People, every show
Everyone on the record, counted once across the shows in view, with the same record a show's own people table carries. Click a header to sort everyone, a name for that person across every show, a show tile for their page there. Corpus average certainty is 3.9/5; deltas show each person against it. Argument clarity carries a full score for 876 people (gold bar, raw unedited tape) and a coarse estimate for 4,440 more (blue bar); on the hub a score rests on the show with the most assessed tape, and the tooltip names it. Roles are as of the recording.
shows 
every show 44 of 44
| Person | Shows | Eps | Statements | Claims | Supported | Partly supported | Contradicted | Full support rate | Certainty | Argument clarity | WPM | Fillers/1k |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
 Guillermo Rauch ×11 Founder & CEO, Vercel |    +4    | 11 | 262 | 124 | 26 | 6 | 3 | 74% 26/35 | 3.8/5 -0.1 | 4.2 | 241 | 11 |
 Dylan Field ×13 Co-founder and CEO, Figma |    +4 | 13 | 202 | 79 | 23 | 2 | 0 | 92% 23/25 | 3.7/5 -0.2 | ≈4.5 | 300 | 38.4 |
 Reid Hoffman ×100 host Partner, Greylock |   +5 | 100 | 672 | 223 | 83 | 12 | 9 | 80% 83/104 | 3.9/5 +0.0 | 4.1 | 243 | 18.2 |
 Brian Armstrong ×17 Co-Founder & CEO, Coinbase |   +5 | 17 | 351 | 176 | 67 | 14 | 6 | 77% 67/87 | 3.7/5 -0.2 | 4.1 | 287 | 13.1 |
 Scott Wu ×16 Co-founder & CEO, Cognition |  +5 | 16 | 239 | 129 | 29 | 6 | 0 | 83% 29/35 | 3.6/5 -0.3 | 3.9 | 311 | 15.6 |
 Eric Ries ×11 Founder & CEO, Long-Term Stock Exchange |   +5 | 11 | 219 | 85 | 37 | 11 | 4 | 71% 37/52 | 3.9/5 +0.0 | ≈4.5 | ≈314 | ≥6.2 |
 Elad Gil ×146 host Founder, Gil Capital | +6 | 146 | 521 | 227 | 62 | 13 | 14 | 70% 62/89 | 3.5/5 -0.4 | 4.0 | ≈262 | ≥2.2 |
 Aaron Levie ×34 Co-Founder & CEO, Box |  +6 | 34 | 447 | 214 | 21 | 2 | 1 | 88% 21/24 | 3.7/5 -0.2 | 4.3 | 283 | 25.8 |
 Sam Altman ×37 CEO, OpenAI |  +8 | 37 | 721 | 281 | 42 | 6 | 3 | 82% 42/51 | 3.7/5 -0.2 | 4.1 | 280 | 16.2 |
 Amjad Masad ×18 Co-founder & CEO, Replit |  +8 | 18 | 450 | 209 | 34 | 15 | 4 | 64% 34/53 | 3.7/5 -0.2 | 4.3 | 249 | 31.7 |
Why this person has no speaking number
We measure speaking style by listening to the audio itself: a model transcribes the tape verbatim, every "um" and "uh" kept, and articulation rate comes from the word timings. A fair number needs at least 2,000 words from one person on tape we have measured, and we have less of them than that. A rate from a smaller sample would be noise, so we print nothing. It says nothing about how they speak.
A plain figure in these columns comes from raw-level tape. A dim figure with ≈ or ≥ comes from the aired feed of an edited show, where the editor cut some hesitation before release, so it is a floor. How it's measured →
every person on the shows in view, most shows first and most checkable claims within a tie; every column header sorts all of them. eps, statements, claims and the record are summed across their shows; eps leaves out compilation re-airs and clips rolled into someone else's episode, the same rule as every person page. Sorting by statements or eps across shows uses the checkable claim count and the show count, the two numbers that exist for everyone at once. certainty is weighted by statements. argument clarity, WPM and fillers come from one show's tape, the one with the most assessed exchanges or measured words; hover a figure to see which. full support rate = fully supported ÷ resolved; sorting it puts records of five or more resolved first, and percentages on fewer than 5 resolved statements are greyed. A ? means under 2,000 words of measured tape anywhere.