gpt-4.1-mini-2025-04-14· OpenAI LMArena Human Preference 140K
You were exactly right on Q9s! the other hands were ranked as such - A8s, K9s - 10%, 55…
- User
You were exactly right on Q9s! the other hands were ranked as such - A8s, K9s - 10%, 55 - 20%, T5s, 84s - 40%
- Assistant
Thanks for the feedback and the correction! That helps clarify the scale and relative hand strengths even better.
So, with your actual solver-based percentiles:
Hand Actual Percentile (% of hands better) Q9s 10% A8s 10% K9s 10% 55 20% T5s 40% 84s 40%
What this tells us:
- A8s and K9s are roughly on par with Q9s at 10%, all considered solid suited hands with decent playability.
- 55 (a mid pocket pair) is stronger than those suited hands, at 20%.
- T5s and 84s are weaker suited connectors/gappers, grouped around 40%.
If you want, I can help with percentile estimates for any other hands or ranges, or explain why certain hands fall where they do in this scale! Just let me know.
Other models, same or similar prompt
Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.