Feb 23, 2011

Golden Cheetah quick start guide

So you have a power meter and you've been riding for a while and uploading your rides to something or other, but you don't really know what any of that means. Well, hopefully this will be of some use to you. First I must get a few details out of the way. None of what I'm about to write was invented by me. I'm merely providing you with the information that I have researched, read and digested. For sake of academic honesty, I will say that most of the things in this blog post have their origins in Training and Racing With a Power Meter, a book I urge you to buy and read cover to cover. 

The first thing that perplexed me when I started riding with my Powertap was what to do with all the data. As I previously posted, TrainingPeaks doesn't offer a Mac version and costs some money, and Golden Cheetah works with just about anything and is absolutely free. The downside of Golden Cheetah is that it doesn't really come with any sort of instructions or explanations of what everything means. This is a gap I'm hoping to fill with this post. There are a few things, however, that this blog won't cover: I'm not going to talk in any great detail about the actual process of uploading rides or playing with settings in Golden Cheetah, you can get all of that from their wiki page. I'm also not going to get into stress balance, short term stress and long term stress, as those concepts deserve a nice long blog post of their own and that will come at some point once I feel I have a firm grasp on the subject myself. In Golden Cheetah this relates to Daniels, Skiba, TRIPM and BikeScore metrics under the Metrics tab. 

Once you've connected your device and downloaded the file to Golden Cheetah, the first tab shows you your ride summary. If you did one continuous ride, it will show you that you've done only 1 interval and will show you several intervals within that interval, for example, distance, avg. cadence, 5 sec max, etc. If you did intervals and somehow split those up on your bike computer, you will get that data for each of the intervals. You can actually choose which totals will be displayed in the summary from the settings menu. From the picture below, you can see the ones I have chosen.

Now, look at the Mertics column above, the first item is xPower. This is what TrainingPeaks refers to as normalized power (NP). If you look at my average power in the Intervals section, you will see it says 169, but my xPower is 221. The latter is the true representation of my effort, while the former is an average of the power outputs at every second of the ride. So, for example, if I were to ride a stationary bike hooked up to a power meter, so that I could produce more even power (without wind, road unevenness, traffic distractions, etc.), and average 221 watts, my xPower would probably be very close to the same number.

If you went out for a ride and did a bunch of intervals, you may want to take a look at them in ride plot mode. You can zoom in and out by clicking on the little arrows next to the "Grid" check mark at the bottom and really get a close up look at each of your intervals. Below is my last hill repeats workout.

Above, you can see my intervals with rest periods in between. I've removed all the smoothing so you can see the power changes at each second of the interval. This is helpful when you are looking at the data close up, but when you want to look at the ride as a whole, you might want to smooth the numbers to get a trend line. You can do this by adjusting the number at the bottom (it currently says 1). The higher the number, the greater the number of seconds over which the data is averaged. Looking at your power intervals in this way can help you determine whether you went too hard in the beginning and faded, or whether you build up in the interval, or if you kept it steady. If you look at my first interval (segment 2), you can see that I started conservatively and gradually increased power as the interval ended. You can also see that I stopped pedaling first and pushed the lap button on my Garmin second, as the power line drops drastically at the end of the interval. I recommend you look at all of your intervals this way and see exactly what happened with your effort within the interval.

Now let's move on to one of the most important things that Golden Cheetah can show you - your critical power plot. This not only shows you your best ever 5 seconds, or your best ever 10 minutes, but you can see your best ever 1 minute 42-second time, or your best 23-minute time. All of these critical power values can help determine what kind of rider you are.

Your power levels are divided into 7 zones: neuromuscular, anaerobic, VO2Max, threshold, tempo, endurance and active recovery. Each is a percentage of your best one hour. Chances are that you've never gone all out with your power meter for one hour because it's hard to do outdoors uninterrupted and very, very boring. The dotted line in the above graph is a critical power curve predicting what your critical power would be for certain times based on the real data represented by the red line (the black line is the ride actually highlighted on the left side of the screen which is not pictured as it is irrelevant here). I will say that the CP curve is not always acurate and it's best to go with your actual values, but if you really have no clue, it will be useful. For example, the CP curve says that my best one hour is 280 watts, but I know that it's actually about 300, so that's the value based upon which I set all of my zones. You can do this in Golden Cheetah by selecting "Preferences" and adjusting your CP value. The power zones will be automatically calculated for you.

After you've been riding for a while and you've accumulated enough data to have some very good best times, or you simply complete a CP test, testing your best efforts at all the critical times, you can refer to the chart below to determine what your strengths are. So let's take my values from the above critical power curve and see what they mean when  plugged into the chart below. My best 5-second time is 1240 watts, which divided by my body mass of 75kg = 16.5, which is at Cat. 3 level. My 1 minute time is 541 watts, which equals to about 7.2, barely reaching into the Cat. 4 category. Interestingly, when we look at my 5-minute time, it's 371 watts or 4.95, which again puts me at the higher end of the Cat. 3 ranking. FT is the equivalent of the 20-minute time and for me equals 297 watts, or 3.92, again in the Cat. 3 category. So how can you use this. 

Each of the times below are representative of different power zones and depending on where you land on your best effort in each, you may have a better idea of what areas you need to train and where your weakness is. Personally, I'm still a bit perplexed at my 1-minute time. It could be that it's low because all of my 60-second efforts come second after my 30-second efforts, or maybe I've always been just a touch too fatigued to give it my true best, or perhaps that's the area that requires the most training, but as I accumulate more and more rides and data-points, I think that will sort itself out.

  Allen & Coggan Race Category Table

5 s
1 min
5 min
5 s
1 min 5 min

7.60 6.40 19.42 9.29 6.61 5.69

7.50 6.31 19.20 9.20 6.52 5.61

11.27 7.39 6.22 18.99 9.11 6.42 5.53
 World Class
11.16 7.29 6.13 18.77 9.02 6.33 5.44
 (e.g., international pro)
11.04 7.19 6.04 18.56 8.93 6.24 5.36
10.93 7.08 5.96 18.34 8.84 6.15 5.28
10.81 6.98 5.87 18.13 8.75 6.05 5.20
10.70 6.88 5.78 17.91 8.66 5.96 5.12
10.58 6.77 5.69 17.70 8.56 5.87 5.03
10.47 6.67 5.60 17.48 8.47 5.78 4.95
 (e.g., domestic pro)
10.35 6.57 5.51 17.26 8.38 5.68 4.87
10.24 6.46 5.42 17.05 8.29 5.59 4.79
10.12 6.36 5.33 16.83 8.20 5.50 4.70

10.01 6.26 5.24 16.62 8.11 5.41 4.62
9.89 6.15 5.15 16.40 8.02 5.31 4.54
9.78 6.05 5.07 16.19 7.93 5.22 4.46
 (e.g., Cat. 1)
9.66 5.95 4.98 15.97 7.84 5.13 4.38

9.55 5.84 4.89 15.76 7.75 5.04 4.29
9.43 5.74 4.80 15.54 7.66 4.94 4.21
9.32 5.64 4.71 15.32 7.57 4.85 4.13
9.20 5.53 4.62 15.11 7.48 4.76 4.05
Very Good
9.09 5.43 4.53 14.89 7.39 4.67 3.97
 (e.g., Cat. 2)
8.97 5.33 4.44 14.68 7.30 4.57 3.88
8.86 5.22 4.35 14.46 7.21 4.48 3.80

8.74 5.12 4.27 14.25 7.11 4.39 3.72
8.63 5.01 4.18 14.03 7.02 4.30 3.64
8.51 4.91 4.09 13.82 6.93 4.20 3.55
8.40 4.81 4.00 13.60 6.84 4.11 3.47
(e.g., Cat. 3)
8.28 4.70 3.91 13.39 6.75 4.02 3.39
8.17 4.60 3.82 13.17 6.66 3.93 3.31
8.05 4.50 3.73 12.95 6.57 3.83 3.23
4.39 3.64 12.74 6.48 3.74 3.14

4.29 3.55 12.52 6.39 3.65 3.06
4.19 3.47 12.31 6.30 3.56 2.98
4.08 3.38 12.09 6.21 3.46 2.90
 (e.g., Cat. 4)
3.98 3.29 11.88 6.12 3.37 2.82
3.88 3.20 11.66 6.03 3.28 2.73

3.77 3.11 11.45 5.94 3.19 2.65
3.67 3.02 11.23 5.85 3.09 2.57
3.57 2.93 11.01 5.76 3.00 2.49
3.46 2.84 10.80 5.66 2.91 2.40
(e.g., Cat. 5) 
3.36 2.75 10.58 5.57 2.82 2.32
3.26 2.66 10.37 5.48 2.72 2.24

3.15 2.58 10.15 5.39 2.63 2.16
3.05 2.49 9.94 5.30 2.54 2.08
2.95 2.40 9.72 5.21 2.45 1.99
2.84 2.31 9.51 5.12 2.35 1.91
(e.g., non-racer)
2.74 2.22 9.29 5.03 2.26 1.83

2.64 2.13 9.07 4.94 2.17 1.75
2.53 2.04 8.86 4.85 2.07 1.67
2.43 1.95 8.64 4.76 1.98 1.58
2.33 1.86 8.43 4.67 1.89 1.50

Now it's time to get really analytical. Golden Cheetah displays a lot of data in histogram form: power, cadence, heart rate and speed. Power and cadence are the most important. The histograms will show you exactly how much time you're spending in each power zone or at each cadence speed. You can also adjust the units by sliding the "bin width" arrow for smoothing or for a more detailed look.

Time (%) vs. Cadence
Looking at the histograms above, you can see how much time I spent in each zone during the ride. As you can also see, the 0 column is very tall in both graphs - this is a good thing. This means that I didn't pedal or produce power during about 18 percent of the ride. This may have been due to descending, or due to drafting, or simply coasting. If you analyze your race data, you should have about 15 percent inactive time, which would represent that you are properly conserving energy during the race and are able to recover and have energy for the hard efforts. I will stress that looking at these histograms in a vacuum is not very useful. Consider how the ride went and then look at the charts. Look at a ride where you felt you performed poorly and then look at another where you thought you did well. Is there a difference? What is it? What can you take away from that data? Did you spend too much time in a certain power zone and burned too many matches, or was your cadence too fast/slow? All of those can be answered with histograms and you subjective assessment of your ride.

The next graph (PF/PV) will help you determine exactly what kind of biker you are. Are you more of a masher, or a spinner. As you can see from the graph below (or you will see once I explain what it means), I'm more of a masher as I produce my highest power during the slower pedaling velocity. That's exactly what the graph is supposed to show. The graph is divided into four quadrants and represents four zones: I - high power, high velocity; II - high power, low velocity; III - low power, low velocity; IV - low power, high velocity. If you look at my chart below, you can see that when I produce high power outputs, the highest concentration of points is in the second quadrant. On the other hand, when I'm turning a lower gear, I tend to prefer a higher cadence, which makes sense - we tend to spin faster when we are pushing a lower gear.

Now, as above, in a vacuum, this data doesn't offer much, but combined with qualitative data, it may provide some helpful insight. How did you feel during the training ride, or a race? How were your intervals? Were you trying to work on your pedal stroke? Think about those things as you look at your PF/PV quadrant analysis chart and see if you want to make some changes in your pedal stroke. Next time you do a stage race with a power meter, compare your crit PF/PV plot with your TT plot. I bet there will be quite a difference between the two.

Now it's time to talk about fatigue resistance. The numbers chart above should have helped you get a better perspective of what your numbers mean on a grand scale, but what do they mean in relation to each other. What good is it to have  1240 watt 5 second sprint if it only takes you 50 yards and the sprint in a crit starts from 400 yards out? How does your power drop relative to time. 

The above graph plots my best times over the last several weeks. I can also set the graph to display the values by days in the drop down menu, but that gets a bit too squiggly, especially for this purpose. You are not looking at the values themselves necessarily, but rather at the distance between the lines. Where the lines are close together, the fatigue resistance is either average or above average. However, where the lines are further apart, chances are that the fatigue resistance is below average. Now, it's important not to get discouraged after learning that your "power numbers" are below average because that's really not what this graph is telling you. How above or below average your numbers are is determined by your critical power graph and the chart above, this simply tracks the changes in those numbers as time progresses. So let's take a look at the graph above. You can see that for the most part, my 10-second, 15-second, 20-second and 30-second times are very close together. This means that there isn't a lot of power being dropped as I go from 10 to 20 to 30 seconds. On the other hand, you can see that the drop from the 1-second peak to my 5-second time is greater, and so is the drop from my 30-second time to my minute time. Of course, there is a much larger time gap going from 30 seconds to one minute than from 20 seconds to 10 seconds, so it only makes sense that the gap between the lines is larger.

Now looking at my aerobic numbers. We can disregard the last set of points, as it seems that I just had one hard 10-minute effort and went easy the rest of the way, which is actually how I remember the last ride. Looking at the remaining data points, you can see that there is relatively little drop going from 20 minutes to 30 minutes, and the gap widens as we get to 60 minutes, or go back up to 10 minutes. This is really not a very good representative sample of my best data because I never really tested myself at my best 30 minutes of my best 20 minutes, so the below are probably not all out efforts, but just power outputs as they happen, which is why the week of February 9 had a higher 30-minute number than the 20-minute number in the three preceding weeks. However, this gives you an idea of how your data will be presented and what it will mean. So if we pretend this is a good sample of data, we can say that my power drops off drastically going from 10 to 20 minutes and drops off very little from 30 minutes to 60 minutes. To verify this, we go back up to my critical power curve and note the angle of the slope as we move between the times. A steeper slope indicates lower fatigue resistance - just note the drastic change in slope at 7 seconds. 

Once you figure out where you are dropping the most power, you can train those times with specific intervals to improve your fatigue resistance at those points. There are many other graphs and metrics you can choose, but most of them (not including the ones about stress) are self explanatory, like amount of time in each power or heart rate zone, or elevation gain, or distance. They don't really require any interpretation.

I realize that this is a lot of information, and there is a good chance I confused a lot of you, or maybe I only confused some of you and dumbed it down way too much. I don't know. However, if there is anything you'd like me to expand upon, or clarify, or amend, or supplement, please comment and I'll do my best to get that done. 


  1. Great work. Very helpful and appreciated. The only part I'll be searching for clarification on is the PF/PV quadrant numbering.

    Is top left 1, top right 2, bottom left 3, and bottom right 4 ?

    If so that would make me think your statement that you produce higher power at lower cadence not seem to match my take away when looking at your graph as there is a higher power dots /concentration in top left.
    Again.. Maybe I just didn't understand that part fully.
    thanks again

  2. Hi, Jon. Thanks for the comment. I think you have the quadrants reversed. Number 1 is the top left, and then rotating counterclockwise from there. Hope it makes a bit more sense now.

  3. Err. Sorry, ignore the above. Quadrant 1 is top right, then rotating counterclockwise.

  4. Any information on understanding the different stress scores?

  5. I haven't put anything together myself (I should), but if you get the book I mentioned in the beginning, Training and Racing With a Power Meter, it explains the concepts fairly well.

    The main idea to keep in mind with stress scores, regardless on which scores you use, is that it's a balance of your short-term stress, long-term stress, and form. You want to be fresh as you peak, but not so fresh that you lose fitness.

  6. "Personally, I'm still a bit perplexed at my 1-minute time."

    I'm guessing that you got this sorted out since you just posted part 2. If not, have you gone out with the intent of only measuring your 1 minute power? I did 2 one minute intervals recently after a warmup, and had noticeable power fade on the second interval.

  7. I'm not sure what you mean by "1-minute time." It's a minute by definition?

    Yes, I've done CP1 tests. I suspect you didn't recover enough between your first and second interval. The rule of thumb is that the shorter the interval, the longer the RBI (rest between intervals). So if I'm doing 10 second bursts, I can take perhaps 50 seconds to recover before the next one. For all out 1-minute efforts, I normally take 2 minutes to recover. However, for those 1-minute intervals I pace myself as if I'm doing however many I'm doing. If I was to go out there and do just two all out 1-minute efforts, I'd probably take about 5 min RBI, just to make sure my systems got back to normal.

  8. This manual is super. I was lost using Golden Cheetah until I found your blog. Thanks.

  9. Thanks so much for these blogs, Open Cheetah needs some documentation, they should point users here, great work on this and particularly the follow up !

    You should link to the 2nd part.

  10. Thank you for the overview.
    One thing to note regarding the histogram. If you are analyzing a training ride it is definitely not a "good thing" for the 0 bin to be so big. Unless you are going down a lot of full speed descents, you should not be coasting around so much. If you are looking at a race, as you suggested, then yes perhaps, but the distinction should be made.

    1. You're absolutely right. But as I said in this blog, this isn't coaching advice, just the basics of how to use GC. If I was giving coaching advice I'd say you scrub your brake while rolling downhill to get more even power output during the workout :-). Thanks for reading!

  11. Thanks for the write up, great info!
    Regarding the issue you have with 1min CP... I seem to have the same issue. I have lots of data and done a CP test, and have the major dip in the power profile matrix at 1 min as well. Coggan describes it as - "an unlikely combination" here - http://home.trainingpeaks.com/blog/article/power-profiling
    Have you any suggestions why 1 min CP is so low? I have been riding/racing long enough to know Im not exceptionally different in comparison to my team mates/competitors. Any ideas?

  12. Hi, great guide, thanks. I think some of the confusion in the replies regarding the QA graph are because in the text where you describe the quadrants, you are using "Power" where you mean "Force". Power is related to the shading. Thanks again for your efforts!