Understand the numbers before you build a portfolio.
OptimumLotto explains coverage, frequency, validation and scoring so users can make informed, budget-controlled decisions without treating random draws as predictable.
Model behaviour guide
Frequency
How often each number has appeared historically. Useful for review, but frequency does not create future certainty.
Balance
Odd/even, high/low and spread checks help avoid over-concentrated portfolios.
Coverage
Reviews how widely your selected lines cover numbers and pairs without excessive duplication.
Validation
Compares saved portfolios against historical draws and random baselines to avoid blind confidence.
Core statistics explained
| Metric | What it means | How to use it |
|---|---|---|
| Hot / cold numbers | Numbers drawn more or less often in a selected window. | Use for context only; do not assume a number is due. |
| Frequency | Historical appearance count for each number. | Check whether a portfolio is over-weighted to one part of the pool. |
| Odd / even | Balance between odd and even numbers in a line. | Avoid extreme patterns unless intentionally selected. |
| High / low | Distribution across lower and higher number ranges. | Review spread across the full game range. |
| Pair coverage | How often two-number pairs appear across your portfolio. | Reduce duplicate pair concentration across many lines. |
| Overlap | Repeated numbers between lines. | Lower overlap normally gives broader portfolio coverage. |
| Anti-crowd score | Flags common human patterns such as dates, straight lines and neat sequences. | May reduce shared-prize risk but does not increase draw probability. |
| Score | Internal ranking value from configured model weights. | Compare lines within the same model version only. |
| Average mains | Average main-number matches during historical scoring. | Use alongside random baselines, never alone. |
| Best main match | Highest match reached during validation. | Shows historic peak, not a future promise. |
| 3+ / 4+ / 5+ hits | Counts of historical validation thresholds. | Check whether results are materially better than random comparison. |
| Cache status | Whether exhaustive ranked combinations have been generated. | Confirms ranking data freshness for the selected game/window. |
Validation approach
Professional use checklist
Decide the maximum draw cost before generating any portfolio.
Avoid portfolios that repeat too many numbers, pairs or human-pattern lines.
Use historical scoring only as evidence of model behaviour, not a future guarantee.