A strength program whose only job is to make you a faster runner
Paste this into Claude. Answer the questions it asks. It builds the plan, and if you have Google Calendar connected it will put the sessions on your calendar for you.
Heavy multi-joint free-weight loading of the lower body is the only strength method that significantly improved race performance in the 2024 meta-analyses. Plyometrics are an addition on top of that base, never a replacement. Calf, hamstring and trunk work is in the program to keep you running, and is labelled as such — injury-prevention literature, not running-economy trials.
It will ask before it builds
Claude checks whether it can see your calendar or training data, then asks about the race, terrain, weekly running, lifting history, equipment, injuries and how many days you can actually lift. It is instructed not to write a program until those answers exist. If your race is too close for a heavy block, it is instructed to say so rather than build one anyway.
Two or three different sessions, phased across the season
A repeated Monday workout is a failed program. The prompt demands a squat-pattern session and a single-leg / posterior-chain session, a ramp-in if you have not lifted recently, and a taper before the race. Heavy lifting is never placed in the 24 hours before a quality run or a long run.
What the program must cover
• Heavy bilateral squat-pattern work — the tier-1 anchor
• A heavy hinge (trap-bar deadlift, Romanian deadlift, or hip thrust)
• Loaded single-leg work, because running is a single-leg task
• Eccentric hamstring work, bent-knee calf loading, and a small trunk anti-rotation dose
• Plyometrics phased in after the heavy base is established, not from week one
What it will not claim
• A bigger or more toned body — concurrent running suppresses hypertrophy, and that is the point
• A higher VO2max — the Storen study left it unchanged, and this program is not for that
• That a side-on phone clip can see load, bar path, or whether you are on your toes
• That the gym plan replaces filming your form — the scorecard is still how you know the running changed
Questions
Why paste this into Claude instead of generating it in the app?
The in-app plan is written from a scorecard and attacks the worst-banded metric on camera. This prompt is gym loading the camera cannot see. Claude can also put the sessions on Google Calendar if you have it connected, which the app does not do.
Will this make me faster?
It is a reasonable bet, not a promise. The anchor study had eight well-trained runners in the lifting group. Recreational runners usually have more to gain and more to lose if the lifting is dumped on top of a week that is already too hard.
Do I still need to film?
Yes, if you want to know whether overstride, cadence or trunk lean actually moved. The gym numbers going up is not the same evidence.
Keep your results
The free check scores one clip and forgets it. An account saves every analysis, writes the 4-week plan around your worst band, and shows whether the number actually moved when you re-film. Five analyses a month, free.