Announcing Version 2.1 WKO+ with Performance Manager!

[Alex Simmons]

I'm slightly confused with some of the values that need to be entered into Performance Manager

The CTL and ATL constants are fine (I've just left them at 42 and 7) but the starting values has me confused.

The recommendation is to multiply average hours per day, lets say 2 hours, by 0.50 to 0.75 which would give 1.0 or 1.50 (am I looking at this the right way

) but only integer numbers can be entered (1 or 2 etc) so how would you enter say 1.5

Any help or advice would be great

I think there is a factor of 100 missing from the advice posted.

Start values are not critical in the medium/long term, just means the first few weeks of the plot may not be as "accurate" for the purpose of validating performance.

Add up your TSS for the fortnight prior to the plot start date and divide by 14. Use that for both start values.

The chart will sort itself out after a few weeks of real data (as long as you have TSS for all days). Hopefully you have more than a few week's worth of data to plot. Otherwise - patience while the data accumulates!

 

[Aztec]

Check out this article I wrote. It has some guidelines regarding those questions.

http://www.cyclingpeakssoftware.com/power411/performancemanager.asp

Hunter

P.S. Andy's MONSTER Science article should be going up today sometime this afternoon.

Can you tell me what the little X is at the left end of every row in the Calendar? I love seeing IF/TSS on there, by the way.

 

[Terry Ferguson]

Can you tell me what the little X is at the left end of every row in the Calendar? I love seeing IF/TSS on there, by the way.

The x is whether or not you want it to show up on the chalkboard. - TF

 

[AshesGlory]

I just upgraded to 2.1 and I'm having a problem with the linking function. In the older 1.x version I could rename my first interval then link the other intervals and they would be automatically renamed. In 2.1 when I link it merely appends :2 etc to the existing interval name so I have to rename each interval individually. According to the user guide:
1) Click on the ranges you want link together.
2) This prevents you from having to retype the name multiple times and it also will give you the rest period between each linked interval in your Summary Section on your Journal page.

What am I doing wrong?

 

[Hunter w/kg]

I just upgraded to 2.1 and I'm having a problem with the linking function. In the older 1.x version I could rename my first interval then link the other intervals and they would be automatically renamed. In 2.1 when I link it merely appends :2 etc to the existing interval name so I have to rename each interval individually. According to the user guide:
1) Click on the ranges you want link together.
2) This prevents you from having to retype the name multiple times and it also will give you the rest period between each linked interval in your Summary Section on your Journal page.

What am I doing wrong?

Nothing. We have a small bug with that. We are working on it.

Hunter

 

[kennethn]

I just upgraded to 2.1 and I'm having a problem with the linking function. In the older 1.x version I could rename my first interval then link the other intervals and they would be automatically renamed. In 2.1 when I link it merely appends :2 etc to the existing interval name so I have to rename each interval individually. According to the user guide:
1) Click on the ranges you want link together.
2) This prevents you from having to retype the name multiple times and it also will give you the rest period between each linked interval in your Summary Section on your Journal page.

What am I doing wrong?

I've found that if you do the reverse - link a bunch of intervals, then rename the first one - that it works fine.

 

[AshesGlory]

I've found that if you do the reverse - link a bunch of intervals, then rename the first one - that it works fine.

Excellent, thanks.

 

[jeffh]

I've found that if you do the reverse - link a bunch of intervals, then rename the first one - that it works fine.

Actually, if you create a set of interval ranges and keep the default name for all but the first one that heads the linked set, the additional ranges will take on the name of the head range.

Basically, create the first set with ctrl+R and give it a meaningful name. Then, select each additional range and press ctrl+R. Finally, select the first range and click the link button for each of the following ranges.

If you name each range individually, then each will retain it's name but append the linked number.

-jeff

 

[zaskar]

Can someone answer in short why my CTL would start out from 106 and continue to dive? even though i was building volume in this period 1-06 -4-30 -06? from 9-1 to 9-30 for 3weeks tss 1000 a week and highest it went is to 105. this is the most tss i have ever done, but CTL wont go over 105.

 

[Alex Simmons]

Can someone answer in short why my CTL would start out from 106 and continue to dive? even though i was building volume in this period 1-06 -4-30 -06? from 9-1 to 9-30 for 3weeks tss 1000 a week and highest it went is to 105. this is the most tss i have ever done, but CTL wont go over 105.

I don't quite follow the dates you quote but generally if training is consistent and your current average daily TSS is higher than your present CTL level, then CTL should rise. Would prob need to see your daily TSS pattern over a few months to properly answer this.

 

[zaskar]

I don't quite follow the dates you quote but generally if training is consistent and your current average daily TSS is higher than your present CTL level, then CTL should rise. Would prob need to see your daily TSS pattern over a few months to properly answer this.

Ok, looks like my starting value is to blame. i thought in starting value it is the avg hours per week you ride? any way i put 0 and now it looks like it makes sense.

 

[lanierb]

Ok, looks like my starting value is to blame. i thought in starting value it is the avg hours per week you ride? any way i put 0 and now it looks like it makes sense.

Definitely not average hours trained per week. It should be average TSS/day (or average TSS/wk/7).

 

[zaskar]

Definitely not average hours trained per week. It should be average TSS/day (or average TSS/wk/7).

I'm confused,on Peaks website it says this....


You might also want to change the STARTING VALUE for both the CTL CONSTANT and the ATL CONSTANT. We recommend that you multiply the average amount of hours per day that you ride, by .70, and that should give you a starting point. Use this value for both the ATL and CTL.


When i do that the chart makes no sense either.

 

[jeffh]

Ok, looks like my starting value is to blame. i thought in starting value it is the avg hours per week you ride? any way i put 0 and now it looks like it makes sense.

Starting values can be tricky. But, if you've been riding consistently for more than a few months (and downloading your PM every ride), then you shouldn't enter a starting value.

In other words, starting values are the ATL and CTL that should be used as of the first date displayed in the chart.

Another "trick" for using the PMC is to choose an ending date into the future. This will give you a sense of how quickly (or not) TSB increases and CTL decreases.

-jeff

 

[jeffh]

I'm confused,on Peaks website it says this....


You might also want to change the STARTING VALUE for both the CTL CONSTANT and the ATL CONSTANT. We recommend that you multiply the average amount of hours per day that you ride, by .70, and that should give you a starting point. Use this value for both the ATL and CTL.


When i do that the chart makes no sense either.

That's definitely wrong. If you ride a 1 hour TT everyday, seven days a week for months on end, your TSS/d would be 100. Your starting TSS/d (if it's even necessary - see previous post) should be your average weekly total TSS volume divided by 7.

Again, if you've been downloading every workout for more than a few months, then leave your beginning values set to zero.

-jeff

 

[joemw]

i would just like to state for the record, that both myself and my cheif analyst are a bit peeved that he can't look at my files i've downloaded to wko on my laptop with his version of the CP software, which is the previous version. lame way to force people to upgrade.

 

[Alex Simmons]

i would just like to state for the record, that both myself and my cheif analyst are a bit peeved that he can't look at my files i've downloaded to wko on my laptop with his version of the CP software, which is the previous version. lame way to force people to upgrade.

I'd be more peeved if my "chief analyst" hadn't bothered to invest in upgrading by far and away the best power analysis tool on the market, all for less than the cost of one decent race tyre....

 

[sidewind]

I'm confused,on Peaks website it says this....


You might also want to change the STARTING VALUE for both the CTL CONSTANT and the ATL CONSTANT. We recommend that you multiply the average amount of hours per day that you ride, by .70, and that should give you a starting point. Use this value for both the ATL and CTL.


When i do that the chart makes no sense either.

Seems for me that they are suggesting a generic estimate for TSS would be 70 per hour ridden. Anyway, it should then be 70 * hours/day, not 0.70

 

[frenchyge]

I'm confused,on Peaks website it says this....


You might also want to change the STARTING VALUE for both the CTL CONSTANT and the ATL CONSTANT. We recommend that you multiply the average amount of hours per day that you ride, by .70, and that should give you a starting point. Use this value for both the ATL and CTL.


When i do that the chart makes no sense either.

Well, now you're talking about the *time constants* used in the calculations, rather than the starting values for CTL and ATL themselves. Anyway....

Yes, that wouldn't make any sense because it would produce the same time constant for both ATL and CTL. Are you sure that's what the guidance really says? IIRC, the CTL time constant should be several times larger than the ATL constant.

 

[jeffh]

Well, now you're talking about the *time constants* used in the calculations, rather than the starting values for CTL and ATL themselves. Anyway....

Yes, that wouldn't make any sense because it would produce the same time constant for both ATL and CTL. Are you sure that's what the guidance really says? IIRC, the CTL time constant should be several times larger than the ATL constant.

The "Starting Values" basically override the calculations of the system (which actually go back and start by analyzing the very first file that you had ever logged). So, they should only be changed by someone who either 1) hasn't been downloading and logging their rides consistently or 2) by someone who just started using a power meter but has been training all along. A good starting value for both ATL and CTL in this case would likely be 70 x Avg daily riding hours over the past few weeks or months that occured immediately prior to the starting date of the chart.

CTL takes approximately "3 x # of days" of data to level out (very generalized math here). So, if you've been recording your workouts diligently, you shouldn't add a starting value of anything. Likewise, if you haven't been training for the past 4-5 months, 0 is also a good value to start with.

Regarding the time constants, 42 days (CTL) and 7 days (ATL) will fit most everyone very well. It's fun to try other values and sometimes worth the look. But, generally 42 and 7 are recommended unless someone is incredibly diligent with their recoveries or an exceptionally fast or slow responder.

-jeff