Carnivore Diet 90 Day Review, Part 5

3 min readApr 24, 2021
Photo by Emerson Vieira on Unsplash

I specifically wanted to look at caffeine / performance and cold / recovery so, before we move on let’s see what those relationships looked like:

Interestingly enough, in all those cases, the relationships (correlations) were weak at best, and not remotely statistically significant! Let’s check check if there were any less significant correlations, say at 99% confidence (p = 0.01).

Discussion: P = 0.01

For RHR the only significant variable was CVL (discussed previously). Regarding AWQ, relationships to sleep routine and fast time were discussed previously. At p = 0.01 there is an additional positive correlation with sleep. This is interesting because several other variables are included in the AWQ, but it is sleep that appears most relevant — or perhaps is most variable. Protein also had a significant (at p = 0.01) relationship with AWQ score at about the same strength as sleep.

With regard to performance (Kb.Load), Kb.Weight has been discussed previously. However, there were several other interesting relationships. Work percent (r = +0.46, p = 0.0091) and rest percent (r = -0.46, p = 0.0091) which suggests that adaptation (over time) has a strong impact on performance. For example, increased duration without a rest day should eventually decrease performance, but the data in this study do not reflect that.

What is more expected is that mood was positively correlated with performance, as was AWQ. The later helps clarify the above; suggesting that consecutive training days to not negatively impact performance so long as adequate recovery (AWQ) is taking place. The relationship between mood and performance is of particular note because it suggests that psychological and emotional sates have a stronger relations with performance than even recovery (AWQ r = +0.48, mood r = +0.51)!

Discussion: Blood Pressure

Over the past year I’ve been trying to pin down causes of my by elevated blood pressure without much success (trials with caffeine, sodium, magnesium, etc…). As I’ve noted in the past, I’ve started using CVL as a measure of heart health, though I’d still like to get to the bottom of blood pressure specifically.

Regarding CVL, there were no additional significant relationships when the confidence interval was decreased to p = 0.01. However, for MAP training intensity had a negative (r = -0.5) relationship; suggesting that more intense activity actually decreases MAP.


Mood and recovery are extremely important variables regarding ongoing, persistent, and consecutive performance which impacts a number of variables (e.g. progression over time relies on the ability to continue to perform over time). Recovery also appears most dependent on sleep and sleep routine. Performance appears most dependent on recovery and mood.

Follow me on Instagram (@savagezen) and tune in to The Kombat Kitchen Podcast!

Originally published at Graphs, plots, and screenshots can be found on Google Photos. Source code and raw data are available on GitHub and Google Drive.




Grappling. Crossfit. Carnivore. The mind is secondary. | All content now on Substack: