The 126th edition of the Boston Marathon saw 40,000 runners depart Hopkington to make their way back to Boston, with Supersapiens Director of Applied Science and Content, David Lipman, being one of them.

David stuck to his plan, seeing the benefit of early discipline when it came to the infamous hilly section through Newton including “Heartbreak Hill” and ran strong from here to the finish, setting a new personal best of 2:48:14 bettering his previous best of 2:49:26 from October’s Rotterdam Marathon.

For the unfamiliar, despite Boston being a net downhill course, it is generally quite rare for people to run personal bests on the course. The mix of crowds, downhill and the hilly section often mean that times are usually slower than personal bests, making David’s achievement even more impressive.

Let’s dig into how David has managed to run a personal best by around 10 minutes in October, followed by another 70 seconds in April on a much more difficult course.

For some context, David is not new to running, having run track in high school (though self admittedly neither too great nor close to his potential), followed by years of very sporadic and causal running, and the running more seriously starting in 2012. He completed his first half marathon in 2013 (1:25) and first marathon in 2015 (3:21).

So What Does David Do In Preparation?

David is self coached and of the belief that consistency is the key to success, as a result maintains a good running volume year round, without significant low volume or high volume periods. This is mirrored in other aspects such as dietary periodization and body composition. He is heavily influenced by Renato Canova’s philosophy (often characterized by long periods of running at marathon pace with short periods of slightly slower running between them) in his training.

David utilized this approach successfully for his recent marathon preparations, allowing him to better understand the fatigue of the race, dial in his pacing and crucially allowing him to both test race day nutrition, compare nutrition products and train his gut.

Using Supersapiens to Gain Insights

Comparing Nutrition Products

David has been using Supersapiens since January of 2021, before joining the company in April 2021.

One of the key insights he initially hoped to gain was the comparison of different sports nutrition products. Some of the initial testing included comparing Vitargo to Cyclic cluster dextrin (CCD) for their impact on his feeling, performance and glucose (predominantly focused on circulating availability and stability).

Training predominantly with CCD, David was aware he would need to be able to utilize gel products for racing given the logistical issues of carrying enough fluids etc for the CCD to be a viable option. As a result he started utilizing Maurten gel products in training, ensuring he could tolerate them and felt like he performed using them (including caffeinated varieties).

Training The Gut

David used a gut training protocol leading into the Rotterdam marathon, concerned about the potential of a hot race day.

Increased heat stress creates an increase in gut stress as blood flow is diverted away from organs, to the skin to aid in cooling and thus meaning there is less blood flow to aid in digestion.

During the warmer months he spent some time at race intensities on warmer days with an intake of 105g/hr of carbohydrates as well as with an increased fluid intake with respect to what he would normally use during a marathon.

Whilst unpleasant, this aided greatly in David’s confidence that his race day nutrition strategy would work regardless of climatic conditions, not to mention aiding him to train his gut to absorb more carbohydrates.

Using Supersapiens to Train the Gut
Gut Training pre-run

Race Day Nutrition

Having decided on using Maurten for his race, the next aspect became checking this during extensive race intensities, under climatic (heat) stress, and deciding on the number of gels to use in the race and their spacing.

This was a decision based on feeling, performance, glucose stability during his marathon intensities and logistics.

He settled on 3 gels per hour, yielding 75g of carbohydrate per hour.

Boston Marathon Fueling Plan

Following long term Supersapiens use, and success using an established protocol, David had the confidence that he could use these learnings going into Boston.

In his preparations for Boston he continued to review his fueling strategy, this time, being more specific around using a race day strategy than previous experimentation. This allowed him to ensure his faster training paces and adaptation post Rotterdam had not changed things in terms of nutrition tolerance and requirements for him. Specifically David was evaluating his circulating glucose availability - a combination of glucose stability and glucose level.

If anything between Rotterdam and Boston, David found that for a similar carbohydrate intake, from the same sources, he was experiencing more stable glucose levels during his training.

CGM Glucose Data from the Boston Marathon

There is a long history of glucose monitoring at the Boston Marathon as detailed here and here but this is the first time we have CGM data from the race.

Supersapiens Data from Boston Marathon
Figure 1: David’s glucose, pace, heart rate and the race elevation from the Boston Marathon

Supersapiens Glucose Score from Boston Marathon
Figure 2: David's Glucose Score from the Boston Marathon

From the above, you can see David maintained a very stable glucose throughout the race. This is no doubt part of why he felt as good as he did and could finish as strong as he did. Crucially, David had no lows and a flat slope–amazing glucose stability.

Of interest is that one of David’s drops is quite clearly associated with the downhill preceding the hilly section of the course, where his heart rate lowers too, as he tried to ease off a little bit in preparation for the upcoming hills. This is both a normal and appropriate physiological response.

For those interested in David’s pacing his 5km splits can be seen below, and whilst the pace chart above looks a little variable this is actually a reflection of the nature of running, the Boston marathon course and David pacing himself up hills. When using a 5km frame for evaluating pacing, you can see his pace was very even, no doubt aiding in his glucose stability.




Pace (min/km)





































Table 1: David’s 5km splits for the Boston Marathon.

Wondering why David ate his breakfast four hours before the race? These two articles may help.

Another factor that played a role in David’s performance and glucose stability was his pre-race carbohydrate loading.

Carbohydrate loading protocols

Research on carbohydrate loading is extensive, as the practice has been around for 50 years.
Various protocols have existed, with different durations, periods of depletion and dosing of carbohydrates. In general it is now accepted that about 10g/kg/day is appropriate.
For the well trained runner, this may be achieved by tapering exercise over the final days before the marathon and ensuring carbohydrate intakes of 10-12 g/kg/day over the 36-48 hours prior to the race.

Figure 3: Example glycogen levels for different scenarios (from Rosenbloom)

Of course, carbohydrate loading is about topping up glycogen stores, in both the muscle and the liver. The use of carbohydrate during exercise is with the goal of either sparing or supplementing this glycogen to allow for continued higher intensity exercise. A great visual representation of the fluctuating nature of glycogen levels (with some causes) is shown above in Figure 3. The goal in this case is increased rest and eating to increase glycogen levels.

For more on glycogen metabolism this paper does a great job summarizing things for coaches and athletes.

At 83kg, David’s carbohydrate loading suggestion is ~830g/kg/day for 1-3 days. He decided on increasing his intake for 3 days, with true emphasis on hitting 800g on the day prior to race day.

Carbohydrate Loading For Boston Marathon

April 15th (Race Day -3)
Cinnamon bun
Barbells salted peanut protein bar- 20g carbohhydrate (CHO)
Salad with veg & chickpeas, smoothie w/ fruit, maple syrup, kale & protein
Bottle with CCD - 80g CHO
Banana - 14g CHO
Poppy seed pound cake slice
Bottle w/Nuun + CCD - 80g CHO
Cookies - 35g CHO
Four Sigmatic Reishi elixir - 2g CHO
Total known CHO 231g

April 16th (Race Day -2)
RX bar - 24g CHO
Chocolate Croissant
Barbells white chocolate protein bar - 13g CHO
1/3rd bottle of Gatorade - 12g CHO
4 handfuls salted almonds
Banana - 14g CHO
Bagel - 44g CHO
Cookie - 13g CHO
Bottle w/Nuun + CCD -80g CHO
Burger + Fries
Probiotic drink - 3g CHO
Meat, veg, hummus, pita & rice
Cookies - 11g CHO
Total known CHO 214g

April 17th (Race Day -1)
Banana - 14g CHO
Bagel - 44g CHO
Rx Bar - 23g CHO
Banana - 14g CHO
Probiotic drink - 8g
Bowl- rice+ quinoa, broccoli, hummus, egg, sweet potato, feta
Half slice of almond strawberry cake
Bottle w/Nuun + CCD - 80g CHO
900g pasta ~ 500g CHO
Half a Green tea drink - 15g CHO
Ginger beer drink
Grapefruit drink
Meatballs & bread
Tagliatelle bolognese
Cookie - 13g CHO
Total known CHO 711g

April 18th (Race Day)
Pre Race:
Banana x2 = 28g CHO
Peanut butter 14g CHO
Bagel 44g CHO
Total CHO 86g

**Note: all carbohydrate values are from packaging or estimates from online nutrition information. Anything without a carbohydrate count is due to portion size issues or no information available. Totals are the total for the values available.

Bussau VA, Fairchild TJ, Rao A, Steele P, Fournier PA. Carbohydrate loading in human muscle: an improved 1 day protocol. Eur J Appl Physiol. 2002 Jul;87(3):290-5. doi: 10.1007/s00421-002-0621-5. Epub 2002 May 28. PMID: 12111292.

Burke LM. Nutrition strategies for the marathon : fuel for training and racing. Sports Med. 2007;37(4-5):344-7. doi: 10.2165/00007256-200737040-00018. PMID: 17465604.

Louise M. Burke, John A. Hawley, Stephen H. S. Wong & Asker E. Jeukendrup (2011) Carbohydrates for training and competition, Journal of Sports Sciences, 29:sup1, S17-S27, DOI: 10.1080/02640414.2011.585473.

Murray B, Rosenbloom C. Fundamentals of glycogen metabolism for coaches and athletes. Nutr Rev. 2018;76(4):243-259. doi:10.1093/nutrit/nuy001.

Dylan Wilburn, Steven Machek, Ahmed Ismaeel. Highly Branched Cyclic Dextrin and its Ergogenic Effects in Athletes: A Brief Review. Short ReviewJournal of Exercise and Nutrition: 2021, Volume 4 (Issue 3): 15

Jeukendrup AE. Training the Gut for Athletes. Sports Med. 2017 Mar;47(Suppl 1):101-110. doi: 10.1007/s40279-017-0690-6. PMID: 28332114; PMCID: PMC5371619.