On October 21, 2010, the IIG Los Angeles conducted a test of the “Power Balance” bracelets in conjunction with Yahoo! News.
The Power Balance products are said to allow athletes to perform at their best, maximizing strength, balance, and flexibility . The active component giving the bracelet this ability is a hologram embedded in the plastic wristband.
To conduct the test, the IIG recruited 16 volunteers as subjects, including former Olympic medalist Dominique Dawes. The subjects ran an obstacle course including a balance beam section, and a section requiring them to carry weight on an obstacle course. Each participant ran the course four times, each time with a different bracelet, giving a total of 64 scores. Each run was timed with a stopwatch.
There was one “legitimate” Power Balance bracelet, and 3 “sham” bracelets that had had the hologram removed from them. The experiment was double-blind, all bracelets were wrapped with tape so no one present knew which bracelet was real and which were fakes. Neither the participants nor the people recording the scores knew which bracelet was which until after all participants had completed their runs and their scores were recorded.
Obstacle Course Results
For the obstacle course the assumption was that if the bracelet was effective, all or some participants should have obtained a better score, i.e. a lower time when wearing the real Power Balance bracelet. In fact, half the participants did slightly better with the real bracelet, and half did slightly worse – exactly as would be expected by chance. Figure 1 shows the results of the obstacle course. Bear in mind that a lower time is better. The “real” Power Balance bracelet was the one identified as “C”.
In Figure 1 we can see that the average performance of all 16 participants appears to actually be slightly worse with bracelets B and C, although the difference isn’t statistically significant. The dashed lines indicate the boundary of results that would indicate a result significantly exceeding chance (either helping or hurting performance compared to “average” performance.)
In addition to the obstacle course, all of the Participants also stretched on a fixture that measured flexibility. There were some known problems with this portion of the test, the main one being that it was fairly difficult to get an accurate reading–there was some subjectivity in reading the position of the participant’s fingertips on the scale, and we realized later there were several ways the reading could be in error, for example with the person twisting their body. So the potential measurement error is larger, meaning even more substantial results would be required to indicate effectiveness of the product. The “zero” point used on the scale for this test is arbitrary, and equal to the shortest reach measured for any participant. Figure 2 shows the results.
Again, the dashed lines indicate significant results, and again we see no statistically significant effect from the Power Balance bracelet “C”. And approximately half the participants did slightly better with bracelet C, and half did slightly worse, just as with the obstacle test. The differences from one participant to the next were substantially larger than any difference due to which bracelet was worn.