This is the paper:
http://zachdechant.files.wordpress.com/2010/07/apre-mizzou1.pdfSome interesting things to note about this study:
1. There was no pre-test. For comparison they used the numbers from the end of pre-season of the previous year.
2. Look at the group differences listed in table 1. While no statistical differences have been found, it is apparent that the APRE group on average was 7 kilogram (15 lbs) heavier than the LP group. While maybe not a
statistically significant weight difference, it might very well still be a
relevant difference.
3. The LP group on baseline was stronger in both bench press tests than the APRE group despite being slightly taller and less heavy on average - those differences were significant, one even being highly significant. It is therefore very plausible that improvements for the LP group were harder to come by than for the APRE group, simply because they might be closer to their maximal genetic potential.
4. In accordance to point 3, the LP group actually performed the same in post-test compared to baseline in the bench-press repetition test, actually the difference was even slightly in the negative. There must have been several athletes whose performance actually declined. Remember how the baseline levels were determined in the first place...
5. The differences in improvement between the groups can easily be attributed to the specific application of linear periodization vs. autoregulatory training. Despite the fact that the authors claim to have been utilizing very similar protocols, this is not apparent from the little information they actually give. One thing I can tell however is that the APRE group regularly went to failure and the LP group did not. Therefore the APRE group had the chance to train close to their actual potential while the LP group did not - because their training numbers were determined based on tests from a year before, which might very well have had little to do with their actual ability.
To summarize this... If you make a program based on the concept of linear periodization and have a very, very poor grasp of what you are doing and how you assign training numbers to your athletes, this is likely going to be inferior to training a group of athletes based on what they are actually capable of doing. Big surprise.
As a scientific paper this study is a complete failure and it has very little application for everyone who understands the slightest bit about training.
However, that does not take away from the fact, that AREG might be effective for certain things. As Lance pointed out this approach can be effective when dealing with reactive or power exercises because they are so much more susceptible to even slighter changes in constitution and because doing them correct and fast is probably more important than doing a certain number of them.
I found that not to be true for performing heavy resistance exercises where bar speed might in fact drop, which however says nothing about if the repetition or set can be completed or not. Strength exercises are for strength, power exercises are for power. That's my idea of it.
