Did training predict actual race performance?

General discussion on Training. How to get better on your erg, how to use your erg to get better at another sport, or anything else about improving your abilities.
Post Reply
Cyclingman1
10k Poster
Posts: 1554
Joined: February 7th, 2012, 6:23 pm
Location: Gainesville, Ga

Did training predict actual race performance?

Post by Cyclingman1 » February 27th, 2019, 12:36 pm

Since this is the training forum and the WRIC just completed, as well as other venues, it would be interesting to see commentary by those who competed about whether their training produced expected results. For example, I've seen a lot of statements that 8 or 10x500m predicts a 2K time. Did it? Or 4 x 1K. Just wondering.
JimG, Gainesville, Ga, 73,189lb,76”.

KeithT
5k Poster
Posts: 553
Joined: February 5th, 2018, 12:41 pm

Re: Did training predict actual race performance?

Post by KeithT » February 27th, 2019, 5:04 pm

Cyclingman1 wrote:
February 27th, 2019, 12:36 pm
Since this is the training forum and the WRIC just completed, as well as other venues, it would be interesting to see commentary by those who competed about whether their training produced expected results. For example, I've seen a lot of statements that 8 or 10x500m predicts a 2K time. Did it? Or 4 x 1K. Just wondering.
Good question - here is what I found:

I did quite a few different interval sessions and some a few times, in the end I think they actually were very predictive but none were right on, instead I found the 500 interval sessions or 750 to be a split or 2 under 2K split and the 1000m about 1 split above as predicted. For example, my time was 6:29.9 - 1:37.5 split at WRIC and here are some of the times I pulled for interval sessions in the 8 weeks prior:

6x500 with 1 minute rest - avg split was 1:35.5
8x500 with 3.5 minute rest - avg split was 1:35.7
5x750 with 3 minute rest - avg split was 1:36.6
4x1000 with 5 minute rest - avg split was 1:38.2

So again, they were pretty good predictors and went about how I heard they would for times - I am a bit better at shorter distances so that allowed me to do the shorter intervals better as well. As much as I dreaded and struggled with the 4x1000, I think it really gives you a good idea at +1.
51 yo, 6'3" 207#
PBs (all since turning 50):
1 min - 373m, 500m - 1:23.2, 1K - 2:59.8, 4 min - 1265m, 2K - 6:29.9, 5K - 17:27, 30 min - 8277m, 10K - 36:30, 60 min - 16036, HM - 1:20:22

H2O
2k Poster
Posts: 297
Joined: March 18th, 2006, 9:51 am
Location: Frankfurt, GER
Contact:

Re: Did training predict actual race performance?

Post by H2O » February 27th, 2019, 7:07 pm

My recent experience:
30 mins:
race = last training (1Month before, exact to the meter -- a fluke)
but much worse than 6K pace a week before (lost 2.5 secs / 500)
1K:
race faster than all training paces by 1 sec/500.

Form has been a little unstable for 6 weeks before the races.

User avatar
NavigationHazard
10k Poster
Posts: 1712
Joined: March 16th, 2006, 1:11 pm
Location: Wroclaw, Poland

Re: Did training predict actual race performance?

Post by NavigationHazard » February 28th, 2019, 11:39 am

Cheers, I doubt that there is anyone on the planet who has done more interval workouts than me in the last 20 years plus. And in my experience, the only 100% accurate 2k predictor that's available is a 2k race or time trial.

IMO, workouts that commonly are presented as 2k predictors (e.g. 6 x 500 on 1'; 8 x 500m on 3:30; 4 x 1k on 3'; etc. etc. etc., arguably at various relations to 2k pace) really ought to be thought of in terms of 2k preparation rather than prediction. I know I'm as guilty as the next person of misrepresenting them in blog and Forum posts. But I would suggest that their chief value lies in providing you with a clue as to what an actual 2k will feel like along the way plus a certain diagnostic value regarding your capabilities. In general, the shorter the rest in these sorts of workouts, the more the aerobic side of your preparatory training will reveal itself. The longer the rest, the faster you'll be able to go and the more the anaerobic side of your training will come to the fore. Over time, depending on factors including individual physiology, training load and frequency, hydration, sleep and nutrition, stress and mental focus, and (not least) rating and pacing on the intervals vis-a-vis how you'd approach an actual 2k, you can acquire a pretty good sense of how a putative "predictor" workout is likely to correlate to a race or trial result. But even at the individual level, it's always possible to surprise yourself both positively and negatively. Attempts to generalize the "predictive" value to larger populations invariably run into the "mileage may vary" problem.

From experience in coaching and in analyzing thousands of interval workouts over the years, I have come to understand that

1) you need to hold as many of the confounding factors I reference above equal as possible to maximize the diagnostic aspects and any "predictive" value that may exist; and

2) it's awfully hard to tell much about a 2k race or trial that's meaningful from any single workout beforehand. Venue conditions can be anticipated but can't be known perfectly in advance. Neither can what the other competitors in a race might be doing. As for time trials, yes they can be controlled a lot more than can venue races. But even there, there's almost always at least some element of variability to contend with.

The bigger the sample size for the "predictor" workout(s), in other words, the better. This is true whether you're repeating the same workout or whether you're essaying a number of different possibilities. And if you look at enough of them, you often start to see a kind of bell-curve distribution relative to average 2k pacing. Some of them you'll discover you're doing faster, some slower, and -- especially when the confounding factors are held relatively constant -- rather more of them will be in the middle....
Last edited by NavigationHazard on February 28th, 2019, 12:53 pm, edited 1 time in total.
63 MH 6' 6"

User avatar
hjs
Marathon Poster
Posts: 8533
Joined: March 16th, 2006, 3:18 pm
Location: Amstelveen the netherlands
Contact:

Re: Did training predict actual race performance?

Post by hjs » February 28th, 2019, 12:42 pm

Not raced in two years, but I "always" know in Advance what time I will pull, within a few seconds.

For me 4x 1k on 4.20 rest, is roughly 0.5 to 1 second above my race result.

8x500 on 1 rest is roughly half a second below my race result.

5k also tells a good bit. Thats 6/7 seconds above 2k for me.


A few but 's, my race results have varied Little over a number of years. Roughly 10 seconds between fastest and slowest.
I am steady in my approach and raceplan.

Over a timespan of 10 ish years I did around 30 races, all early morning, which I hate! From those 30 twice went wrong, once I wanted a result, and halfway I knew I would not get it and beng I stopped and walked away. :oops:

Second time, After a false start, the restart was super quick, missed it, had to pick up the handle, and pulled too fast for too long, saw the numbers, but it felt ok, although I ofcourse knew that was bull... Really really hurted from halfway on. Placed me 4th, which should have been a first, this at a euro open.

For the rest, roughly half where absolute 100% efforts. The others, always close.

My last racing season, I did three races, in a timespan of 10 weeks, and got three results with just 0.1 second difference. Thats twice exactly the same result plus one 0.1 slower.....

Other point, over the years I got a bit stronger, different training, and pulled faster 500/1k results, but those did not gave me better 2k s. The longer work was a bit more varied, but on average a bit slower.
For my training see twitter @Hjsrowing

Cyclingman1
10k Poster
Posts: 1554
Joined: February 7th, 2012, 6:23 pm
Location: Gainesville, Ga

Re: Did training predict actual race performance?

Post by Cyclingman1 » February 28th, 2019, 1:41 pm

Interesting feedback.

Henry, I think you are probably unusual in predicting so closely to race results and having very similar results. You have a lot more experience than many for sure. I can see that your 8x500, R1 would get you very close to 2K pace.

Jonathan, I find myself in nearly total agreement with your lengthy response. I've said elsewhere that prediction workouts are really more targeted training than predictions, unless one is highly experienced in doing so. As you say, there are so many factors that go into any rowing result, whether training or a race. Easy to take a result on a day when one feels great and everything is aligned and take that as the standard - might never happen again. Also, intervals with a lot of rest at high SPM are questionable as far as predicting 2K, though they may be good training. No rest in a 2K. The difficulty of taking training to a strange venue miles away can hardly be minimized. Olympic athletes get time to acclimate, but weekend warriors don't get that luxury.

Hopefully, more will respond with what they predicted and what they did. It's a hard thing to do.
JimG, Gainesville, Ga, 73,189lb,76”.

User avatar
hjs
Marathon Poster
Posts: 8533
Joined: March 16th, 2006, 3:18 pm
Location: Amstelveen the netherlands
Contact:

Re: Did training predict actual race performance?

Post by hjs » February 28th, 2019, 1:51 pm

Agree Jim,

But I do think its also a matter of doing things right, or at least the same. If I do a race and do intervals to prepare for that, my mindset is almost as focussed and like doing a race.

Doing other intervals, rate restricted, using other rests all muddy the waters. Simply biting the bullet and doing a serious 2k every now and than seems like a good approach.
For my training see twitter @Hjsrowing

ccooper
Paddler
Posts: 37
Joined: August 6th, 2010, 2:55 pm

Re: Did training predict actual race performance?

Post by ccooper » February 28th, 2019, 2:26 pm

I find that the Pete Plan intervals are reasonable predictors for me = 8x500/3:30r and 4x1000/5:00r.

But more importantly, the trend over time is accurate. If you do these interval workouts over the course of a season you will hopefully see your times improve. Improve by 1 second and it usually translates into 1 second improvement in pace for 2k. Match that up with occasional actual 2k's and you will be able to project your current 2k pace at any time during the season.
--
Chris Cooper M65-69 Hwt 2km world record 6:39.5 in 2017

KeithT
5k Poster
Posts: 553
Joined: February 5th, 2018, 12:41 pm

Re: Did training predict actual race performance?

Post by KeithT » February 28th, 2019, 2:44 pm

ccooper wrote:
February 28th, 2019, 2:26 pm
I find that the Pete Plan intervals are reasonable predictors for me = 8x500/3:30r and 4x1000/5:00r.

But more importantly, the trend over time is accurate. If you do these interval workouts over the course of a season you will hopefully see your times improve. Improve by 1 second and it usually translates into 1 second improvement in pace for 2k. Match that up with occasional actual 2k's and you will be able to project your current 2k pace at any time during the season.
I think this is key - I think the workouts can be good predictors if you are already basing it on a known 2K time and doing them as you would a 2K piece. I found as the workouts improved the 2K time would improve. Now, if you took someone who never tested a 2K and had them do a workout it might not be too accurate but I think you can definitely have a good idea of what kind of 2K pace you can keep by trying the workouts IF you have some experience to how they correlate.
51 yo, 6'3" 207#
PBs (all since turning 50):
1 min - 373m, 500m - 1:23.2, 1K - 2:59.8, 4 min - 1265m, 2K - 6:29.9, 5K - 17:27, 30 min - 8277m, 10K - 36:30, 60 min - 16036, HM - 1:20:22

RayOfSunshine
2k Poster
Posts: 424
Joined: December 15th, 2017, 9:45 am

Re: Did training predict actual race performance?

Post by RayOfSunshine » February 28th, 2019, 2:49 pm

hjs wrote:
February 28th, 2019, 1:51 pm
Simply biting the bullet and doing a serious 2k every now and than seems like a good approach.
I did this and ended up 0.6 slower in the race. So, it's a pretty good marker :lol:

Seriously though... someone, probably Henry, said the best way to know is to do one.
48M, 175cm/5'9", 74.8kg/165lbs
Neptune Beach, FL/Boone, NC

Cyclingman1
10k Poster
Posts: 1554
Joined: February 7th, 2012, 6:23 pm
Location: Gainesville, Ga

Re: Did training predict actual race performance?

Post by Cyclingman1 » February 28th, 2019, 5:32 pm

NavigationHazard wrote:
February 28th, 2019, 11:39 am
The bigger the sample size for the "predictor" workout(s), in other words, the better. This is true whether you're repeating the same workout or whether you're essaying a number of different possibilities. And if you look at enough of them, you often start to see a kind of bell-curve distribution relative to average 2k pacing.
hjs wrote:
February 28th, 2019, 1:51 pm
But I do think its also a matter of doing things right, or at least the same. If I do a race and do intervals to prepare for that, my mindset is almost as focussed and like doing a race.
Doing other intervals, rate restricted, using other rests all muddy the waters.
ccooper wrote:
February 28th, 2019, 2:26 pm
But more importantly, the trend over time is accurate. If you do these interval workouts over the course of a season you will hopefully see your times improve. Improve by 1 second and it usually translates into 1 second improvement in pace for 2k.
Lot of agreement that consistency and enough data points are the key to predicting 2K. Of course, coupled with actual 2K times, either time trials or races.
JimG, Gainesville, Ga, 73,189lb,76”.

Post Reply