The RW Race-Time Predictor



  • I used a time trial time from today (4.2K) to predict what I should do a half marathon in and it said 1:37:58.  I did a PB a month ago in 1:37:57.  I would say that I have trained well for both distances (lots of short fast runs and lots of long runs) so this was impressively accurate!!

     The good thing is, I feel that I held back in the 4.2K today so it may give me a bit of confidence if I do quicker next time and then attempt another Half Mara....

  • Forgot to update this thread -

    From 13.1 miles time of 1:51:53 (March 16th 2008) - 26.2 miles prediction 3:53:16. I know the predictions only 'work' if you train 'properly' for the event you are doing so I'll do my best to stick with the training and get my 20 miles runs in then update on this thread.

    On 25th May 2008 - my time was 3:59:44. I'm sure I didn't do enough long runs - 15, 19, 18.7 , 21 and 15. I'm sure the extra 6 minutes would have been there if I'd more LSRs in me image

    Maybe this year!

  • The value of the Predictors like RW and MacMillan is when they DON'T they demonstrate the races you should be concentrating on ....every one seems to slavishly assume that we all need to run 10ks...NOT TRUE !
  • Well, it confirms how terribly I did in my first half marathon last fall; I ran 1:32:57 and completely fell apart at 5 miles. 2 weeks later when slightly hung over I ran a 37:20 10k, which works out to a half marathon over 10 minutes faster (and much closer to my original goal time). Based on my 10k pb it comes up with 2:43 which is highly unlikely but at least makes me optimistic of sub 3 hrs this fall.
  • I think it's definitely on the fast side for the girls.  2 formulae (1 for men and 1 for women) might give better results.
  • I reckon that it's on the fast side, but not by a great deal. It gave my predicted marathon time of 2.50 on an actual half marathon time of 1.21. At Taunton I ran the full in 2.55. Maybe I was not as fit as possible, but 5 mis is still a large jump.


  • Pammie*Pammie* ✭✭✭
    My recent 5km time was 24:23 just checked what it says for a half marathon it reckons 1:52:02

    On Sunday i ran a half in 1:53:09 which aint bad as i was not properly trained for it and i walked through the drinks stations

    What is important though is that it is just a pridictor of what you can do and not an actual fact You can do all the training under the sun but the reality is some will be suited to the shorter distances and some to the longer

    What they do tell you is the distances you are most suited to i would say it is fairly accurate for me up to the half marathon distance (maybe 15 miles) but beyond that it is way out although less than 5km is a bit out as well as i don't have the raw speed
  • I'm in a bit of a quandry now....

     Been strictly following a sub-4 hour programme for the last 14 weeks for FLM and so my race day pace was supposed to be 9min miles, to go just under 4.

     But my Reading HM time a few weeks ago was 1:48:41 and the predictor says I should be aiming for roughly 3:46, which would roughly 8:40 pace.....

     I don't know if it's a stupid idea to change my plans with only two weeks to go until London, or to just go for it?

    What do you think?


  • I tried the calculater but oddly I am faster than this. I train 15km-16km, but when I ran my half-marathon I was 10 minutes faster. And this was because I had some problems in the first 5km (I expected those) and on the 19th km (didn't really expect those). On the other hand, I am probably not the typical runner - I start very slow and gain speed, I tend to be approx 25% faster in my last 5km when training and keep speeding up afterwards. For some reasons it takes me 3-6km to get in to running mode and then I start enjoying and wanting to actually run and add speed to my run image

     I suppose the calculator is good for people who run in a standartised manner but doesan't work for everyone ... image


  • I do think that the predictions are too fast.  If I put in my 10k PB it predicts a 1:40:00 half for which my PB is 1.43.  Putting that in to predict my full marathon time it comes out at 3:30 and I'm anticipating about 3:55 in the FLM.
  • Well, when I plugged in my 10K time and predicted my 1/2 marathon time, it was accurate to within 20 seconds! Hopefully the predicted marathon time comes true as well.
  • I find that this predictor is way off.

    My 5k time is 17:57 which predicts a marathon of 2:51:59

    My 10k time is 38:05 which predicts a marathon time around 2:55

    My actual marathon pb is 3:15 - a long way a way from sub 3 - maybe this yearimage

  • It's optimistic in its marathon time from 10 km time, and its half time is still a bit optimistic. But also my WAVA grades for 10 k, half and full are 67%, 65% and 60%, so maybe it is me that's the problem. It would be interestesting to see a correlation calculated between the RW prediction and the WAVA 2006 tables used as a prediction.
  • I think a 38.05 10k predicting a 2.55 marathon is pretty sound. However, it only works if you are doing marathon training as opposed to 10k training.

    Certainly as you go up distances the predictor has more margin for error, however, what it is suggesting is that if you can run X for one distance you can do X for another distance - IF you do the appropriate training. Some people on this thread seem to have forgotten that bit!

    To be honest, the reason why most people's predictions are off on the longer distances is because they don't do enough miles - imho.
  • I think that what you say is true Bazza. I should put the 3:15 marathon pb in context. It was about 10 weeks after an ironman triathlon, so while there was a lot of endurance training completed, most of it was for cycling rather than running, also would have been tired after the IM tri. All my pb's for shorter run distances (5k, 5mile and 10k) have come since January, and I have been training more for running. I will start an 18 week marathon programme in June, with a view to having a good tilt at the Dublin Marathon in October. Maybe the sub 3 hours will come then. I'll keep ye posted!
  • You could be right in the general case (who can say) but definitely not me specifically, Bazza - I've just done back to back 16-week 3:45 mara programmes and run a 10k, half and full so far this year without any specific shorter distance training (and all three were pbs with at least the mileage required by RW's own mara programmes - about 1000 km so far this year) so my response is as relevant as anyone's could be to the original question.
  • I have found this calculator amazingly accurate.According to my 10k time acouple of weeks ago I should of ran 34.07 at Fridays 5 mile race.I ran 34.06.Using my 10k time I am about 10 seconds off a pb for a half marathon.10 mile spot on.Marathon time is predicted as 3.17 I ran 3.32 3 years ago.So with proper training theres hope I can still break 3.30.
  • I think the prediction is sound as long as you train properly for each event.In laymans terms if you training for a 10k then run a marathon you are proberly going to fall short.

    Mind you I am sure I am stating the obvious and you all know this already.

  • Stevie  GStevie G ✭✭✭

    this year I must say my timings have all pretty much fallen into line.

    Although I've just done a 17:23 5k, which is hyping the pressure on to bring my 10k 37:27 down to a possible 36:10  cripes!!

     37:27 wasn't an easy course tho...but then the 5k course wasn't an easy course either!!

  • BirchBirch ✭✭✭

    Just spotted this - very interesting.

    The predictor for my 10K best of 34:58 indicates a 2:40:41 marathon, my half mara best of 1:15:58 indicates 2:38:23 for marathon. My actual best is 2:46:00, suggesting I underachieved at the full distance.  I must say I have always trained appropriately, so perhaps I just don't get the distance, so to speak, or have been too cautious. This is now academic, as the aforementioned half was in 1990, the marathon in 1991, and the 10K in 1994 (age 39).  

    Fascinating, though.

  • PhilPubPhilPub ✭✭✭
    Ah,  big thread so I don't know if this has been covered already but in my experience the RW predictor is a little optimistic when predicting longer race times from shorter, but the McMillan predictor works better.
  • Stevie  GStevie G ✭✭✭

    whoops...didn't realise the rw one was different to the macmillan, as my timings/predictions above are from that...

    am currently in the great position fitness going well/non injuries of knowing my 5mile pb will be shattered next race out and that the next flat 10k should mean a new pb too.

    this is the good time of running!

  • Checked out both the RW calculator & the McMillan one, and the McMillan calculator is staggeringly accurate, all of my distances from 15k downwards are to within 3 or 4 seconds, my 10 mile & half marathon times are out by a couple of minutes - maybe down to a lack of endurance on my part.
  • I have tried the calculator based on a couple of recent race times and it always predicts I will finish faster than I do. Seems like one of us is out somewhere.
  • I just tried using the calculator to predict my time for the forthcoming marathon in Berlin, which will be my 6th marathon.

    Funny thing is, whether I use my PB on 10K or half marathon (30:59 and 1:08:22 respectively), I get the EXACT same marathon time! (2:22:32)

    So, I don't know yet how accurate the calculator is in my case, but it certainly seems consistent with my different PB's image

  • I have only just discovered this!  A lot to read through but it looks as though it gets a lot of positive comments.  I hope it is accurate as I am doing Amsterdam marathon and my predicted time is 3.45 based on my PB half marathon earlier this year.  As I am aiming for and have trained for a sub 4 I just hope I can do it on the day!  Maybe the power of positive thinking will help too!
  • I can run a mile in a little under 5 minutes... so it says I could potentially with training do a marathon in the 2:30s region. I swear one day I will do this and better it!... I believe now - thanks probably innacuarte calculator, you always believed in me.

  • I ran my most recent 10k in 48:20 plugged in predicted time for half marathon actual time was 1:44:24.   
  • I recall when this subject last came up 3 or so years ago the discovery was that the faster you were the more accurate it was likely to be.

  • I guess I better get faster!!!
Sign In or Register to comment.