Jabbr

• If you need independently validated accuracy metrics (published benchmarks, audited error rates) immediately—public materials emphasize announcements and provide limited third-party validation.
• If you require publicly available enterprise security documentation and certifications (SOC 2 Type II, ISO 27001) as a procurement prerequisite.
• If your workflow cannot support multi-camera capture or you only have inconsistent single-angle footage (computer-vision performance may degrade).
• If your primary need is developer-accessible boxing schedules and fighter metadata rather than video-derived statistics—an API provider may be a better fit.
• If you are looking for a single statistic to replace formal fight scoring: even detailed punch metrics are an imperfect proxy for judging.

Limitations evidenced in public materials and community discussion

• Accuracy and definition disputes are common: differentiating "landed" versus "blocked," grazing shots, and handling occlusions are ongoing challenges.
• Punch statistics have conceptual limitations for judging close fights; counts do not encode damage or the qualitative effect of strikes.
• The vendor's public security posture is limited to baseline privacy and terms pages; enterprise-grade security details are not publicly documented.

Roadmap / future features

• Vendor communications suggest expansion beyond simple counts into additional metrics (for example, footwork, balance, stance) and automation of highlights and streaming. Buyers should request a dated roadmap and reference customers for feature commitments.

Differentiators (partly validated by community interest)

• Video-derived automation compared with manual counting methods, and an approach that aims to provide auditability via video clips tied to counts.
• A more detailed strike taxonomy than simple jab/power categories (for example, quality/impact tiers) which may better reflect aspects of fight dynamics.

Where competitors still have the advantage

• CompuBox: entrenched distribution and familiarity in many boxing broadcasts.
• UFC official statistics: official status and historical continuity within its domain.
• API aggregators: easier self-serve onboarding and transparent pricing.

Jabbr publishes news posts describing deployments; these items function as vendor-published case studies.

Case: Tyson Fury vs Oleksandr Usyk (May 18, 2024) — live fight statistics (vendor statement)

• Vendor statement: DeepStrike provided live fight statistics for the heavyweight unification bout on May 18, 2024.
• Critical evaluation: Independent corroboration from the broadcaster or promotion would strengthen the credibility of the claim; publicly accessible methodology and validation details are limited.

Case: TNT Sports Box Office AI stats launch

• The vendor described a live-stats deployment with TNT Sports Box Office, stating that DeepStrike logged punches from multiple camera angles and slow-motion clips.
• Critical evaluation: Multi-angle and slow-motion processing are plausible operational approaches for improving detection accuracy. The vendor post does not provide quantified accuracy metrics or independent audits.

Case: DAZN deployment (AJ vs Dubois)

• The vendor reported metrics such as one fighter outlanding another on certain impact categories.
• Critical evaluation: These vendor-generated metrics illustrate the product's output, but buyers should request access to raw data, definitions, and reproducibility tests before relying on them for operational decisions.

Jabbr has been discussed in threads sharing AI-derived punch statistics and comparisons with CompuBox; these threads include both expressions of interest and skepticism.

Positive / interest signals (examples)

• Some users express a preference for an automated approach compared to perceived human bias in manual systems. Example comment from a thread: "I personally prefer Jabbr cause it uses AI and isn't biased..." (user-generated content; not independently verified).
• Some users indicate a preference for Jabbr while simultaneously questioning the general utility of punch statistics.

Critical / skeptical feedback (examples)

• Some users question whether AI punch counting is sufficiently accurate, reflecting broader skepticism of punch statistics as an analytic tool in close fights.
• In discussion threads, users debate whether discrepancies are due to errors in Jabbr's processing or in CompuBox's human-generated counts; this implies a validation and trust challenge for automated systems.

Why customers/users chose Jabbr (inferred from discussions)

• Reported motivations include a desire for increased transparency (for example, vendor statements about publishing punch-count video clips), consistency and auditability compared to human operators, and a more detailed classification of strike quality.

Why some might not choose it (inferred)

• If stakeholders do not accept third-party automated statistics as authoritative, or if promotions and commissions distrust external analytics providers.
• If a broadcast workflow cannot support multi-angle capture or slow-motion clip logging.

What the product does (objective value proposition)

• Jabbr is an AI/computer-vision analytics platform for combat sports (boxing and MMA) that generates automated fight statistics and related content (for example, highlights) from video. The product name most frequently referenced is DeepStrike (the company's computer-vision AI).

Target market & positioning

• Primary buyers/users (inferred from messaging and deployments):
  • Broadcasters & streaming platforms (referenced in announcements about live AI stats launches with partners).
  • Fight promotions / event producers seeking on-screen statistics packages and enhanced broadcast graphics.
  • Gyms & training facilities (vendor mentions of custom hardware and gym streaming/stats workflows).
  • Fans, media and analysts consuming post-fight statistics.
• Positioning: The company presents the product as an option to update analysis workflows and to increase transparency in combat sports judging and analysis by providing consistent automated counts and additional metrics.

Company background (founding year, HQ, leadership)

• Public profiles commonly describe Jabbr as based in Copenhagen, Denmark, and at an early (seed) stage. Regional press coverage and company databases align on this.
• Founders listed across company databases include Allan Svejstrup Nielsen and Elias Obeid.

Funding & growth (rounds, notable investors, caveats)

• Regional press and third-party databases report a recent seed financing event; coverage references venture and angel backers. Exact totals and investor lists vary by source.
• Practical caveat: treat aggregated funding figures as directional unless confirmed via a primary source such as a vendor press release or regulatory filing.

Pricing model (tiers and costs)

• No clear public pricing was found on the vendor's public site for DeepStrike or for broadcast/gym deployments; site content is largely product narrative and news.
• Practical implication: buyers should assume quote-based enterprise pricing with negotiation dependent on usage scope (events, camera angles, automation level, rights, and whether custom hardware is included).

Total cost of ownership (TCO) considerations

• Implementation & operations: If DeepStrike requires custom hardware and multi-camera ingest, TCO likely includes hardware, setup, calibration, operator workflows, and training.
• Ongoing costs: model updates, support SLAs for live events, integration with broadcast graphics packages, and storage/compute for video processing.
• Risk cost: if automated statistics are disputed publicly, there can be reputational impacts for broadcasters and promotions.

Key competitors / alternatives (4–6) Combat sports statistics and analytics is a fragmented category; competitors are often category-adjacent rather than direct like-for-like:

1. CompuBox (boxing punch statistics produced by human operators; de facto incumbent in many boxing broadcasts)
2. UFC Stats / FightMetric ecosystem (official MMA statistics for UFC; not a boxing solution and not computer-vision based)
3. Boxing Data API (developer-focused boxing schedules, fighter data, and where available, aggregated statistics)
4. Cage Metrics (advanced MMA analytics built on official statistics; adds derived metrics)
5. MMA Labs (UFC fighter database and statistics tooling)
6. Other emerging AI-based judging/stats applications and analytics sites (varied credibility; not always primary data providers)