Recommendation Engine
The recommendation engine ranks possible fixes by impact and confidence so you get one high-leverage priority fix — not twenty competing tips.
The recommendation engine is the layer that turns a list of identified faults into a prioritized action plan. Once the rules engine and heuristic analysis have identified the fault patterns present in your swing. The recommendation engine ranks them by two criteria: impact (how much improvement addressing this fault is likely to produce) and confidence (how certain the diagnosis is given the available data). The highest-ranked combination — highest expected impact, highest diagnosis confidence — becomes your primary recommendation. Secondary faults may be noted, but only the primary recommendation is the focus of your training plan until a retest confirms it has been addressed.
The ranking logic reflects the "one fix, one plan, one retest" commitment that is central to SwingVantage's approach. Athletes who receive twenty tips at the same time do not know where to start, spend practice time on lower-priority issues, and rarely close the loop on any single fix. By surfacing one high-leverage recommendation, the engine ensures that your practice time is spent on the change most likely to improve your performance — and a defined retest protocol checks whether it worked before you move on.
Your profile still shapes the plan. Skill level does not change which fault ranks first or curve your score. It can change how the fix is explained, which drills you get, and how your practice plan is built.
Tips by skill level
Trust the single recommendation surfaced for you rather than trying to work on all identified faults at the same time. The recommendation engine ranked those findings deliberately — the one at the top is where your practice time will produce the most improvement per hour.
Once a retest reads Improved, you can lock in your next fix or start a new swing check. The sequence of recommendations is effectively a development roadmap — each resolved fault clears the way for the next most impactful intervention.
Example
Facing both a face issue and a strike issue in launch monitor data, the engine puts the higher-ranked fix first and holds the other back.
Why it matters
One clear priority beats a pile of tips. The recommendation engine turns analysis into a plan you can actually act on.
How it shows up on video
The recommendation engine output is visible in the Today launchpad as the Focus, Practice, and Retest cards. These are the personalized drill and retest recommendations generated from your fault analysis and progress history combined.
Common mistakes
- Overriding the recommendation to work on a fault not at the top of the priority list. The recommendation engine weighs severity and impact order on purpose. Fighting the priority order usually produces slower improvement.
- Treating the recommendation as a permanent prescription — it updates with new data, so retest after the recommended practice to get your next step.
- Not uploading new sessions regularly — the recommendation engine uses session history to personalize; without fresh sessions, recommendations stay anchored to old analysis.
- Expecting the recommendation engine to account for factors it cannot observe (injury, mental fatigue, equipment changes) — supplement its recommendations with your own context when these factors apply.
In SwingVantage Motion Lab
In SwingVantage, the recommendation engine determines which finding appears at the top of your report as your primary focus and what the associated training suggestion is. The secondary findings visible in your report are those the engine ranked lower — present, but not your focus until the primary is retested and resolved. The main issue shows its priority and why this fix was chosen, so you understand why it leads.
Frequently asked questions
What if I disagree with the top recommendation?
You can see the ranked findings in your SwingVantage report. If the top recommendation does not fit today, Today's "work on something else" option offers ranked alternatives and asks why. Your answer helps shape what Today suggests next. The recommendation is a starting point informed by the data, not a mandate.
How does the recommendation engine handle completely new users with no history?
For a first session, the engine uses only the data from that submission — there is no historical baseline to draw on. The recommendation from a first session should be treated as an initial read, not a definitive diagnosis. A second session recorded the same way shows whether the same fault appears again, which is a better check than one read.
Related terms
- Primary vs. Secondary FaultThe primary fault is the swing issue with the highest severity and impact on performance; the secondary fault is the most likely co-existing issue.
- SeveritySeverity is how much a swing fault is expected to cost you in performance. SwingVantage shows it as a priority label, such as High priority, and Motion Lab adds a Fix or Watch status.
- RetestA retest is a short follow-up protocol attached to each recommendation. It says how to repeat the test, what to compare, and what success looks like, so you can see whether a fix worked.
See it in a real analysis
Try SwingVantage on one of your own swings — free to start.