Methodology

This page explains how the Domainer Legends scoring models are constructed and what they are — and are not — intended to represent.

1. What These Models Are

Each model is a rule-based heuristic inspired by a specific domainer's publicly visible behavior: blog posts, interviews, conference talks, and well-known domain purchases or sales. The goal is to answer:

• "How compatible is this domain with what this person says and seems to like?"
• "Does this name fit the patterns of their public portfolio and philosophy?"

The output is a 0–10 score that reflects alignment with each domainer's publicly discussed criteria. It is not a price estimate or appraisal.

2. What These Models Are Not

• They are not affiliated with, endorsed by, or approved by any domainer.
• They do not reflect private deals, unpublished strategies, or confidential data.
• They do not guarantee liquidity, value, or sale probability.
• They are not financial, investment, or legal advice.

3. Shared Inputs Across All Models

Every model looks at the same basic features of a domain and then applies different weights per domainer:

• Extension (TLD): e.g., .com, .net, .org, others.
• Main label (before the dot): cleaned of punctuation to compute length.
• Length: how many alphanumeric characters the core label has.
• Structure: whether the label includes hyphens or numbers.
• Keyword signals: presence of commercial, geo, service, or brand-related keywords.
• "Single strong string" feel: whether the label looks like one tight concept or a messy mashup.

3.1 Scoring Scale

All models map their internal score to a 0–10 range:

• 9–10: Very strong alignment with that domainer's public style.
• 7–8: Solid alignment with multiple strong factors.
• 5–6: Mixed; some elements fit, some don't.
• 3–4: Weak alignment.
• 0–2: Very low alignment.

4. Shared Feature Logic

4.1 TLD Preference

Almost all domainers listed here are strongly biased toward .com. The models reflect this by giving .com a positive weight, sometimes a very strong one. Classic extensions like .net and .org may get a smaller boost for some domainers.

4.2 Length

Shorter labels are usually better: easier to remember, type, spell, and brand. The models generally:

• Reward "short" ranges (e.g., up to 8–12 characters).
• Give a smaller boost to medium ranges.
• Apply a penalty to very long labels, especially for premium-focused domainers.

4.3 Hyphens and Numbers

Hyphens and numbers tend to lower perceived quality in high-end domains. The models:

• Use strong penalties for ultra-premium brand-focused domainers (e.g. Rick, Andrew, Brent).
• Use milder penalties for monetization/geo-oriented models (e.g. Page Howe, Adam, Yun Ye).

4.4 Keyword Groups

Each model uses one or more keyword lists, such as:

• Commercial keywords: finance, travel, property, jobs, health, etc.
• Brand / startup suffixes: cloud, labs, media, capital, group, etc.
• Geo / service terms: plumber, dentist, realtor, cleaning, etc.

When a domain's core label contains one of these keywords, the model adds a positive score to reflect: "This sits in a valuable or clearly monetizable category for this domainer."

4.5 "Single Strong String" Feel

Many legendary domains are just one powerful word or a tight phrase. The models approximate this by:

• Checking if the label is one piece (no hyphens).
• Ensuring the length is within a "brandable" range.
• Rewarding those labels with a small extra boost.

5. Model-Specific Notes

6. How to Use These Scores

• Compare how the same domain scores across different domainers.
• Use high scores as a signal that your name fits a particular investing philosophy.
• Use low scores as prompts to ask "why?" — too long? wrong TLD? cluttered structure?

A low score doesn't mean a domain is "bad", and a high score doesn't guarantee a sale. The tools are designed as educational lenses, not verdicts.