Engagement Weight Calculator

What Are Engagement Weights?

Not all engagement is equal. When you like, reply, or retweet a tweet, the algorithm assigns each action a different value. These values—called engagement weights—determine how the algorithm scores every tweet in your feed.

How it works: The Heavy Ranker (Twitter's scoring model) predicts 15 different types of engagement for every tweet. Each prediction gets multiplied by its weight, and the sum becomes the tweet's score. Higher score = higher ranking in your feed.

Why weights matter: These weights reveal what Twitter optimizes for. They're not just scoring mechanisms—they shape:

• What content gets amplified - High-weighted engagement (replies: 13.5) drives visibility more than low-weighted engagement (likes: 0.5)
• Your algorithmic identity - Your engagement history (weighted by these values) trains your cluster assignment weekly
• Creator incentives - Creators optimize for high-weighted engagement, not just high volume
• The feedback loop - High-weighted engagement → stronger cluster assignment → see more similar content → more high-weighted engagement

The Core Insight: Conversation Over Consumption

Twitter massively prioritizes conversation (replies, especially with author engagement) over passive consumption (likes, video views):

Conversation (active):
- Reply with author engagement: 75.0 (most valuable)
- Reply: 13.5
- Good profile click: 12.0

Passive consumption:
- Favorite (like): 0.5
- Retweet: 1.0
- Video playback 50%: 0.005 (nearly worthless)

The math: One reply with author engagement (75.0) equals 150 likes (75.0 ÷ 0.5). This isn't a bug—it's the business model.

Why? Conversation drives time on platform:

• Reading replies takes time
• Writing replies takes time
• Waiting for responses keeps users checking back
• Back-and-forth conversations = more ad impressions

The Shape of the Weights

High-value actions (10.0 to 75.0): Deep engagement requiring effort—replies, profile exploration, meaningful clicks. These signal genuine interest.

Low-value actions (0.5 to 1.0): Passive engagement—likes, retweets, basic viewing. Easy to give, minimal signal.

Negative actions (-74.0 to -369.0): Explicit dislike or policy violations. Catastrophic penalties that last weeks to months.

Experience How Engagement Weights Work

Use this calculator to see how different engagement patterns score. Notice how high-weighted engagement (conversation) massively outweighs passive engagement (likes) even when volume is much lower.

The 15 Engagement Weights

These weight values come from X's ML training repository. The algorithm multiplies predicted probabilities by these weights.

Tweet Score Calculator

Compare how different tweet types score. Each scenario represents realistic engagement probabilities. Notice how conversation-driven tweets can outscore viral content.

How Weights Train Your Feed

The Technical Details

The Scoring Formula

Every tweet's base score is calculated as a weighted sum of predicted engagement probabilities:

tweet_score = Σ(P(engagement_i) × weight_i) + epsilon

where:
- P(engagement_i) = Heavy Ranker's predicted probability (0.0 to 1.0)
- weight_i = configured weight from the table above
- epsilon = 0.001 (small constant to avoid zero scores)
- i ranges over all 15 engagement types

Concrete Example

Let's score a tweet with realistic predictions:

Heavy Ranker predictions:
- P(favorite) = 0.20 (20% chance user will like)
- P(reply) = 0.05 (5% chance user will reply)
- P(reply_with_author_engagement) = 0.01 (1% chance of reply-back)
- P(retweet) = 0.10 (10% chance of retweet)
- P(good_click) = 0.15 (15% chance of meaningful click)
- P(negative_feedback) = 0.02 (2% chance of "not interested")

Weighted contributions (using March 2023 weights):
- Favorite: 0.20 × 0.5 = 0.10
- Reply: 0.05 × 13.5 = 0.675
- Reply w/ author: 0.01 × 75.0 = 0.75
- Retweet: 0.10 × 1.0 = 0.10
- Good click: 0.15 × 11.0 = 1.65
- Negative: 0.02 × -74.0 = -1.48
- Epsilon: 0.001

Total score = 0.10 + 0.675 + 0.75 + 0.10 + 1.65 - 1.48 + 0.001 = 1.796

Key observation: The 1% reply-with-author-engagement probability (0.75 contribution) contributes more than the 20% favorite probability (0.10 contribution). This is by design.

Why These Specific Weights?

From Twitter's official documentation:

"Each engagement has a different average probability, the weights were originally set so that, on average, each weighted engagement probability contributes a near-equal amount to the score. Since then, we have periodically adjusted the weights to optimize for platform metrics."

Translation:

• Initial design: Normalize by rarity (rare actions get high weights, common actions get low weights)
• Current state: Tuned to optimize Twitter's business goals—time on platform, conversation depth, user retention

Configurability: All Weights Are Tunable

Every weight is defined as an FSBoundedParam, meaning X can adjust them without deploying new code:

// From HomeGlobalParams.scala:788-930
object ReplyEngagedByAuthorParam extends FSBoundedParam[Double](
  name = "home_mixer_model_weight_reply_engaged_by_author",
  default = 0.0,        // Not actual value - just placeholder
  min = -10000.0,       // Can be negative (penalty)
  max = 10000.0         // Can be very positive (amplification)
)

Where the actual weight values come from:

• The open-source algorithm repo shows parameter structure (default = 0.0 is a placeholder)
• X's ML training repo (the-algorithm-ml/projects/home/recap) documents the actual production configuration
• The values we show (0.5, 13.5, 75.0, etc.) come from that ML repo
• These represent X's documented weights, but current production may differ

What this means:

• X can A/B test different weight configurations
• Weights can be adjusted per-user or per-region
• Production values may differ from the documented ML repo values
• X can pivot platform priorities by adjusting weights (more video, less conversation, etc.)

The Complete Weight Table (March 2023)

Note: Bookmark, Share, Dwell, Video Quality View, and Video Extended weights are configurable but not disclosed in March 2023 snapshot

The Nuclear Penalties

Negative Feedback V2: -74.0

Equivalent to -148 favorites of negative value

Triggered by: "Not interested" click, "See less often" click, muting author

Duration: 140-day linear decay

Day 0: 0.2x multiplier (80% penalty - nearly invisible)
Day 70: 0.6x multiplier (40% penalty - recovering)
Day 140: 1.0x multiplier (penalty expires)

Impact: A single "not interested" click suppresses content from that author for 5 months.

Report: -369.0

Equivalent to -738 favorites of negative value

Triggered by: Explicit report for spam, harassment, misinformation, or other policy violations

To overcome ONE report, a tweet would need:

738 favorites (at 0.5 weight each), OR
28 replies (at 13.5 weight each), OR
5 reply-with-author-engagements (at 75.0 weight each)

In practice: Impossible to overcome. Reports are a platform-level harm signal. Content that gets reported is effectively dead.

Code References

What The Algorithm Doesn't Know

The Heavy Ranker predicts engagement probability, not:

• Truth: No fact-checking in the scoring model
• Quality: Only engagement likelihood, not content value
• Intent: Can't distinguish productive debate from toxic conflict
• Context: Replies could be agreement or furious disagreement—same weight either way

The algorithm optimizes for engagement, not for truth, quality, or societal value. This is a business decision, not a technical limitation.

Key Implications