Understanding the Movement Mesh Network, Predicted Sets, and Calibration Sets
Volt’s Movement Mesh Network is the intelligence behind how we recommend weights for your workouts — even when you haven’t performed a specific movement recently. It helps power Cortex®, the engine behind Volt’s strength recommendations, to suggest more accurate and personalized weights.
Instead of treating each lift in isolation, Volt connects related movements across your training history to deliver smarter, more accurate loading recommendations. The result: fewer guesswork sets, smoother progress, and better training continuity.
What is the Movement Mesh Network?

The Movement Mesh Network is Volt’s system for understanding how different exercises relate to one another.
Many movements train similar muscle groups or movement patterns — for example, back squats, front squats, and goblet squats. The mesh network uses these relationships to estimate your current strength more accurately, even if you haven’t logged a specific movement in a while.
Rather than relying on outdated data or forcing unnecessary calibration sets, Volt looks at your most recent performance on related movements to keep your training on track.
This system powers our Predicted Sets feature.
What are Predicted Sets?

Predicted Sets are weight recommendations generated using the Movement Mesh Network.
When you start a set, Volt determines the most accurate load by:
- Using your most recent e1RM for that movement, or
- Referencing closely related movements if you haven’t performed that exercise before
This allows Volt to recommend appropriate weights even when your direct history for a movement is limited.
In short: Predicted Sets use what Cortex already knows to save you time and guesswork.
What are Calibration Sets?

Calibration Sets appear when Cortex doesn’t yet have enough data to confidently predict your starting weight.
This typically happens when:
- You’re performing a movement for the first time
- You haven’t logged a similar movement before
- There’s not enough recent data from related exercises
In a Calibration Set, you’ll be prompted to enter a weight you’re confident you can lift for the given number of reps.
This set helps Cortex learn your strength so it can guide you more accurately in future sets and workouts.
In short: Calibration Sets help Cortex learn you.
How Calibration Sets and Predicted Sets Work Together
These two features are designed to work as a progression:
-
No relevant data yet?
→ Cortex uses a Calibration Set
-
Some related or recent data available?
→ Cortex uses a Predicted Set
-
You log your performance
→ Cortex updates your e1RM and improves future recommendations
Over time, as you train and log sets with weight, reps, and RPE, you’ll see fewer Calibration Sets and more accurate Predicted Sets.
How Predicted Sets Help Your Training
Predicted Sets are designed to:
- Reduce unnecessary calibration sets
- Prevent big jumps or drops in loading
- Keep workouts flowing smoothly
- Maintain training momentum across cycles
As you log sets with weights, reps, and RPE, Volt continues to refine your e1RM — making future predictions even more accurate.
Will Predicted Sets replace my actual performance data?
No. Your real performance always takes priority.
Predicted Sets are simply a starting point. Once you log a set, Volt uses that data to update your movement-specific e1RM. Over time, recommendations rely more heavily on your direct history with that movement.
Think of Predicted Sets as a smart bridge — they help you train confidently until enough direct data is available.
In Short:
- The Movement Mesh Network connects related exercises across your training history
- Predicted Sets use that network to recommend smarter weights
- Calibration Sets help Cortex learn when there’s not enough data
- Volt adapts automatically as you train and log data
- You spend less time guessing — and more time progressing
If you ever have questions about a recommended load or feel something doesn’t match your ability, trust your judgment and adjust as needed. Volt will learn from it.