Raise the Average

Breakthrough→Raise the Average→Breakthrough

Ötillö Casco Bay 2023

As we settle into 2026, I wanted to bring up something I found myself talking about a lot with athletes in 2025: the idea of raising the average of our performances.

This was often brought up in contrast to primarily assessing athletic progression based on personal bests (PB).

A quick personal anecdote:

In the fall of 2007, I raced Ironman Florida and finished in tenth place in 8:40. It was a 17-minute PB for me, the first time I broke 3 hours on the marathon, and I was ecstatic. It was a major breakthrough day for me where I felt like all my training and preparation had come together and I was thriving.¹

In the fall of 2008, I raced Ironman Arizona and just never seemed to have ‘it’ on the day. Everything felt harder than it should, my motivation was waning and I certainly did not feel I was thriving.

I also finished this race in 8:40.

I was initially pretty disappointed in the day as I felt like I was capable of something more, but after some time I came to realize that a time and performance that I once considered to be me performing at my peak (8:40) had now become something I could do when I felt pretty mediocre.

I mean sure, it would have been better to race faster and see some improvement when I was at my best, but I started to realize that I had seen improvement: I had raised the level of my average performance.

In January of 2023, I wrote an article titled Most Days Are Just Days where I tried to underscore the importance of understanding that race days are not magical days. They are just days like any other day, but we still need to try and get the most out of ourselves on those days, no matter what the circumstances may be.

This is why I think a superior way to assess athletic progression is to shift away from solely judging it on our single best performances and giving more credence to the rolling averages of our performances. This is not because I think personal bests don’t matter, it is because I think that raising the average makes new personal bests more likely.

There are a couple of ways that I think about raising the average:

1. What the average actually amounts to

2. The depth of the average

Here’s what I mean:

What the average amounts to

This idea focuses on the numbers themselves. For example, a swimmer who does several 200 races in 2:19, 2:23, and 2:27 has an average swim time of 2:23. However, the range is 8 seconds and the number of reps is small.

The depth of the average

Let’s say there is another swimmer who races the 200 with an average of 2:23, but they have seven races with times of 2:22,2:22,2:23,2:23,2:23,2:24,2:24. This second swimmer technically has a slower PB than swimmer number one (2:22 vs 2:19), but the depth of their average performances is stronger and the range of performances is tighter (2 seconds vs 8).

Now, there are strengths to both athletes. One has a faster PB and one has a deeper average, but I would posit that if athlete number two breaks through to a new PB, they are more likely to repeat performances in a similar vicinity of that new level based on their consistency.

The legendary running coach Jack Daniels wrote about this observation in his book Daniels’ Running Formula. In the book, Daniels describes that once athletes break through to a new level of performance, they are more likely to go on to repeat it.

I agree with his observation, but I might add to it by saying that athletes that show the ability to replicate performances are more likely to find themselves in positions down the road to make further breakthroughs in performance.

Paradoxically, the ability to make a new level of performance into an average performance leads to the opportunity to supersede it.

Not only that, but I believe that one of the benefits of raising the average is that it translates into robustness in athletes.

Robustness

• Robustness (n): the ability to withstand or overcome adverse conditions or rigorous testing.²

The conditions of competitions will always vary, whether that is from:

1. Countless variables like dynamic environmental conditions (heat, cold, wind, etc.), venues, schedules, etc.

2. The level of competition and the stakes that are at hand (e.g., the difference from local comps to championships).

One of the benefits of athletes that have repeatedly shown their ability to perform is that they are more likely to withstand either of those challenges.

This comes from athletes knowing they do not need magical days to perform, just an average one.

Last year, I saw two athletes advance to a championship competition that would consist of a qualifying round and a final.

Athlete one had a faster qualifying time going into the championship, but was somewhat new to the sport.

Athlete two had a slower qualifying time going into the championship, but was much more experienced in the sport.

Upon entering the qualifying round, neither athlete posted a time that was as fast as their qualifying time, but athlete two still posted a time that was fast enough to advance to the finals while athlete one failed to do so.

Raise the average of training

So far, we have talked specifically about raising the average of performances, but it’s also important to understand this concept in the context of training. Like racing, training has days where it all comes together, but many of those days fall below that standard.

A memorable moment from Alexi Pappas

When Alexi Pappas appeared on the Rich Roll Podcast³ in February 2021, there was an interaction in their conversation that went viral where she discussed her coach’s rule of thirds.

According to Pappas, when she was training for the 2016 Olympics in Rio de Janeiro, she felt lousy during one particular workout and shared that with her coach.

Pappas’ coach, in response, told her not to worry because training will often fall into one of three buckets:

1. One third of workouts will feel good

2. One third of workouts will feel ok

3. One third of workouts will feel bad⁴

Generally speaking, I like this framing of not expecting a disproportionate amount of training to feel good.

However, I would probably widen the OK bucket in the middle quite a bit so that you would be left with something like:

1. 15% will feel truly good

2. 70% will feel ok.

3. 15% will feel truly bad

Regardless of whether you use Pappas’ coach’s numbers or my own, it still requires an understanding that we cannot only assess our progression on the good days because most days will fall below that threshold.

I used to describe a lot of those OK training days as the following:

The training may not have been “good” but it was “acceptable.”

It may not have met the threshold of ‘good,’ but I knew it was still moving the needle.

Final note

To be clear:

I am not suggesting that you cast aside the aspirations of new personal bests, only that you also apply the concept of improving your average performances when evaluating progression.

What we can do on an average day directly contributes to what we can do on our best days.

• Breakthrough→Raise the Average→Breakthrough

1. In 2008, I wrote a two part series on a website that was then called Xtri.com. The first part was about improving from a 12:55→9:20 IM. The second part was about improving from 9:20→8:40 which happened at IMFL in 2007. While the original articles are no longer published you can find copies of Part I & Part II on my old website.

2. as defined by Google.

3. Episode 569 : 1:36:53 timestamp

4. She technically used the term “crappy”