The Secret of Running

How to get faster and fitter!


Do you have any comments or remarks? Please place them here.


  • Andre Pletschette says:

    Hello, I am really interested in your work. And I defintely want to study your book for my next Marathon 🙂 I just wonder why there are so many differences for heart rate zones. Most authors I know suggest <70% MHR for RECOM, and for example here I read that I should train <70% MHR for base training.

    • Dear Andre,

      You should not worry too much about the heart rate zones. The training effect of <70% MHR is limited to the 'groundwork"of buildling a foundation base and to put sufficient mileage in the tank. The higher training zones will give you the additional training impacts of increasing FTP, VO2 max, speed and RE. In practice during training the zones will overlap.

      Best regards and good luck with your training and marathon!

      Hans van Dijk

  • Fabio Pillon says:

    I have another question.
    When I finish a running session, I have 2 data about elevation, gain and loss. It’s normal that I have session where I start and arrive in same place, so how I can use those elevations for calculate power of my session? I must use your formula P=(i/100)*mgv one time for gain and another time for loss elevation and I use difference of those two results or I must do difference of it and use it for calculate inclination (that become usually near zero) or what?

    • Dear Fabio,

      For calculating Pc you should differentiate between level gain and level loss, as the muscle efficiency is different. Uphill the work is positive, so the muscle efficiency is higher. Downhill, the work is partly negative (braking forces) so the muscle effciciency is lower. This is explained in more detail in our chapter on hills.

      Best regards,

      Hans van Dijk

      • Fabio Pillon says:

        I read your book, but I didn’t find an answer about my question that I think is good.
        In my Valencia Marathon (42,43Km in 3h30’59”), some days ago, I have elevation between 0 and 34m (correct with Garmin correction) and I have total gain of 408m and total lost 413m record on Garmin Connect (with Garmin correction). So at the end in your calculators is correct for you if I use a pendence of (408-413/2)/42430=0.0047 or 0,47% (downhill is neat to half gain than time lost with uphill, right?) and my total weight in formula is my wight more shoes and wear? And about wind, if I know that I had wind between 5 and 10Km/h I must use (5+10)/2/3.6=2,08m/s or less because was not only wind in front of me, but sometime back or at my side? Thanks

      • Fabio Pillon says:

        In same marathon I see that I have quote from 0 to 34m. So maybe is more better use simply 34m like gain (but it’s not total gain and total less, so I don’t think that it’s ok in calculator…) on half distance for calculate energy? I have 34/21215=0.16% of pendence, more less than before. You think that is better calculate like this or how you calculate?

  • Fabio Pillon says:

    I had read your book and I have many questions about it.
    I found some errors about formulas, and I show you it in my next letter.
    First of all I want ask you something about air resistance: I see that you use for CdA 0.24mq, and that you tell that you use this valor with Cd at 0.9. It mean that A is 0.267mq but this area is really too much small for man body. I’m 1,76m for 58Kg and with Autocad I calcolate from my frontal photo that I have 0.53mq of surface. I read some test about ski man that my tall when a man is not band is around 0.48mq, that is really near about my calculate. This test tell that Cd is 0.76 for ski man, so at the end CdA for me is around 0.40 and not 0.24. How is possibile?

    • Dear Fabio,

      Sorry for replying this late, this was caused by spam attacks to our website.
      The answer to your question is twofold. First of all, we have determined the CdA of 0.24 m2 from the results of wind tunnel research in literature (Davies and Pugh). We believe this provides the best estimate of the air-resistance in running. The second part of the answer is why is so low, because indeed one might expect of CdA of around 0.40. The explanation for this is the fact that the muscle efficiency is higher in case of positive work, such as in running with a head wind and also in running uphill. This means thet the power required to surmount the air-resistance is lower than calculated with the theoretical formula. WE have explained all this in more detail in a paper on the Dutch page of our website.

      Best regards and good luck with your running!

      Hans van Dijk

  • Brian Dye says:

    Dear Hans and Ron,I have read your book in detail and your latest articles published in Dutch on the new Stryd power meter (through translation) and in particular your treadmill and intervals experiments. I have the new Stryd; and on 29 September I am running the Berlin marathon (nearly flat for the whole distance so hills are not a factor) and only a couple of weeks ago I did a 32.2k race using the old Stryd and I have the wind and other details for that and so have recent personal data for a comparable long distance. If there any experiment concerning this real life situation you might be interested in collaborating on that would be helpful, please let me know: you could use the email address I’ve supplied to contact me, if you wish. If this is not helpful, it’s no problem. Kind Regards and Best Wishes and Good Luck for your own training, Brian

    • Dear Brian,

      I hope your marathon went well! We are always interested to discuss experiences with our readers. This is fun and it may provide new insights, so we wellcome any data and thoughts that you want to share.

      Best wishes and good luck with your running and studies!

      Hans van Dijk

  • Steve Middlebrough says:

    Hi Hans and Ron Really interested in your power calculation PDF , and I am trying to use it to verify some tests I am undertaking in a masters thesis on performance in athletes
    In the pdf ,I understand P=Pr+Pa+Pc , Pr and Pa can see the calculation but
    it didnt explain how to calculate Pc = imgηv as it assumed a level course
    in PC I understand m and v weight and speed but not i,g and η
    can you share this info

    many thanks


    • Dear Steve,

      First of all, we want to apologize for replying this late to your question as a result of spam-attacks to our English website.
      On al level course Pc= 0 because the gradient i (which is the level difference devided by the length in %) is 0. G is the gravitational constant (9.81 m/s2) and eta is the hill factor, which is (45.6+1.1622i)/100.
      We hope this answer will still be of use to you.

      Best regards and good luck with your running and thesis!

      Hans van Dijk

  • Brian Dye says:

    Dear Hans and Ron,

    Congratulations on your book, and on your work, from me as one of your English fans. I hope your training and running is going well.

    I thought it would be useful to draw to your attention a two part analysis done by a Mr Ian Williams, published in our Guardian newspaper (link below). Mr Williams did a statistical analysis of 1071 marathon runners who logged their training data on one of the running websites in the UK. He found that the Riegel formula, reliable for shorter distances, broke down as a reliable predictor of marathon performance: 95% of the runners analysed were outside its prediction. Hans, I think you say, you yourself fall outside the Riegel predictions in the case of marathons. Mr Williams also sought to analyse, from the website’s data, training factors which were helpful enabling some of the runners to come close to meeting the Riegel prediction, or not, as the case may be. It may be food for thought as to whether, in the case of non-elite, runners, it might be reasonable to adopt a two exponent approach: Reigel for shorter distances, say up to 1/2 marathon, and another exponent for a distance such as the marathon where the risk of fatigue or glycogen depletion is a heavily influential factor. I think you’re already thinking to some extent along these lines in the book. For interest, therefore, here is the main link from which other links can be found to explore this topic:

    Kind Regards,
    Brian Dye

    • Hi Brian,

      First of all, I want to apologize for replying this late: our English website has been experiencing spam-attacks as a result of which we have been down for a long periond.
      Second, I want to thank you for drawing our attention to the interesting work of Mr. Williams, who definitely has a point with the Riegel exponent for the marathon as compared to shorter distances. Indeed, as you noticed, I myself have had problems in many a marathon, so it is certainly a good idea to distuingish between the shorter distances and the marathon.

      Kind regards and best of luck with your running!

      Hans van Dijk

  • Zak Aston says:

    Hi there.

    Your online calculator has been great for finding out what goals I should be aiming for, now that I’ve finally started to run consistently.


  • Michael Lewandowski says:

    Running in a group will help to spend less power against the wind. But what, if the group runs a little bit faster than my race-pace? How many seconds faster should I run to stay in the group? And at which difference should I leave it to keep my pace on my own?

    • Hi Michael,
      This is a very tricky aspect. The advantage of less power depends on your speed, but is only 1 or 2 % for recreational runners.
      Of course you need to avoid going too fast as this will create big problems. In my experience you need to be cautious. On a lucky day you will find a group running at your pace, so you can follow them while saving energy and power. At the end of the race this will pay off and you mau achieve a PB.
      Best of luck,
      Hans van Dijk

  • Thomas says:

    Dear Hans and Ron,

    Your book and information at this website are THE reference for running with power besides Jim Vance’s book. I really appreciate that finally the ‘full physics’ of running are explained in a comprehensible way in one reference.

    However, I wonder about your assumptions in the online calculator regarding FTP and VO2max, understanding that you derive VO2max from FTP by the factor of 0.072, here.

    I end up with a calculated FTP that agrees well with my STRYD CP, but VO2max (using the aforementioned factor) does not agree with that calculated by the Jack Daniels formula, whereby the latter actually provides me with very good estimates for my duration-dependent race paces.

    Should your calculator not rather use the factor 0.081 to derive VO2max from FTP (as stated in Chapter 75 of your book) instead of 0,072? At least for me, this provides a VO2max estimate which is closer to that calculated by the Jack Daniels formula (-> realistic race time estimates).

    Thanks your comments and kind regards,


    • Thomas says:

      After doing some further testing with the calculator and my own calculations, I understand that you are scaling the factor to relate VO2max to FTP by 10 min to 60 min, even though race times provided by the user are likely always different.

      Hence, you receive different VO2max values for race paces that actually perfectly match the calculations after Peter Riegel. Scaling the factor by race time vs. VO2max time solves this issue from my point of view. Consequently, each realistic pace for a given race distance would provide the same VO2max value.

      I am looking forward to your comments and would be grateful about more details on your calculations.

      Thanks, Thomas

      • Dear Thomas,

        Thank you for your compliments and question.
        I am glad see that you have figured out the answer by yourself already….
        Indeed, we calculate the VO2max from the given race pace for a given distance.
        The relationship between FTP and VO2 max is the factor 0.072.
        This can be easily understood by the different duration times (FTP= 1 hour and VO2 max= 10 minutes, the Riegel ratio is 0.88).
        The relationship between FTP and VO2 is theoretically a factor 0.081 as explained in chapter 75.
        So the VO2 at 1 hour FTP/0.081, consequently the VO2max at the FTP is FTP/(0.081*0.88) which equals FTP/0.072

        I trust this will be clear now?
        Best regards,

        Hans van Dijk

  • Chris says:

    Good afternoon,

    I have purchased a hard copy of the English translation from Amazon and I’m finding it absolutely fascinating, so thank you for taking the approach you have!

    I find it a bit large and heavy to read in bed, so I wondered if you did a reduced price ebook or PDF version for people who have purchased the hardcopy? I could then read it much easier on my iPad.

    Thanks again, fantastic work,

    • Thanks for your compliments, Chris!
      I have submitted your question to our publisher Meyer& Meyer
      Unfortunately, they have informed me that due to legal and other matters, they cannot offer you a reduced price for an ebook.

      Best regards,

      Hans van Dijk

  • Laurent says:

    Hello Hans and Ron,

    Thank you for the book! I’m enjoying reading it. I’m not following a following in Chap 20 where you relate FTP and Vo2max. Vo2max is expressed in ml/kg/min and DeltaG in KJ/L of O2, it seems than when doing the conversion, we should end up with Vo2max in L/kg/Min vs. ml/kg/min? could you help me there? thx L

    • Hi Laurent,
      Thanks for the question and we are happy that you enjoy reading our book.
      The answer is that delta G is indeed expressed in kJ/LO2 which is equivalent to J/mLO2.
      So you can convert the VO2 max in mlO2/kg/min to 0.25*19.55J/kg/min.
      As 1 Watt is 1 J/s, you also have to divide by 60 (seconds in 1 minute) to come to Watt/kg which is the unit of the FTP.
      Best regards,
      Hans van Dijk

  • Ron van Megen Ron van Megen says: