A better fighter is one that performs more reliably for your documented role and investment, not simply a card with a higher-sounding label. Record the card, develop one weakness, compare repeated matches, and reroll only with cash above the progression reserve.
Develop the current card or recruit another?
Decide what better means before opening a recruitment screen: reaching a visible gate, doing the same job with less preparation, or improving results against a particular opponent type. Keep a useful existing card while evaluating a candidate if the current game permits it; this guide does not assume selling, refunds, or transfer of training investment.
| Situation | Reasonable next step | Cost to include |
|---|---|---|
| A useful card has a visible weakness | Test a bounded training block. | Training plus the next comparison match. |
| The roster lacks a role you can describe | Inspect the recruitment options and set an attempt limit. | Acquisition and initial preparation, not just one reroll. |
| The current card is uncaptured | Save its baseline before replacing it. | Already spent training is not recovered by forgetting the card. |
| A new card only looks rarer | Compare visible fields and the intended role. | Unknown performance cannot justify unlimited rerolls. |
Define better for the roster
The official description advertises random fighters, real fighters, and real legends, but it does not publish a complete roster or universal ranking. Define the role or progression problem before comparing cards.
- Current league or match need.
- Visible strength and weakness.
- Training cost and time already invested.
- Performance against comparable opposition.
Record the card before developing it
- Exact name and card group.
- Visible starting values, strength, and weakness.
- Acquisition source and displayed cost.
- Screenshot date and update title.
Give one fighter a controlled test
A new card cannot be compared fairly after receiving a completely different budget and opponent path. Keep the training investment and match context as similar as the live game allows.
- Train one weakness.
- Record total cost.
- Run several comparable matches.
- Keep losses in the dataset.
When a reroll is reasonable
Use the current reroll tier and displayed price, then calculate whether the attempt fits above the protected reserve. Exact pull odds are not published in the official public description.
- Do not chase an unpublished probability.
- Record every sequential result, not only legends.
- Stop when the reserve would be crossed.
Build a source-backed roster
Save fighters in Fighter Lab, assign private tiers only after your own evidence, and export the notebook before changing devices. Personal rankings remain local and are not published as universal claims.
- Separate observed facts from your preference.
- Version records after updates.
- Submit clean evidence when a public claim can be reproduced.
- The official description advertises 250+ cards, including random fighters, real fighters, and real legends.
- Training strengths and weaknesses and watching match results are part of the official loop.
- The complete roster, card odds, and universal ranking are not publicly documented.
- A fighter's value depends on card data, investment, opponent, and update context.
Sources and scope
Official links support the systems and labels described below. Decision tables, checklists and labeled worked examples are this site’s practical guidance, not measured game rules. Publisher and community material retains its own source status.