perfectgame.org — AI Site Grade

Analysis

Perfect Game USA — AI-Visibility Audit

The site's most consequential AI-visibility problem is not blocked crawlers or missing content — it is a complete absence of structured data across every page examined, leaving the site's massive player database and event catalog semantically invisible to AI engines.

Crawler Access

All major AI crawlers — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, ChatGPT-User, Bytespider, Applebot-Extended, anthropic-ai — receive a full 200 response with identical byte payload (253,505 bytes) as a browser. No UA-based blocking exists. The robots.txt contains no AI-specific directives; the wildcard User-agent: * rule disallows /MyPG/, /admin/, /Players/PlayerProfile_*.aspx (the individual player profile sub-pages), and other admin paths, but the homepage, rankings, events, and article pages are all permitted. The site runs on Microsoft-IIS/10.0 behind Cloudflare DNS, with no CDN/WAF layer that discriminates by bot. llms.txt returns a 404, meaning no AI-friendly content map exists.

Schema Posture

Zero JSON-LD schema was found on any page tested — homepage, About page, FAQ, player profile, rankings page, testimonials, or scout blogs. No Organization, WebSite, SportsTeam, Person, Event, or FAQPage markup exists. The FAQ page (/faq.aspx) contains 8,439 words of genuine Q&A content but is not marked up as FAQPage. Player profiles contain structured data (PG grade, position, stats, commitment status) rendered in HTML tables but carry no Person or SportsTeam schema. The homepage's og:type is website — no og:type=article or profile on content pages.

Cold-Knowledge Gap

The LLM knows Perfect Game as "the largest amateur baseball scouting and event organization in the U.S.," founded in 1995 by Jerry Ford and Rob Naddelman, and accurately describes the National Showcase, All-American Classic, and the player database. However, the model also surfaces reputational signals the site itself never addresses — high participation costs, pay-to-play dynamics, and overuse injury criticism. The site's About page and Testimonials page (featuring Brian Cashman and Ben Cherington quotes) present an unbroken positive narrative. The cold model knows about controversies the site's own content omits entirely, creating a credibility gap for AI engines that synthesize both sources.

Content Architecture

Every page carries a massive, repeated navigation sidebar (~1,000+ words of identical menu HTML) that inflates page weight to 250KB+ and dilutes the content-to-noise ratio. The homepage's only <h1> is "SPEND MORE. SAVE MORE." — a promotional banner, not a brand or value-proposition heading. The actual brand statement ("THE WORLD'S LARGEST AND MOST COMPREHENSIVE SCOUTING ORGANIZATION") appears as uppercase body text, not in any heading or schema. The sitemap is impressively deep — 30+ sub-sitemaps indexing over 100,000 URLs including player profiles, events, and teams — but none of those pages carry structured data. Player profile pages (e.g., Playerprofile.aspx?ID=1) contain scouting grades, metrics, and commitment data in HTML that AI crawlers can read but cannot semantically classify.

Findings

1. Zero JSON-LD schema on any page High

   No JSON-LD structured data was found on any page tested, including homepage, About, FAQ, player profiles, rankings, testimonials, and scout blogs. This leaves the site's content semantically invisible to AI engines.

   What to change: Add JSON-LD structured data for Organization, WebSite, SportsTeam, Person, Event, and FAQPage schema on relevant pages.

2. FAQ page lacks FAQPage schema Medium

   The FAQ page contains 8,439 words of genuine Q&A content but is not marked up as FAQPage schema, missing an opportunity for rich results and AI extraction.

   What to change: Add FAQPage schema to the FAQ page, wrapping each question-answer pair in the appropriate structured data.

3. Player profiles lack Person schema High

   Player profile pages contain scouting grades, metrics, and commitment data in HTML tables but carry no Person or SportsTeam schema, preventing AI engines from classifying the data.

   What to change: Add Person and SportsTeam schema to player profile pages, including properties for name, position, stats, and team commitment.

4. Homepage H1 is a promotional banner, not a brand statement Medium

   The homepage's only H1 is 'SPEND MORE. SAVE MORE.' — a promotional banner. The actual brand value proposition appears as uppercase body text, not in any heading or schema.

   What to change: Change the homepage H1 to reflect the brand and value proposition, e.g., 'Perfect Game USA — World's Largest Baseball Scouting Service'.

5. llms.txt returns 404 Medium

   The llms.txt file returns a 404 error, meaning no AI-friendly content map is provided for language models to discover key pages.

   What to change: Create an llms.txt file listing the most important pages for AI consumption, such as About, FAQ, rankings, and event pages.

6. Reputational credibility gap between site content and AI knowledge Medium

   The LLM knows about controversies (pay-to-play, overuse injuries) that the site's own content omits entirely, creating a credibility gap when AI engines synthesize both sources.

   What to change: Consider adding a balanced narrative or FAQ addressing common criticisms to align site content with external perceptions.

7. Heavy navigation sidebar inflates page weight Low

   Every page carries a massive, repeated navigation sidebar (~1,000+ words of identical menu HTML) that inflates page weight to 250KB+ and dilutes content-to-noise ratio.

   What to change: Optimize the navigation sidebar by lazy-loading or reducing repeated HTML to improve page weight and content ratio.

8. Missing og:type on content pages Low

   The homepage's og:type is 'website', and no og:type='article' or 'profile' is used on content pages like blogs or player profiles, limiting social and AI understanding.

   What to change: Add appropriate og:type tags (e.g., 'article' for blog posts, 'profile' for player pages) to improve semantic classification.

What's working

• All major AI crawlers have full access — All tested AI crawlers receive a full 200 response with identical content as a browser, with no UA-based blocking or discriminatory directives in robots.txt.
• Deep sitemap with 100,000+ URLs — The sitemap indexes over 100,000 URLs including player profiles, events, and teams, providing excellent crawl coverage for search engines.
• Player profiles contain rich HTML data — Player profile pages include scouting grades, metrics, and commitment data in HTML tables that AI crawlers can read, even if not semantically classified.
• Strong LLM brand awareness — The LLM correctly identifies Perfect Game as the largest amateur baseball scouting organization, founded in 1995, and accurately describes key events and the player database.
• Testimonials from credible baseball figures — The Testimonials page features quotes from Brian Cashman and Ben Cherington, providing authoritative third-party validation.