Specific 2k Pacing (Per 500m)

[Yankeerunner]

Mike--

It strikes me that you are questioning an ingrained cultural bias--almost a matter of faith--and facts are often of limited influence in such instances.

Tom

Good point.

In my running days I became a practitioner of even pacing or negative splitting, but it wasn't easy. Many times (I shudder to think how many times it took to get through my thick skull) the sound of a starting gun would infuse me with so much adrenaline that I'd shoot out fast, feeling incredibly easy doing it, out with the leaders, convinced that this was my breakthrough day when all the hard training would finally place me at my rightful position in the front of the pack. Then without fail there would come a point, as little as 5% into the race, that like a tire with a slow leak I would go flat and stagger through the rest of the race.

Finally I decided that at the start of races I would not run the first mile any faster than the other runners in the race who normally beat me. I immediately began finishing higher. At races where splits were available it turned out that I ran a relatively even pace, but in big races could go from something like 100th place at the mile to somewhere in the teens at the finish. I seriously doubt that starting faster would have placed me higher at the finish, which is the one split that really counts.

Rick

[mpukita]

After reading this thread for the first time after being pointed here by Mike from the UK forum I have two comments:

1) For the first time ever in a venue race I was able to hold a relatively steady pace through +1700 and kick. Prior to this, my race plan was too aggressive. I would always die in the final 500M. I feel I was able to do this because I now have a coach (although one could do this by themselves, with some discipline) who has me on a structured program ... part of which is test pieces - mile, 1500, 1800, etc. Over time, he sees what I can do and maps it against hundreds of Suunto files from other athletes. His predictions of what I should be able to do are now very, very close to what I do. As well, I see and feel the improvements, see what happens when I push a workout and "blow up" on the final piece, etc. I now have a better feel for what is probable, what is possible, and what is impossible. This gave me great confidence in my last race. I knew I could unquestionably hold our race plan pace for the entire 2000M ... so much so that at 500 I was saying to myself, "Should I pick it up now?" And then said the same thing at 400 to go, 300 to go, and finally went at 250-200 to go. 1:48.7, 1:49.1, 1:48.7, 1:46.7 = PB by 5.9 seconds. I had never done anything close to this, from my memory, as, unfortunately, none of the other venue races I attended (all here in the US and Canada) provided stroke data. It seems to me the avrage pace was always slipping up from 500M in.

2) This took more confidence in my fitness and training than any race prior. Having it, I was able to execute almost exactly to plan. I see a negative split approach, as Mike advises, as the "next step up" in a racer's confidence. While we all see the data, it takes a great deal of confidence to say to one's self, "Self, I am going to set a PB today. And, I'm going to do it by negative splitting the piece. How do you like that?"

Of course, Mike encourages negative splitting workouts in the WP, a classic "train as you race" strategy. Still, it seems to me that one would need a tremendous amount of confidence in one's training as well as strong mental discipline to step out on the ledge and try this the first time.

Has anyone that's not of an elite rowing status tried this in a venue race? What was your experience?

[Mike Caviston]

The peak of the indoor racing season approaches, and inevitably people will be posting questions along the lines of, “How should I pace myself when racing for 2K?” Perhaps some of those people will find this thread. Some of them may even learn something from the information provided here. A while ago I finished analyzing data from the past three CRASH-Bs for open lightweight women & men, but I haven’t gotten around to putting the results here till now. The data include times from 266 women and 406 men. As recorded previously for openweight athletes, the tables & graphs below divide the field into 20% increments and in 500m increments show average times/paces, Watts, deviation (W per 500m), and % of final power (W). The scatterplots compare final 2K Watts to the 1st 500m (as a % of final).

[Mike Caviston]

The following tables & graphs compare three different racing strategies (explained/defined previously): Negative Split, French (classic OTW fast-cruise-fast approach), and Positive Split).







The following histograms compare 1st 500m to final Watts. Each bar represents a 2.5% increment of 1st 500m as a % of final Watts. Values on the left represent the most conservative starts and values on the right represent the most aggressive (fly-and-die) starts. Numbers at the top of each bar are the average Watts for each increment; numbers at the bottom in parentheses are the number of athletes in each increment.




To summarize some of the main points supported by the data:
• The fastest 2Ks occur when the opening 500m is a little slower than the final average.
• There is a solid relationship between the pacing of the first 500m and the final time. Starting at 120% is clearly a recipe for disaster, but generally speaking starting at 101% is worse than starting at 100%; 102% is worse than 101%; and so on, with results getting worse and worse as the opening 500m gets faster and faster.
• Classic OTW race strategy is less effective than the Negative Split strategy. If this is true on the ergometer, it follows this is true OTW as well.

Enjoy.

Mike Caviston

[Bob S.]

To summarize some of the main points supported by the data:
• The fastest 2Ks occur when the opening 500m is a little slower than the final average.
• There is a solid relationship between the pacing of the first 500m and the final time. Starting at 120% is clearly a recipe for disaster, but generally speaking starting at 101% is worse than starting at 100%; 102% is worse than 101%; and so on, with results getting worse and worse as the opening 500m gets faster and faster.
• Classic OTW race strategy is less effective than the Negative Split strategy. If this is true on the ergometer, it follows this is true OTW as well.
Mike Caviston

Mike,

I appreciate your advice and have found that slight negative splitting has served me very well for 2ks and other pieces as well. The main problem I have is knowing what would be a suitable first split pace. All of my training is down at 4000ft of altitude, where my absolute best 2k time is 8:17.1. At sea level, I have had a couple of sub 8 2ks and 3 in the range 8:03-8:06. I don't know whether to try to start at a 1:59 pace or to take it a bit easier, like 2:01. I can't test it here. I tried a 2k a month ago with a 2:04 start but quickly blew up and ended up with a 2:09.8 average pace. Last week, doing 4x2k intervals with 4'r, I ended up with that same average. A couple of days later I did a 10k at a pace that was less than 4 seconds slower (i.e. 2:13.6)

What would you suggest for a 2k starting pace at sea level?

regards,

Bob S.

[robhen]

Mike,

I believe in your analysis and strategy - I followed it in my last indoor race and got a PB!

Does it matter if races are 1000m that are commonly held for masters? Or do you have an opinion for 1000m races?

I race 1000m in masters races and I am having trouble convincing my partner who is the stroke in my double scull to drop the effort down 1 or 2% so we have enough in reserve to sprint faster at the end. In our last raced I was stuffed in the last couple of hundred, as a result of going out too hard, and I am sure this cost us a major title.

[Mike Caviston]

The main problem I have is knowing what would be a suitable first split pace... What would you suggest for a 2k starting pace at sea level?

Bob, that is the $64,000 question. If you should start your 2K a little slower than your average pace, how do you know what your average pace will be? That can/should be the topic of another thread, and I’ve dealt with it pretty extensively on the old WP discussion thread. (The basic principle involves selecting a race pace based on recent workouts – the best 2K predictor for me being an all-out 4 x 1K.) For an experienced athlete such predictions can be remarkably accurate, but adding the variable of altitude to the mix makes prediction much more challenging. And I don’t have enough personal experience with altitude to give you any better advice than you can come up with on your own. I know this has been discussed frequently on the forum as well, but my short visit with Dennis H. in CO a couple years ago seemed to suggest about a 12-sec difference between 2K at altitude and sea level.

What I can’t stress enough is that whenever in doubt, it pays to be conservative. If you start too slowly, it’s true you lose time that can never be made up. But if you start too quickly, the end result is about three times worse. In other words, starting one second too slow costs you a second, but starting one second too fast costs you three seconds.

Does it matter if races are 1000m that are commonly held for masters? Or do you have an opinion for 1000m races?

Most definitely. Optimal racing for 1K requires good pacing (i.e., even/negative splitting, NOT an all-out start). See the post at the top of the previous page.

Mike Caviston

[robhen]

Mike,

Thanks for your response.

Something lit up like a 1000watt bulb. When I raced in my younger days I used to go out really hard, be on the pace and have absolutely nothing left for last couple of hundred meters and get thrashed. A saying, "putting a wise head on young shoulders" springs to mind.

Im on a downer now as I could have got more out of my career.

Mike, dont take this the wrong way - have you ever thought of using your brain power to analyse the stockmarket? Instead of crunching the numbers on thousands of races you could of picked a basket of good shares to put in a portfolio.

[Nosmo]

....All of my training is down at 4000ft of altitude, where my absolute best 2k time is 8:17.1. At sea level, I have had a couple of sub 8 2ks and 3 in the range 8:03-8:06. I don't know whether to try to start at a 1:59 pace or to take it a bit easier, like 2:01. I can't test it here.....

Bob S.

Bob,
IF this was me I would try to make as good a test as I could at sea level before the actual race. If I was getting to sea level the day before I would try an all out 2K (or close to it) if possible. If I was getting to sea level the same day as the race or could not erg at sea level before hand, I would focus my warmup on testing the pace. Do some tests of 750 to 1250 meters at an effort that feels like one or two second pace slower then you max effort. You may be more tired then you want to be, but having an accurate idea of what your pace is so much more important. Also realize that when doing heats or multiple races, the second or even third race is often the fastest.

Good luck and let us know how it goes.

Nosmo

[RowingScience]

What a very interesting and thorough brief on pacing. It must have been a facinating project to undertake. Thanks Mike for sharing it.

One thing on the physiology side I'll add in. The initial energy system used in the start is the ATP-CP system. The breakdown of the CP to replenish ATP lasts only for a few seconds BUT the phosphate remaining is able to act as a buffer for the lactic acid later produced. The effect doesn't last long, but an initial hard start of a few seconds (I coach 5 strokes all out) actually HELPS you because of this. Much longer though and you'll pay. Physiological research on pacing suggests even splits are optimal - that of course is quite aside from your psycological arguments that I'm sure are important.

Alan Thomas
Rowing Science Weblog
http://daily-ergworkout.blogspot.com

[Mike Caviston]

I am gradually recording scores from the 2007 CRASH-B for Open & LW men & women, which I will fold into the data from 04-06 to provide a more complete picture of optimal pacing for championship indoor rowing. But as that may take a little time yet, I thought I’d do a quick update of another analysis I did previously on this thread. Most of the data I’ve presented over the past several months is cross-sectional in nature – that is, I’ve compared averages of sample populations (“winners” and “losers”). This works with large samples that can withstand the statistical impact of a few outliers. But it is even more instructive to compare single individuals or crews under different conditions (fast start vs. slow start). Laboratory research that has attempted to compare pacing strategies under controlled conditions generally shows that even- or negative-pacing is a more effective strategy than positive-pacing. The drawback of such studies is that they can’t accurately reproduce true race conditions (preparation, motivation, etc.) OTOH, it is hard to analyze the same athlete in different races and control for training, motivation, injuries, etc. Still, with that limitation in mind, if we analyze elite athletes in championship contests we can reasonably assume a certain amount of consistency in factors (besides pacing) that affect performance. Earlier on this thread I looked at different performances by Xeno Muller (Olympic/WC 1x), the University of California varsity women’s eight (NCAA), and Graham Benton (British & World indoor rowing championships). Graham has recently reported a Personal Best 2K of 5:43.7 (details on UK forum), so I decided to re-examine his collective championship performances in comparison to his recent test.

Below is a table summarizing seven of his 2Ks, including the recent PB.


This graph compares the average pacing of the first six 2Ks with the pacing of the PB. The average opening 500m of the first six races is 106.3% of final; the opening 500m of the PB is considerably slower: 102.8%.


This graph plots final 2K Watts against opening 500m (as % of final). The general trend shows that the faster 2Ks begin with a more modest opening 500m.


This is of course consistent with the data gathered on hundreds of other indoor races. The data has shown (and will continue to show, as I update it) that the optimal strategy for 2K performance involves an opening 500m that is less than 100% of final power (in other words, a negative split). Graham has long been known for his fast starts, though his best performance to date has come with a (relatively) more conservative strategy. I believe that if he hopes to take his 2K to sub-5:40 territory, his best chance will come by tweaking his race plan even more. I certainly wish him the best of luck in future races, whatever strategy he uses.

Mike Caviston

[bscastro]

I'm new on the forums, but I wanted to congratulate you on a great time - especially for your size! As a smallish type guy myself I am very impressed.

Congrats!

Bryan

[Mike Caviston]

The women’s NCAA rowing championships were held a couple weeks ago, and since JAMCO recorded the racing splits I’ve analyzed the 1V Petite Finals as well as the 2V and 1V Grand Finals. The results, as usual, illustrate some pertinent concepts regarding effective pacing. In the Petites, Minnesota provided another example of coming from way back to win a race. I want to stress again that being behind early is not in itself the objective of good strategy. I would rather lead from the start and steadily increase the margin until the race is over. But when racing crews of equal or better ability than yours, it is essential to stay within yourself and follow your own optimal strategy, which often results in being behind early. Don’t panic, stay focused on your own performance, and trust that the other crews will fade as the race progresses. If they don’t, they were so much faster than you that nothing you could do would make the difference. But good strategy can put you into position to capitalize on other crews’ mistakes, and even if you don’t come in first you’ll be a lot closer to the winner. When analyzing a race, remember the key is not to look at the absolute time at the start, but at the start as a percentage of final speed. See how much more conservative Minnesota started compared to Dartmouth (the early leader) and the other crews.

[Mike Caviston]

The 2007 NCAA championships were interesting in that overall team champion Brown didn’t win a single race but placed well enough in all events to gather the most team points. While Yale ran away with the Varsity 8 championship, the battle for the team title was between Virginia and Brown. Virginia won the Varsity 4, and merely had to finish ahead of Brown in the 1V, or finish one place behind Brown in the 1V but two places ahead in the 2V. But Virginia finished behind Brown in both the 2V and 1V, and it is instructive to look at how each team approached those races. Virginia established the early lead, “theoretically” putting them into position to control the race and stay ahead of Brown, but that did not happen. Meanwhile, the 2V winner (Minnesota) and the 1V winner (Yale) each led from wire to wire. Does that suggest the best strategy is to grab an early lead and hang on? No. Neither Minnesota nor Yale was “hanging on” at any point during their races. They were clearly the dominant crews, and they had enough ability to lead early without adversely affecting their ability to finish well. Their relative starts are not significantly faster than the other crews. Note also Ohio State’s move from 6th to 2nd in the Varsity 8 final.

[Mike Caviston]

I was reading an account on Row2K of one of the races at Munich:

…the buzz in the grandstands irrespective of nationality or allegiance was unquestionably focused on the first of the men's quad, which came down to a photo finish. There was a heap at stake in the race; only the top 11 crews from this World Championships get an automatic berth in the Olympics, so a failure to make the A-B semi squashed any hope for earning one of these spots. The United States came through from a somewhat distant fourth to sprint into second two strokes before the line to take the second of only two semifinal qualifying spots, 6/10 of a second behind UKR, and 4/10 ahead of Canada. Their final 500 was a full 1.5 seconds faster than any else in the rep.

"We were in fourth about a length down, and I could hear them, but I couldn't see them," said US men's stroke Mark Flickinger after the race. "So we started rowing higher at around 700 to go, and just kept going up and up. We had nothing to lose at that point, and at the 500 we were at 38, but we still stuck to our plan in the final 500 and I think that made us able to go faster as the rating came up. I was pretty proud of our crew that we could stick to the plan, and we were able to get through.

Looking ahead to the rest of the week, Flick might move some of the crew's speed around a little bit: "I'd be happy if we were a little faster in the first 1000," he said with a quick laugh; "but for today I'll take it."

The full story can be found here:
http://www.row2k.com/worlds/features.cfm?ID=1594

When all the finals have been run, I’ll analyze the results as I have in the past. But this race caught my interest for a reason. Accounts such as this, where a crew comes from behind to win or qualify (I’ve read many such accounts during the college racing season) tend to make it sound as if the crew were at a disadvantage by being behind early. They were in a huge hole and needed a miraculous effort to come from behind. In these situations, athletes and coaches assume that if they had gone faster in the first half of the race they would still have the same speed in the second half, but that is highly unlikely. I suggest that if the American quad had started much faster in this race, the Canadians would be the ones advancing to the finals. Note the fate of the fast-starting Croatians.