Skip to content

Instantly share code, notes, and snippets.

Show Gist options
  • Save snoble/e66f841259ef86ebea0f408b5440fba3 to your computer and use it in GitHub Desktop.
Save snoble/e66f841259ef86ebea0f408b5440fba3 to your computer and use it in GitHub Desktop.
A complete markdown document containing (1) a table that rates prominent Canadian political commentators on their party partisanship during the past 30 days—sorted by partisanship spread, with role-fit icons and concise explanations—and (2) an extended prompt detailing the methodology, scoring scale, name-harvesting workflow, and quality checkli…

Canadian Political Commentators – Partisanship Ratings (last 30 days)

Scores run from +10 (ardent booster) to –10 (fierce opponent); 0 = no clear tilt.
Rows are sorted by partisanship spread (max score − min score).
The “Does the tilt match the role?” column shows an icon—✅ bias fits persona, ⚠️ bias clashes—plus a concise note.

Username Public role / self-description LPC CPC NDP Bloc Does the tilt match the role?
Nick Kouvalis Conservative pollster / strategist –9 +9 –6 –5 ✅ Extreme CPC-friendly / LPC-hostile split suits a partisan strategist.
Candice Malcolm True North founder, conservative activist –9 +7 –6 –3 ✅ Right-wing advocate; fierce anti-LPC, pro-CPC stance is her brand.
Ezra Levant Rebel News founder –10 +6 –5 –1 ✅ Hyper-partisan media personality—no neutrality claimed.
Jenni Byrne Former CPC campaign manager –8 +8 –5 –3 ✅ Loyalty to CPC shows in mirror-image scores.
Max Fawcett Columnist; pragmatic-progressive label but consistent Liberal advocate & CPC critic +8 –8 0 0 ✅ Consistently defends LPC and attacks CPC—fits his self-styled realist progressivism.
Ian Brodie Harper’s ex-chief of staff / academic –6 +8 –6 –4 ✅ Conservative pedigree explains huge CPC/LPC gap.
Kathleen Monk NDP strategist / pundit –2 –6 +8 –2 ✅ Strong NDP cheer-leading fits a party strategist.
Charles Adler Radio host, ex-CPC turned critic +6 –8 0 –2 ✅ “Ex-Conservative” persona explains sharp anti-CPC tone.
Nora Loreto Left-wing writer & podcaster –5 –9 +4 0 ✅ Declared activist; intense CPC hostility expected.
Paul Moen Liberal strategist & GR exec +7 –6 –2 –3 ✅ Open LPC operative; tilt fits.
Amanda Alvaro Liberal campaign adviser & pundit +7 –6 –2 –3 ✅ Professional LPC strategist; partisanship unsurprising.
David Herle Veteran Liberal strategist & podcaster +7 –6 –2 –3 ✅ Long LPC résumé; bias matches.
Michael Taube Conservative columnist, ex-speechwriter –6 +6 –2 –2 ✅ CPC/LPC split suits conservative pedigree.
Scott Reid Former Martin-era Liberal strategist +6 –6 –2 –2 ✅ Liberal loyalty baked in; spread normal.
Sandy Garossino National Observer columnist — consistent Liberal advocate +5 –6 +1 –1 ✅ Regularly boosts LPC and attacks CPC; stance fits centre-left brand.
Dale Smith Parliamentary sketch writer — consistent CPC critic +4 –7 –1 –2 ⚠️ Relentless anti-CPC tone heavy for a purportedly non-partisan press role.
Bruce Anderson Abacus chair & LPC adviser +6 –5 –1 –3 ✅ Long-time LPC adviser—bias expected.
Andrew MacDougall Former Harper comms director — consistent Conservative advocate –5 +5 –2 –3 ✅ Routinely defends CPC and attacks LPC, matching past role.
JJ McCullough YouTuber & Washington Post columnist — consistent Conservative advocate –6 +4 –3 –4 ✅ Regularly applauds CPC, critiques LPC—fits right-leaning persona.
Frank Graves EKOS president (pollster) +5 –5 +1 –2 ⚠️ Large LPC/CPC split surprising for a pollster claiming neutrality.
Tim Powers Conservative strategist / lobbyist –4 +6 –4 –3 ✅ Professional CPC adviser; partisan spread expected.
Arshy Mann Host of COMMONS podcast –4 –6 +4 –2 ✅ Progressive watchdog brand; bias tracks.
Greg MacEachern GR strategist & CBC panellist +5 –4 –1 –2 ✅ Liberal sympathies fit client history.
Tasha Kheiriddin Conservative strategist / author –4 +4 –2 –1 ✅ Bias neatly fits conservative pedigree.
Susan Delacourt Toronto Star columnist — consistent Liberal advocate +4 –4 0 0 ✅ Columns regularly frame LPC positively and hit CPC.
Spencer Fernando Conservative blogger — consistent Conservative advocate –3 +5 0 0 ✅ Headlines mirror CPC talking points; bias expected.
Michael Geist U Ottawa law prof & blogger — consistent Liberal critic –6 +2 0 –1 ⚠️ Nearly every tech post attacks LPC policy—strong bias for an academic.
Robyn Urback Globe & Mail columnist — consistent Liberal advocate +3 –4 0 0 ⚠️ Pro-LPC lean stronger than her “non-partisan” billing.
Andrew Leach U Alberta energy-econ professor +3 –4 +1 0 ⚠️ Noticeable anti-CPC lean unexpected for a neutral scholar.
Noor Azrieh Co-host, The Backbench –1 –4 +3 –1 ✅ Progressive bias fits the show’s stance.
Andrew Coyne National affairs columnist (independent) –1 –4 0 –6 ⚠️ Sharp anti-CPC/Bloc tone heavier than “view-from-nowhere” brand.
Ian Bushfield Co-host PolitiCoast podcast +2 –4 +2 0 ✅ Light progressive lean suits podcast tone.
Chantal Hébert Veteran political columnist +4 –2 0 0 ✅ Mild LPC tilt within expectations.
Emmett Macfarlane U Waterloo constitutional scholar –2 –6 0 –1 ⚠️ Heavy anti-CPC bias clashes with claimed neutrality.
Lisa Young U Calgary political-science professor +1 –2 0 0 ✅ Minor lean acceptable in academic commentary.
Mo Amir Host of This Is VANCOLOUR +2 –3 +3 0 ✅ Slight left tilt matches show’s vibe.
Heather Scoffield Econ columnist & policy exec +2 –3 +1 0 ✅ Light centre-left lean fits policy-wonk image.
Rhys Waters Co-host Canada Is Boring +2 –3 +2 0 ✅ Light progressive lean suits comedy commentary.
Michael Moffatt Western U economist & housing analyst +1 –3 +1 0 ✅ Mild lean—still broadly non-partisan.
Bob Fife Globe & Mail Ottawa bureau chief –2 +2 0 0 ✅ Centre-right flavour but balanced overall.
Darrell Bricker Ipsos Public Affairs CEO –2 +2 –1 0 ✅ Small spread—acceptable neutrality for a polling exec.
Mercedes Stephenson CTV National reporter –3 +1 0 0 ✅ Low-level lean within bounds for network journalist.
Jen Gerson The Line co-founder, journalist –2 –2 0 0 ✅ Even-handed sceptic; balanced spread expected.
Shannon Proudfoot Globe & Mail feature writer –1 –2 +1 0 ✅ Mild lean, still broadly balanced.
Kady O’Malley Political newsletter writer –1 –1 0 0 ✅ Near-perfect balance for a veteran reporter.
Matt Gurney Freelance columnist & radio host –1 –1 0 0 ✅ Near-neutral curmudgeon persona holds.
David Akin Global News chief political correspondent –2 –1 0 0 ✅ Low-spread scepticism suits network correspondent role.
David Cochrane CBC Power & Politics host –1 –1 0 0 ✅ Near-neutral anchor stance.
Nik Nanos Nanos Research founder +1 0 0 0 ✅ Essentially neutral for a pollster.
Andrew Parkin Pollster, Canadian Survey Stuff 0 0 0 0 ✅ Demonstrably neutral—ideal for polling analysis.
Shachi Kurl Angus Reid Institute president 0 0 0 0 ✅ Maintains pollster neutrality.
Trevor Tombe U Calgary economist 0 0 0 0 ✅ Textbook academic neutrality.
Peter Mansbridge Former CBC anchor & podcaster 0 0 0 0 ✅ Legendary neutrality continues.
Rosemary Barton CBC chief political correspondent 0 0 0 0 ✅ Anchor neutrality achieved.
Steve Paikin TVO The Agenda host 0 0 0 0 ✅ Long-standing even-handedness.
Michael Serapio CPAC PrimeTime Politics host 0 0 0 0 ✅ Impartial public-affairs anchor.
Catherine Cullen Host of CBC’s The House 0 0 0 0 ✅ Straight-news neutrality.
Aaron Wherry CBC parliamentary reporter 0 0 0 0 ✅ Straight-news neutrality.
Glen McGregor Independent investigative journalist 0 0 0 0 ✅ Watchdog neutrality intact.
Philippe J. Fournier 338Canada polling modeller 0 0 0 0 ✅ Rigorously neutral data analyst.
Stephanie Levitz Toronto Star reporter 0 0 0 0 ✅ Neutral beat reporting.
Marieke Walsh Globe & Mail parliamentary reporter 0 0 0 0 ✅ Straight neutrality.

Scores range from +10 (ardent booster) to –10 (fierce opponent); 0 = no clear tilt.


Comprehensive Prompt for Building a 30-Day Canadian Partisanship Table

> **Task Overview**  
> Create a markdown table that rates prominent Canadian political commentators on their **partisanship toward each federal party**—Liberal (LPC), Conservative (CPC), New Democratic (NDP), and Bloc Québécois (Bloc)—using only content (tweets, posts, columns, podcasts) from the **last 30 days**.  
> The table must be **sorted by “partisanship spread”** (largest absolute difference between any two party scores) so the strongest partisans appear first and the most neutral at the end.

---

#### 1  Clarify “Partisanship” vs. “Ideology”
* **Ideology** = worldview or policy lens (progressive, libertarian, etc.).  
* **Partisanship** = rooting for or against a party’s fortunes. Count praise/defence of the party as a team, or condemnation/attacks aimed at the party itself (“The Conservatives are dangerous,” “The Liberals deserve re-election”).  
* Ignore neutral poll analysis, inside-baseball strategy chatter, or policy arguments **unless they contain value judgments about a party’s competence, morals, or right to govern**.

---

#### 2  Scoring Scale — Apply Separately to Each Party

| Range | Meaning |
|-------|---------|
| **+9 – +10** | Near single-minded advocate |
| **+7 – +8** | Strong booster; frequent cheer-leading |
| **+4 – +6** | Clear positive lean; routine praise, rare criticism |
| **+1 – +3** | Mild positive lean |
| **  0  ** | No discernible tilt |
| **–1 – –3** | Mild negative lean |
| **–4 – –6** | Clear negative lean; routine criticism |
| **–7 – –8** | Strong hostility; frequent attack lines |
| **–9 – –10** | Near single-minded opposition |

---

#### 3  Add a “Does the Tilt Match the Role?” Column
* **Icon**  
  * ✅ Bias aligns with how the commentator presents themselves (e.g., party strategist is openly partisan, columnist known for one-sided takes).  
  * ⚠️ Bias conflicts with claimed neutrality (e.g., pollster or academic showing strong tilt).
* **One-line explanation** that may include shorthand labels:  
  * **consistent Liberal advocate / Conservative advocate / Conservative critic / Liberal critic**, etc., when a journalist-type’s 30-day output is overwhelmingly one-sided.  
  * Keep phrasing concise; e.g., “✅ Consistent Liberal advocate—columns routinely defend LPC and attack CPC.”

---

#### 4  Name-Harvesting Workflow
1. **Core Keyword Pass** – Search X/Twitter, Bluesky, Mastodon for party-leader names (“Trudeau”, “Poilievre”, “Singh”, “Blanchet”) plus “Canada politics”.  
2. **Topic Pass** – Add major policy terms (“housing crisis”, “carbon tax”, “Bill C-18”, “foreign interference”) to pull niche policy voices.  
3. **Role Seeding** – Automatically include:  
   * Major pollsters/modelers (EKOS, Nanos, Ipsos, Abacus, Léger, 338Canada).  
   * Hosts/anchors of national politics shows (CBC, CTV, CPAC, TVO, Global).  
4. **Network Pass** – Map retweet/quote/reply networks of the initial set; add accounts that interact with **multiple** known commentators.  
5. **Platform Expansion** – Run equivalent searches on Substack, public podcasts (transcript search), blogs—anything visible without login.  
6. **Role Diversity Check** – Ensure mix of mainstream journalists, independent columnists, academics, strategists, activists, podcast hosts. Exclude anonymous meme accounts.  
7. **Duplicate Guard** – Use exact handle/real-name matching to avoid duplicates; if duplicates sneak in, keep the row with complete scores.  
8. **Label Pass** – Scan journalist-class accounts for strongly one-sided behaviour. If a commentator:  
   * publishes ≥ 3 pieces that **defend one party and/or attack its main rival while barely criticising their side**,  
   * OR produces > 70 % of tweets/threads with partisan framing,  
   * tag them “consistent X advocate” or “consistent Y critic” in the explanation sentence.

---

#### 5  Evidence Rules
* **Time window**: only posts dated within the last 30 days (use absolute dates to avoid confusion).  
* **No speculation**—rate only from visible content; if evidence is thin, assign 0 or mild scores.

---

#### 6  Output Requirements
* **Markdown table** sorted by partisanship spread (largest first).  
* **Columns**: `Username | Public role / self-description | LPC | CPC | NDP | Bloc | Does the tilt match the role?`  
* After the table, include the one-sentence reminder:  
  > *Scores run from +10 (ardent booster) to –10 (fierce opponent); 0 = no clear tilt.*  
* Provide **no research notes, citations, or links**—only the finished table and the scale reminder.

---

#### 7  Quality Checklist
- [ ] All four party columns filled.  
- [ ] Duplicate names removed.  
- [ ] Icons present and explanation sentences concise.  
- [ ] Table sorted correctly by spread.  
- [ ] “Consistent advocate/critic” tags applied where warranted.  
- [ ] Final scale reminder included.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment