This is a pretty good summary of the situation!
I agree with your basic idea.
However, the idea that periodization must be some rigid, unchanging monster is becoming outdated.
The linear periodization concept is most certainly just that…fixed and with little room for variation.
But other methods involving the conjugate sequence system, such as Verkhoshansky’s concentration of loading, or Arosiev(sp?)'s pendulum approach (now being implemented effectively by CT) or Westside, show how periodization can be implemented in a very fluid, as you’re saying autoregulating, manner. Autoregulating training, better known to most as cybernetic periodization, plays a large role in such models.
This is really arguing over semantics, though. The concepts are the same, only the wording is different.
Yes, I see what you’re saying but working to fatigue dropoffs is not used in Westside or Pendulum styles as far as I’ve seen. So how do you know when you’ve lifted enough. Is it 5x5 or 3x8 or 8x3? Why 5 sets? Why not 6? When do you know to stop? It’s guesswork.
Intensity is a significantly more powerful variable than volume. Volume at any given intensity follows the Law of Diminishing returns.
Yes, diminishing returns…when do you know to stop? What if someone has the work capacity to do 10 triples at 90%? Should they only do 5 sets because that’s what’s on the board?
When do you know you’re recovered? Do you guess the same time next week? Muscle recovery? CNS recovery?
In weightlifting I advocate an increase in frequency NOT volume (per session).
I said there were two extremes - one where autoreg is solely used, and one where periodisation is solely used. The extreme of periodisation is an unchanging model where the athlete must adapt to an unchanging plan.
Other methods are somewhere in the middle, and all have different benefits and problems associated with them. The perfect system would take this into account and apply the correct model at different stages of the development phase.
Maybe we as a group should list all the different types of training systems, and their perceived benefits and problems. This would allow us to place each on the autoreg->periodisation continuim and have a good place to start discussion.
A wiki would be brilliant if that were available on the site, as it wouldnt be lost in specific forums.
The one thing that strikes me is that no one is applying the same concept to other fields - nutrition, regeneration etc - the concepts work anywhere. Autoreg to me is just that - a concept - and can be applied in many situations. There are situations where it is not applicable, as opponents have bought up.
I’ll let someone more knowledgeable start this in the right place if everyone is interested.
Depends on your objective. If it’s sprinting, then the critical training times must be monitored, with all training componants adjusted to raise these performances on a regular basis and to be sufficiently recovered for each componant to be carried out at a time within the microcycle that will maximally complement all other componants. Forward planning AND adjustment as necessary.
Completely correct. I am quite often wrong, but could it also be possible that when one wants to increase a specific motor quality - lets say absolute maximal speed - then an autoreg approach may offer benefits over a periodised approach? This could of course only be applied when one doesnt need to peak in the near future, and other training components such as endurance/accel etc arent needed to be trained.
Charlie - what do you think?
There are problems with a solely autoreg approach to max speed:
1: When, in training, should losses in endurance and acceleration not, at least, be minimized? Losses in these areas will only have to be re-gained later at a time when the competition for CNS energy with speed work will more than negate any percieved benefit.
2: Maximal speed session recoveries can extend well beyond the period when re-stimulation of the CNS must occur to optimize tone for the next max speed session.
3: Max Speed development requires the complete extension and body position that only occurs at the highest speeds, so a performance drop-off method for determining an optimal recovery period, would not be practical.
4: The subtleties of max speed work (rythem, coordination, relaxation, and the search for hundredths, if not thousandths of a second) require learning, in digestable bits, over a prolonged period.
While, in the short term, it would be better to operate in a purely reactive way, rather than from purely forward planning, prolonged improvement requires both.
To avoid an early plateau, speed work must be applied in a manner that will create a training rebound, not only by session, but by mesocycle and macrocycle.
This is done primarily via the compression and spreading of the timing of maximal speed work within the constant frequency of speed sessions optimal for CNS re-stimulation, maintained via sub-maximal sprinting in the overall program and secondarily by the adjustment of individual session volumes.
If you understand the limitations of autoreg in peaking, you can see its problem at the meso and macro level.
That said, you must also be prepared to react to changing circumstances affecting recoveries as necessary.
Eric Sevareid once said: “The chief cause of problems is solutions.”
Dogmatic approaches from either extreme will limit development in the long run.
Sometimes, when I read charlies posts, I think Charlie Francis is a genius.
Great info Charlie.
I read ‘1984’ a couple of weeks ago, and it’s my favourite novel so far, but I still prefer speed trap!! 
(I’m not saying any of speed trap is fiction just a good read!)
Sorry no more arse-kissing after this, promise!
I’ll input my comments:
I agree. Now you are getting to my area of expertise. I have written software that deals with this exact problem for other animals which are measured and tested far more easily than humans. What we found was that there are two conflicitng problems in optimisation of organism function:
- The optimisation of all elements reduces the attainment of maximal performance in any single element. i.e. if you want to gain (or maintain) in all elements, then the maximal attainable level in any single element is decreased.
- The focus on one particular element tends to decrease the performance in other elements that have high correlations between them.
What could be useful would be to use the solution in (1) for most of the training year, and the solution in (2) for brief periods for the most lacking motor quality present in the individual.
This doesnt mean that an autoreg method couldn’t take that into consideration, but does mean that finding this period for the individual would be paramount. Again, another problem with applying autoreg at the wrong time.
Thats just one way to do an autoreg method. There are many measures which could be used, such as distance covered at maximal speed for example.
Time is just another variable which could be used, along with everything else which normal systems usually use.
Excellent analaysis! I understand the problems associated with it, but believe it may offer some unique qualities in organism optimisation that normal periodisation cannot possibly attain. Whether it is the right tool, or a cross-spectrum approach must be used is another question.
Extreme approaches will limit development if used in isolation - but maybe not if they are used at the correct time for periods which are useful. Why couldn’t all systems live in harmony -its just a matter of figuring out when each is appropriate 
Joe
Great points Charlie, and to expand on that, pretty much any neurally-dependent quality can be more or less placed in the same boat.
I’m of the belief that as long as goals are made at the mesocyclic level, the rest can be more or less fluid. In general I prefer some structure of the microcycles, but during certain phases even those can go by feel, depending on what’s being trained as the focal point (I’m a big believer in Verkhoshansky’s concentration of loading).
As far as general vs. specific capacities, the most successful training systems work very generally in the off-season to develop everything to some degree, so strength, speed, endurance, agility, technique etc all get some attention (the conjugate sequence system). The idea being that you want to keep the the entire organism in good shape. As it gets closer to competition, you’ll pick the qualities needed and shift the training towards specificity until you’re eventually performing under competitive conditions (or you could use the concentration/intensification approach, either way). In other words, you develop the entire pool of abilities, then pick the ones that are needed to truly dial in.
RE Joecole:
A few clarifications:
1: You state: “If you want to gain (or maintain) in all elements, then the maximal attainable level in any single element is decreased.”
This is, of course, true- if you leave out the “or maintain” part of your equation.
Experience in the taper/maintenance phase shows how little CNS stress may be required to, at the least, delay the drop-off of other qualities. This fits neatly with the fact that you must re-stimulate the CNS more frequently than recovery from maximal speed sessions anyway (EOD vs every 6 to 10 days for top athletes).
2: You state: “The focus on one particular element tends to decrease the performance in other elements that have high correlations between them.” (high intensity- CF)
This corollary to point 1 is 100% correct, and is precisely why training must be balanced between all critical componants, both in terms of the duration of concentration on a given element in the plan- and the need to address the cost on other required elements before it becomes too high.
You suggest that time at maximum speed can be considered in an autoreg system. This is a bit tricky to define. When considering the advancement of capacity for absolute speed, the duration for top sprinters has not been shown to be beyond 20 meters, or 1.66 sec, requiring complete rest between reps. Likewise the number of reps possible will be limited. However, you can think of this 20m segment as a curve, with the highest point in the middle. Initially the curve (vertical axis = vel in MPS, horizontal axis = distance in Meters) will not be symmetrical, as the second 10m segment will drop off more that the first segment rises. This will begin to even out over the training period by repeated exposure to sessions of the SAME type. ( There is a good example of this shown on the GPP DVD)
Much has been made of the concept of a longer duration of “top speed” shown by Maurice Greene, but his coach, John Smith, has been careful to point out that this maintenance is slightly sub-max for him (.84 to .85 per 10m segment vs .83) or, in other words based on the acquisition of speed reserve, which, by definition, must be developed after the highest max speed is already in place.
I don’t understand how maintenance doesn’t do the same thing. Your optimisation function still has the same parameters, just no force to change, so it keeps the same values.
Maintaining the current state of the athlete [in every parameter] may not take much training to do, but it still should inhibit the maximal attainable level in any single element.
Completely agree.
Again, this in an example of many such measures you could use. Can you think og any that might be more suitable?
I guess the interesting thing from an autoreg perspective would be how many reps are able to be completed by an athlete under a specific time on a given day? The number of reps will be limited - to the individuals capacity on the day in question. Measuring the time is the biggest problem - it needs to be electronic as stopwatches have more error than the 6% or whatever error measure you are using.
Completely agree.
joe
[QUOTE=joecole]I don’t understand how maintenance doesn’t do the same thing. Your optimisation function still has the same parameters, just no force to change, so it keeps the same values.
Maintaining the current state of the athlete [in every parameter] may not take much training to do, but it still should inhibit the maximal attainable level in any single element.
CF: The best way to think of this is why we use a maintenance phase for the maximization of performance for competition in the first place. We stimulate qualities, not for gains, but just enough to minimize losses that would rapidly ensue if we didn’t touch these qualities at all. Think of substituting the development of speed as a replacement for the competitions. This might, as you say, inhibit the maximal attainable level in any single element in the shortest term, but the maintenance of complimentary qualities over a longer term will allow for the advancement of the specific quality for a much longer period at only a slight loss of rate of gain.l
[Quote=Joecole]: Completely agree.
Again, this in an example of many such measures you could use. Can you think og any that might be more suitable?
I guess the interesting thing from an autoreg perspective would be how many reps are able to be completed by an athlete under a specific time on a given day? The number of reps will be limited - to the individuals capacity on the day in question. Measuring the time is the biggest problem - it needs to be electronic as stopwatches have more error than the 6% or whatever error measure you are using.
CF: I usually work in sets of 4 with recoveries determined by athlete level and a “flying 20” would use the same recovery as a maximal rep of the total distance (ie 20m + 40m build-up is treated the same as a full-out 60m).
As a determinant, I listen to the foot strikes of the athete in the 20m pass. Any increase in volume is an indicator of a drop in hip height, which would be too slight to be observed. A top athlete might be able to complete 2 sets of 4 with recoveries of 7 to 10min between reps and 15 to 20min between sets, though they would be stopped as soon as there was any indication of deterioration. If, by rep 3 in the frst set, I thought there might be a pending issue, I’d stop the set at 3reps and try again after the full set break for a maximum of 3 more.
Electronic timing MIGHT help, but there are problems with timing trap systems (what part of the body or arms crossed the line to start and stop the clock). Some people have tried counting frames on VHS, which can measure at the same point on the torso like true ET and are accurate to ± .02 (PAL) or .03 (NTSC)
Sure, obviously we were talking about two different phenomena. My communication skills at work again
So in a sense, you are applying a type of autoreg, but instead of using a quantitative measure, you are using a highly tuned & experienced qualitative measure.
joe
So in a sense, you are applying a type of autoreg, but instead of using a quantitative measure, you are using a highly tuned & experienced qualitative measure.
joe[/QUOTE]
—Isn’t that what great coaching is?
—Isn’t that what great coaching is?[/QUOTE]
Yeah completely.
What I am interested in personally is finding ways of taking what the great coaches know from years of practice & experience, and putting it in software so that many other coaches can use it. This conversation validated the software model I am using, so thats great.
The software will never beat the experience of master coaches - but will help all coaches and athletes optimise their training effort to a greater degree.
In our beta testing we have found performance patterns which are interesting and have never been identified by others. We also have found things that are specific to the individual that would take years to uncover by experience or some very good statistical analysis by coaches.
Regards, joe