with things and the key I believe is being positive and realistic.  I think the coaches body language and expressions of getting data and the plan after the information will shape how the athlete responds and time will shape how the athlete interprets this. I think when data is suppose to predict, it can lead to some interesting responses and the above four components of athlete and coach interaction can help direct the mental outlook to where it’s needed to go.

Mladen: Postural Restoration Institute (PRI) is all the rage lately. What is your take on their postural restoration and ‘corrective movement’ and ‘movement gurus’ in general, along with the newest breathing re-education? How should these “soft” skills be implemented with hard-core high-performance athletes and clubs?


Carl: Remember every rage means it’s a trend, and trends come and go. I remember in the late 1990s I was an intern with the then Tampa Bay Devil Rays (I joke all the time since when I left they dropped the Devil from their team name) and I was suggested to go on the “World Wide Web” and see what they were doing because posture was again trending because of Paul Chek and NASM at the time with Mike Clark. Ken Crenshaw was the assistant trainer at the time and was ahead of his time, but we need to see the impact of the therapy by stepping back and seeing the results. Since everyone is talking about Bulletproof bodies and even drinking bullet proof coffee, where are all of the legions of resilient athletes playing unholy minutes with zero injuries? If someone goes to Kenya and introduces the breathing techniques will we see a wave of sub 2 hour marathoners?

Do I believe the information has value? Yes. Do I believe the information has limits in a practical manner? Yes. What are those limits and how do we properly evaluate them? I think we need to recalibrate our expectations with any educational option out there. We have athletes that are out of shape conditioning wise breathing fire for circus performances or whatever they think they are doing still getting injured. What is amazing is how therapist and coaches on social media is hyping every workshop or seminar like they are resurrecting people or making freak athletes like Dr. Frankenstein. A healthy approach is to see what investment organizations should place into those methods and see what real world outcome one has. People always talk about value but nobody quantifies the value in real world coaching terms.

I think an integrated approach to breathing is needed since many coaches will be reinforcing habits and actions with athletes during training. In July of 2003 I blogged about breathing because I was seeing a spike of asthma medications with a team and noticed none of them got tested or evaluated, just prescribed drugs and inhalers. Regardless of one’s beliefs, the medical responsibility of the coach is to be educated on what he or she should do and when to refer out, something I am seeing less of. Anyway we all focused on strategies of coping with intense exercise and recalibrating what is hard. Most of the athletes simply were soft and went to inhalers when things go tough and I never intervene with medical situations besides following doctor’s orders. The moral of the story was clear, when training improved and mindset was integrated about dealing with strain some athletes were revaluated to find that symptoms don’t always mean problems. A Rugby or NFL player with broken ribs in the past may need some reeducation here but after speaking with Dr. David Lain on the phone years ago, we can’t make assumptions until one tests an athlete properly. Spirometry, capnography, and other pulmonary testing should be done holistically and have realistic plans in place. The goal is that athletes acquire better breathing by better program design and coaching reinforcement, not by spending hours on balloon sculptures.

Corrective exercise is a wide stroke of options coaches have to improve something and that can range from a self directed ankle mobility sequence to a clean lift off exercise for the upper back, glutes, and hamstrings. Like breathing training, we need to create a balance of what options will create safeguards against injury and get back to improving performance or at least sustaining output. I am a fan of the Dan Baker and want to ensure my speed, power, range of motion, and conditioning are sufficient before I look at smaller or micro problems. It feels good to the ego to see problems others can’t, but when I see athletes who are out of shape getting hurt, do the basics before we start fishing around for dysfunctions.

I am not a team strength coach but have witnessed some very progressive and creative solutions when I visit teams to learn or to install custom hardware or other technologies I can quickly assess who is organized and who is pretending. The best teams have a mapped out hierarchy of interventions based on impact of the options and created a culture to support the implementation by having clear policies, straightforward transparency, and a little marketing for compliance. When one follows those suggestions the effectiveness of any treatment option will rise, truncated, or be excluded from a system.


Mladen: You are big on using foot pressure mapping and TMG. How these tools are used to predict performance and injury and how are they used in return-to-play programs? How are these linked to manual therapy and corrective work (see PRI question above) and orthopedic corrections?  What is their “logistic burden” on the team sport clubs compared to individual sports?



Carl: After the Sydney Olympics was over, I was informed about the “muscle tester” by a few athletes and about custom spikes by the top sprinters. I was very skeptical and every Olympic cycle athletes coming back from Europe would tell me about the “electronic muscle fiber tester” and about pressure mapping for foot function evaluation.  Years later at the BSMPG a private workshop lead by two leading experts in TMG and pressure mapping evaluated an athlete to showcase why something subtle down the kinetic chain may be more important than we know. Remember movement screening is usually slow, unloaded with forces, and very unspecific to problem identification. It’s a good tool, but most practitioners never talk about other tools besides their movement screen and seem to forget gait labs at Universities are not fiction.

So that the readers are informed, TMG is tensiomyography, a way of evaluating muscle status and pressure mapping is a method of seeing foot kinetics using cells that collect levels of pressure. Some systems with pressure mapping have 12 markers and some medical and research grade options may have hundreds and sample at high rates while some clinical models that are more walking speeds and sample the data slower. TMG falls under the category of MyoAnalytics or muscle diagnostics, a growing area that we are seeing interest. Elastography and Myoton readings are exploding with some consultants because at the end of the day athletes don’t tear their CNS and pull mitochondria, they hurt their soft tissue. Physiological monitoring is still important and the data is far more simple to manage and should be done, but managing soft tissue and joint function is essential. You have asked three good questions in this topic and I will do my best to explain what is best practice and what is reactionary and frustrating. Logistical burden can be a logistical nightmare if not done right and I have witnessed cautionary tales and legendary stories from a high-resolution evaluation and treatment plan. What can be learned with implementation of many data sets can be true with surface EMG, motion capture, video, and now biochemical and physiological data. One has more information that can identify underlying causes that may slip through “movement screening”. Testing, analyzing, and interventions with this data is hard work, please don’t mistake me.

This is the most loaded question and perhaps most controversial one I will attempt to address. Perhaps this question is the most important one I have answered in any interview so far. You have asked a question that I hope will show why making changes in sport is so hard. I will actually show real examples of how I have used the technologies and metrics.

Injury Prediction
Increases of risk from mechanical strain on the body can be calculated but the body is a living organism that can adapt. Models or algorithms that claim to predict injury are usually very crude and impossibly short in calculation and data sets. The idea of risk analysis is to work backwards based on etiology of the injury and what biomechanical, loading, and biochemical risk factors are known to increase the chance of specific joint or muscle injury.  I am talking about internal biomaterial disruption from projected biomechanical variables. Sounds complicated but to simplify what does the research suggest with regards to movement and specific site injury.

  •          MyoAnalytics (Tensiomyography and myoton metrics) – This data is more fleeting and variable and can be used to do preseason screening and monitoring of manual therapy and training interventions. Also combined with Player tracking data and motion capture, a real profile of athletes can be made. Movement signatures with force plates or accelerometers is like painting with a wide brush for houses and trying to do a portrait.  One high profile team is including military grade thermography surveillance cameras as early warning tools and then validate with muscle diagnostics. When patterns calibrated by research and estimated coefficients are added, the early warning system alarms or tags the event.
  •       Pressure Mapping (In-shoe and barefoot)- Injuries from the plantar fascia to lower back can be linked to ground reaction forces that may not a good fit with the athlete. Remember athletes are now over-competed and under prepared, a death sentence coaches and medical teams are trying to manage. Asymmetries can be absorbed from the amazing nervous systems of athletes, but the workhorses are muscles that may not have enough ability to handle rapid eccentric forces. Pressure mapping alone has merit, but sEMG and motion capture connects the dots. If you look a professional soccer they are on a runaway train to imploding with higher outputs of both speed and conditioning, and small problems may be fine driving around a lazy Sunday to church, but on the autobahn going 180 kilometers per hour, alignment issues are sometimes exponentially problematic.  One example of this is an athlete that had one foot injured as a child with morphological and structural changes that caused him to drift 10% while sprinting. For every ten meters he veered a meter. As he got faster he experienced muscle strains and had him get pressure mapped and the COP trajectory of foot strike was radically different and this caused the drift each step. Combined with the analytics run on his fiber testing and jump tests, his fatigue pattern didn’t create a solid buffer zone. The patterns of muscle status cross validated the predicted patterns from EMG and motion capture, and the pressure mapping identified the potential cause.


Return to Play

Return to play is sometimes a removal of pain story. Pain is a growing subject of debate. I am appreciative of the science but chronic pain and chronic injury is not the same. Healing takes time and who has that in modern sport? Can one play and can one play effectively is up to the team coaches, performance staff, medical team, and the athlete of course.  I see return to play as a process of personal pain and function. I understand pain is in one’s head but injury or trauma to tissue is real. Imaging isn’t perfect but I like the idea of using TMG as a return to play tool rather than a monitoring device. One can compare the less injured and fatigued group bilaterally to baseline and normative data from past historical data. Teams can use pressure mapping for ankle and foot sway analysis and combine it with jump and running tests to ensure treatments actually did something.

Manual Therapy Integration

When people think about Podiatry they think orthotic prescription. Podiatry is similar to Psychiatry and people assume that you meet and greet and get drugs or an orthotic. Podiatry is about managing foot mechanics and sometimes physical therapy is prescribed. Sometimes podiatrists manage problems or do rehab with manual therapy with manipulations but all of this is fleeting. I am shocked how many barefoot gurus and therapist do youtube infomercials for their courses and think they are making amazing changes to foot performance. Some changes to foot mechanics can be done by bodywork and strength training and I encourage everyone to not think a modification to a shoe or change in footwear will do everything. Small felt inserts and orthotics may help, but manual therapy and other modalities are a small part of the process if individualized properly. For example the FAST protocol by ASPIRE showed different foot loading from exercises and EMS, but performance didn’t change. With injuries rising coaches and medical professionals are likely interested in maintaining performance and minimizing risk since most athletes come premade at elite levels.

Sports Medicine is getting more data driven and those that are against it are usually bad at technology. Many hands on people are brilliant and can juggle everything in their minds but are isolationists and the key is working collaboratively and future medical health records will integrate google glass and voice recognition. One team preparing for the world cup has a consultant doing this already  and the key with soft tissue therapy and joint work is to see what it’s doing objectively. Hands on work is perhaps placebo and neurological relief, but I have seen amazing data showing the connection and value of it. Combined work must be evaluated and I find myoton readings with massage to be a great predictor and evaluation of treatment because communicating tissue tone needs objective status for everyone to be on the right page. Like R2P strategies one needs to evaluate the treatments with objective data otherwise it becomes an ego war and the loser is the athlete.

Logistical Concerns

Sometimes outsourcing or referring out is needed when dealing with specialized knowledge and experience with technologies. SaaS models are growing for good reason. Nobody can do everything and know everything and teams are more likely to act as case workers. Teams should invest into internal budgets and external budgets and just look at sport science and sports medicine a potential service. Planning is the most essential part of screening and monitoring, and usually such exhaustive approaches are done after repeated failure and sometimes not much can be done to resurrect players after enough structural damage and deconditioning is present. Screening during preseason when people are healthy or less injured is key. Companies who can work like a SWAT team and be in and out quickly and thoroughly are going to be the winners in this space. More data, less time, and less annoyance is essential. Every private vendor loves what they do but never seems to get what the constraints of a team is dealing with and never see a perspective that is in the trenches. Any vendor that is making a silent impact with less invasive approaches is going to be coveted. A culture of fun, enriching, and minimally disruptive and educational will change sport.

I sometimes send data to Chicago, Copenhagen, and Johannesburg to allow the groups to use their knowledge to create a more comprehensive evaluation. My Chicago guy does more injury and pathomechanics, the expert in Denmark helps with therapeutic integration, and my bioengineering contact looks a biomaterial remodeling and strain calculations.  All three help to see if the solution investment is going to work in the long term. I do all of this part time as am a track and field hobbyist and if I can do it teams with far more resources should as well.

Mladen: Let’s talk about new monitoring tools that emerged lately – CheckMyLevel and MOXY. Can you share your opinions on both and how do you use them?



We are going to see an explosion of non-invasive technologies from start-ups and CheckMyLevel and MOXY are two players in this space. Crowd funding campaigns with Kickstarter and Indiegogo are producing a bunch of amazing products and some   I worked with a start-up from Germany testing body suits and without getting into detail, consumer products are not always team friendly or even athlete friendly. Non-invasive solutions are estimations to internal chemistry and are convenient but one must be warned. Even biochemistry is an estimation of what is going on for the most part so lactate is limited.

CheckMyLevel is a device that uses muscle stimulation to the hand and measures the response time and motion with an accelerometer to the thumb. Based on that data, the voltage and response is trying to gage an estimate of total body neuromuscular fatigue. The question remains how to this data is compared to and what prior approach we can learn from to see. Reaction time or finger tap tests are not new, but this device is trying to control for motivational variables with a controlled contraction and using the convenience of mobile devices it seems on paper a good solution. What will be needed is actual product validation and more development on the algorithm. I have used this for ten months and similar technologies for years and everything looks promising.

The MOXY Monitor is one of the many Sm02 and Sp02 products that are starting to come down the pipeline and are all promising to be the lactate tester killer in some form. Several studies have shown to support various lab tests but the question remains how does one use local muscle testing (Sm02) and regular conventional Sp02 work in practical settings. Look at the product MuscleSound.com and ask yourself is this going to be the Holy Grail doing this everyday? I think the MOXY Monitor needs to be micronized more so athletes don’t feel like Robocop or Inspector Gadget.  Also the placement of Insight to the calf is interesting because most optical sensors don’t use that location and specific details such as arteries and capillaries must be factored in.  Estimating hydrogen from calculated optical sensor data is prone to a lot of artifacts and any algorithm should be made public or at least validated. Look at the Mio Alpha and see that this technology isn’t perfect.

“Alpha is the world's first strapless, continuous coronary heart rate check you can dress in on your wrist. It has been tested accurate even although you are operating at performance speeds of up to 12mph (20km/h)."

Anyone trying to get intra rep recoveries of lactate readings will want to get the data if it’s valid as it’s a pain to get repeat speed endurance reps. Still, I am interested in the following?:


  •       I would like to see MOXY or a similar product used during speed tests and player tracking devices more in order to see how offensive linemen and defensive linemen fatigue specifically in American football. GPS and accelerometer data is like Jackson Pollock paintings, it’s popular and people believe they interpreted it right but I question the value of it’s use. We need more muscle fatigue information with sEMG and other data to see what is going on in the trenches. Also, let’s see how well those breathing “workouts” are transferring between reps and between sessions. Certainly we should se some changes in recovery that is showing up somewhere.
  •     As for the CML device I want to see it cross-validated with CNS testing from Omegawave and have both compared to some intensive analysis. While central fatigue and the peripheral fatigue are different, both have the same general purpose of seeing nervous system fatigue interact with performance incompetence and injury from fatigue. I love jumping tests but can’t do them daily so we want a passive way to look at general explosive ability status of the body. Perhaps some field tests mixed with some non-voluntary testing as well as POMS like scores can show what is more precise, assuming both are valid and reliable.


Of course all of this data needs to be merged to provide better enlightenment to what is really going on. Either device is one data set, and algorithms that can analyze and provide indications of risk that are realistic would be great.

Mladen: I have seen some examples of the clubs that have bought (and/or can buy) and used Athlete Management Software solutions (like SMARTABASE, Apollo, EDGE10, SportsOffice) switch back to using Excel and Dropbox. Do you think this might be the case because of staff rotation (new staff every 6-12 months in some clubs) or complexity of such tools? What is your experience in using those and can you make quick comparison? What do the coaches want in Athlete Management Software solutions?



Carl: My biggest warning to coaches is not to buy most AMS that are currently out on the market and visit other coaches. A big difference exists between custom solutions with a vendor versus the cookie cutter products we see available. Since professional teams are demanding and require very powerful looking tools for sales, the vendors are providing products to satisfy an array of professional teams. When you try to make everyone happy you tend to make nobody happy. On the other hand, pro teams are a small percentage of customers in sport, and that means to be profitable you need to saturate a 1% market with an expensive tool that is designed by business people to get adoption and this will frustrate coaches. I don’t think a mob hit out for me, but I have had conference calls on the behalf of few teams and asked some very uncomfortable questions to providers of Dashboards and AMS tools. I strongly suggest seeking out medical data software vendors or hire competent bodies to build your own tools. I can’t stand the garbage that is out there. Dashboards that are showing cartoony pie charts and emoticons for $30,000.00 USD? AMS tools that don’t work offline or are glorified and skinned Opensource CMS tools with team logos and colors? Embarrassing! Talk about American Hustle but most of the products are from the UK.

You mentioned the revolving door of coaches and of course change in rosters being a problem. Data dies on the vine if not warehoused properly and few teams have historical data to make future decisions correctly. Excel files may not be high tech or sexy, but we are not being held back by it when it takes teams weeks to investigate if an athlete got fat during the season. Red tape is choking internal problems that already exist and the smarter option is not fight rapid changes and keep things simple and fast. Some star athletes will services and specific data while others avoid it like the plague.

For example what does a team in the NFL do when they are subscribing to HRV, Catapult, Zephyr, and countless other data sets. No company wants to be a data provider, they want to be a data aggregator and want team eyeballs on their software. Look at the dismal API status with all the devices or data feeds. Very few companies are seeing what teams are doing with all of the data, not just their own. SaaS models need to be cleaner and feed to custom solutions like Haloview. This is normal and I agree with them. The problem is no vision exist except a underworld of “Mercenary High Performance Advisors” that come to teams and offer their expertise on how to manage it and are usually not their a year after. Why? People are drowning in junk data or “noise” and are dying of thirst for valid and effective data.  

Now we have the software itself. Who is making this stuff? Who is signing off on the user experience with it? Most AMS software is designed by lazy programmers who don’t have a passion for elite health and performance and the results show it. How many people that code and design are like you and understand training? Not to stereotype most programmers but they are not all educated on sport science and coaching. The vision of software should be by the power users and not the pretenders. Paradoxically the most vocal complainers of software should be listen to. Many products don’t get used and people are so embarrassed that they don’t bother at all, or even worse, just use a brief period to pretend they are using the product! Coaches want a product that does what the marketing material promises. Coaches are likely to want simple and clear data with very little annoyance factor by the athletes and flexibility for changes down the road. Excel and Dropbox are not as fancy but the cloud tools listed I can replace with a few subscriptions to online tools and mobile phones.  Sometimes it’s better to do less and do it well, then have a diluted product that tries to do everything and is subpar. Most coaches want some sort of simple and clear reporting, and like mentioned earlier most of the PDFs and Dashboards I have seen are so bad they need repeat workshops with Stephen Few and Edward Tufte.

A note to owners, general managers, and athletic directors- Most of the problems are getting good data and not storing it. Algorithms are going to be the next big purchase because everyone has one now that can predict the future but without data to enter they are powerless. Listen to your medical and performance staff and ask what data is hard for them to get and invest their with human power and then rethink the AMS use. Some great products exist but please be an educated shopper.  Vendors of AMS and Dashboards need to realize they are more warehousing and displaying data and that is a feature, not a full product. Platforms are the real tools and that is a very long discussion but look for that change in the next year or two.


Data not Doping


Tuesday, March 25, 2014

How to [pretend to] be a better coach using bad statistics

How to [pretend to] be a better coach using bad statistics

How to [pretend to] be a better coach using bad statistics

Here is a simple scenario from practice: Coach A uses YOYOIRL1 test and Coach B uses 30-15IFT (for more info see recent paper my Martin Buchheit, which also stimulated me to write this blog) to gauge improvements in endurance

Coach A: We have improved distance covered in YOYOIRL1 test from 1750m to 2250m in four weeks. That is 500m improvement or ~28%

Coach B: We have improved velocity reached in 30-15IFT from 19km/h to 21km/h in four weeks . That is 2km/h improvement or ~10%

If you present those to someone who is not statisticaly educated he/she might conclude the following:

  • Coach A did a better job, since the improvement is 28% compared to 10% of Coach B
  • YOYOIRL1 test is more sensitive to changes than 30-15IFT

As a coaches, we needs to report to a manager(s), so which one would you prefer reporting? 28% or 10%? Be honest here!

Unfortunately, we cannot conclude who did a better job (Coach A or Coach B), nor which test is more sensitive (YOYOIRL1 or 30-15IFT) from percent change data. A lot of managers and coaches don't get this. At least I haven't until recently.

What we need is Effect Size statistics, or Cohen's D. But for that we need to know variability in the groups, expressed as SD (standard deviation). Let's simulate the data and use usual SDs for YOYOIRL1 and 30-15IFT

require(ggplot2, quietly = TRUE)
require(reshape2, quietly = TRUE)
require(plyr, quietly = TRUE)
require(randomNames, quietly = TRUE)
require(xtable, quietly = TRUE)
require(ggthemes, quietly = TRUE)
require(gridExtra, quietly = TRUE)

set.seed(1)
numberOfPlayers <- 150
playerNames <- randomNames(numberOfPlayers)

# Create YOYOIRL1 Pre- and Post- data using 300m as SD
YOYOIRL1.Pre <- rnorm(mean = 1750, sd = 300, n = numberOfPlayers)
YOYOIRL1.Post <- rnorm(mean = 2250, sd = 300, n = numberOfPlayers)

# We need to round YOYOIRL1 score to nearest 40m, since those are the
# increments of the scores
YOYOIRL1.Pre <- round_any(YOYOIRL1.Pre, 40)
YOYOIRL1.Post <- round_any(YOYOIRL1.Post, 40)

# Create 30-15IFT Pre- and Post- data using 1km/h as SD
v3015IFT.Pre <- rnorm(mean = 19, sd = 1, n = numberOfPlayers)
v3015IFT.Post <- rnorm(mean = 21, sd = 1, n = numberOfPlayers)

# We need to round 30-15IFT to nearest 0.5km/h, since those are the
# increments of the scores
v3015IFT.Pre <- round_any(v3015IFT.Pre, 0.5)
v3015IFT.Post <- round_any(v3015IFT.Post, 0.5)

# Put those test into data.frame
testDataWide <- data.frame(Athlete = playerNames, YOYOIRL1.Pre, YOYOIRL1.Post, 
    v3015IFT.Pre, v3015IFT.Post)

# And print first 15 athletes
print(xtable(head(testDataWide, 15), border = T), type = "html")
Athlete YOYOIRL1.Pre YOYOIRL1.Post v3015IFT.Pre v3015IFT.Post
1 Shrestha, Ezell 2000.00 2520.00 18.50 19.50
2 Cha, Gequan 1440.00 2080.00 20.50 20.00
3 Brown, Hindav 2360.00 2040.00 19.50 19.50
4 Venegas Delarosa, Destinee 1640.00 1800.00 19.50 21.50
5 Simon, Barrington 2240.00 1800.00 19.00 19.50
6 Williams, Hyeju 2200.00 2280.00 18.00 19.00
7 Gutierrez, Sabrina 1760.00 2400.00 17.50 23.00
8 Wilder, Johannah 1920.00 1640.00 19.00 22.00
9 Martin Dean, Jillian 1440.00 1960.00 21.00 22.00
10 Thomas, Neil 1840.00 2400.00 19.00 20.50
11 Nosker, Andrew 2080.00 2120.00 19.00 21.00
12 Blackford, Matthew 1760.00 2880.00 19.50 21.50
13 Mata, Rachel 1600.00 2640.00 18.50 19.50
14 Cheng, Ryan 1560.00 2440.00 17.50 21.00
15 True, Ashley 1720.00 2240.00 21.50 21.00

To plot the data and to do simple descriptive stats we need to reshape the data from wide format to long format using reshape2 package by Hadley Wickham

# Reshape the data
testData <- melt(testDataWide, id.vars = "Athlete", variable.name = "Test", 
    value.name = "Score")

# And print first 30 rows
print(xtable(head(testData, 30), border = T), type = "html")
Athlete Test Score
1 Shrestha, Ezell YOYOIRL1.Pre 2000.00
2 Cha, Gequan YOYOIRL1.Pre 1440.00
3 Brown, Hindav YOYOIRL1.Pre 2360.00
4 Venegas Delarosa, Destinee YOYOIRL1.Pre 1640.00
5 Simon, Barrington YOYOIRL1.Pre 2240.00
6 Williams, Hyeju YOYOIRL1.Pre 2200.00
7 Gutierrez, Sabrina YOYOIRL1.Pre 1760.00
8 Wilder, Johannah YOYOIRL1.Pre 1920.00
9 Martin Dean, Jillian YOYOIRL1.Pre 1440.00
10 Thomas, Neil YOYOIRL1.Pre 1840.00
11 Nosker, Andrew YOYOIRL1.Pre 2080.00
12 Blackford, Matthew YOYOIRL1.Pre 1760.00
13 Mata, Rachel YOYOIRL1.Pre 1600.00
14 Cheng, Ryan YOYOIRL1.Pre 1560.00
15 True, Ashley YOYOIRL1.Pre 1720.00
16 Inouye, Connor YOYOIRL1.Pre 2080.00
17 Tatum, Janice YOYOIRL1.Pre 1600.00
18 Latour, Pearl YOYOIRL1.Pre 1720.00
19 Tripathi, Juan YOYOIRL1.Pre 1360.00
20 Moore, Michelle YOYOIRL1.Pre 1880.00
21 O'Sullivan, Johanna YOYOIRL1.Pre 2160.00
22 Sharp, Gregory YOYOIRL1.Pre 2200.00
23 Blum, Jennifer YOYOIRL1.Pre 2000.00
24 Doering, Darius YOYOIRL1.Pre 1200.00
25 Sohn, Kendle YOYOIRL1.Pre 1880.00
26 Horton, Grant YOYOIRL1.Pre 1880.00
27 Waynewood, Nicholas YOYOIRL1.Pre 1640.00
28 Pallen, Raymundo YOYOIRL1.Pre 1800.00
29 Montoya, Simon YOYOIRL1.Pre 1480.00
30 Clark, Bryce YOYOIRL1.Pre 1960.00

From the tables above it is easy to see the difference between wide and long data formats.

Let's calculate simple stats using plyr package from Hadley Wickham (yes, he is a sort of celebrity in R community) and plot them using violin plots, which is great since they show the distribution of the scores

# Subset YOYOIRL1 tets
ggYOYO <- ggplot(subset(testData, Test == "YOYOIRL1.Pre" | Test == "YOYOIRL1.Post"), 
    aes(x = Test, y = Score))

ggYOYO <- ggYOYO + geom_violin(fill = "red", alpha = 0.5) + theme_few() + stat_summary(fun.y = mean, 
    geom = "point", fill = "white", shape = 23, size = 5)


# Subset 30-15IFT tets
ggIFT <- ggplot(subset(testData, Test == "v3015IFT.Pre" | Test == "v3015IFT.Post"), 
    aes(x = Test, y = Score))

ggIFT <- ggIFT + geom_violin(fill = "steelblue", alpha = 0.5) + theme_few() + 
    stat_summary(fun.y = mean, geom = "point", fill = "white", shape = 23, size = 5)


# Plot the graphs
grid.arrange(ggYOYO, ggIFT, ncol = 2)

plot of chunk unnamed-chunk-3


# Calculate the summary table
testDataSummary <- ddply(testData, "Test", summarize, N = length(Score), Mean = mean(Score), 
    SD = sd(Score))
# Print the summary table
print(xtable(testDataSummary, border = T), type = "html")
Test N Mean SD
1 YOYOIRL1.Pre 150 1749.60 309.40
2 YOYOIRL1.Post 150 2246.13 318.41
3 v3015IFT.Pre 150 18.86 1.12
4 v3015IFT.Post 150 20.98 1.09

From the table above we can calculate the percent change.

YOYOIRL1.Change <- (testDataSummary$Mean[2] - testDataSummary$Mean[1])/testDataSummary$Mean[1] * 
    100
v3015IFT.Change <- (testDataSummary$Mean[4] - testDataSummary$Mean[3])/testDataSummary$Mean[3] * 
    100

print(xtable(data.frame(YOYOIRL1.Change, v3015IFT.Change), border = T), type = "html")
YOYOIRL1.Change v3015IFT.Change
1 28.38 11.22

But as mentioned in the beginning of the post, percent change is not the best way to express change and sensitivity of the tests (although it is great to impress the managers or your superiors, or claim that your test is more sensitive).

What we need to do is to calculate effect size (ES). ES takes into account the difference between the means and SD (in this case of the Pre- test, but it can also use pooled SD).

YOYOIRL1.ES <- (testDataSummary$Mean[2] - testDataSummary$Mean[1])/testDataSummary$SD[1]
v3015IFT.ES <- (testDataSummary$Mean[4] - testDataSummary$Mean[3])/testDataSummary$SD[3]

print(xtable(data.frame(YOYOIRL1.ES, v3015IFT.ES), border = T), type = "html")
YOYOIRL1.ES v3015IFT.ES
1 1.60 1.88

From the data above we can conclude that they are pretty similar and that 30-15IFT might be a bit more sensitive (or the Coach B did a better job).

Anyway, to summarize this blog post - start reporting ES alongside with percent change. If someone claims high improvements in testing scores to show how great coach he is, or how great his program is, ask to see ES or the distribution of the change scores or Pre- and Post- tests. Besides we need to also ask for SWC and TE, but more on that later.