← Watch feed
Increase probabilistic scorer defaults 5x
Public commit record
90/100 · Strong
What the developer wrote
Authored by benthecarman
Increase probabilistic scorer defaults 5x
Increase the base and historical-liquidity penalties fivefold. This
applies to both the fixed and payment-size parts, so their balance stays
the same. The router will now pay more to avoid longer routes and routes
the scorer considers less likely to work.
I tested this using the graph and scorer pulled from my ldk-server. The
scorer was kept unchanged during the test. Codex randomly assigned 600
different high-degree nodes to the old or new settings, with 300 probes
per setting. The test was balanced across 100 rounds and used amounts of
1,000, 10,000, and 30,000 sats. Only one probe ran at a time.
The old settings reached 167/300 targets (55.7%) within 10 seconds. The
5x settings reached 214/300 (71.3%), an improvement of 15.7 percentage
points. Results were 69% vs 82% at 1,000 sats, 47% vs 72% at 10,000
sats, and 51% vs 60% at 30,000 sats.
Across the 587 targets where both settings found a route, the quoted fee
rose by 2.4 sats on average. These probes used one node and did not
settle, so they do not guarantee payment success elsewhere.
✓ Descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Explains rationale or failure mode✓ Mentions testing or verification
What changed, and why it matters
This commit changes default numerical tuning knobs in the Lightning Dev Kit's payment router. It makes the router prefer shorter, more reliable payment paths by increasing several penalty values fivefold. The change is presented as a performance improvement (better routing success rates in tests), not as a security fix. There is no evidence in the commit of a vulnerability, exploit, or security incident.
Recommended action
No security action required. Treat as a routine routing optimization. Operators who rely on the previous penalty defaults should review whether the new defaults affect their fee or path preferences.
Security signals we found
01
No security framing in commit message or diff
02
No vulnerability pattern (overflow, injection, auth bypass, etc.) present
03
Change is a tunable default value adjustment with documented rationale
04
No advisory, CVE, or incident references in commit or supplied materials
Evidence from the diff
The patch modifies ProbabilisticScoringFeeParameters defaults in lightning/src/routing/scoring.rs. It multiplies base_penalty_msat, base_penalty_amount_multiplier_msat, anti_probing_penalty_msat, historical_liquidity_penalty_multiplier_msat, and historical_liquidity_penalty_amount_multiplier_msat by 5. The commit message frames this as improving routing success probability based on 600-node probe tests. Test expectations are updated to match the new defaults. No code logic, bounds checking, or cryptographic behavior is changed.
Changed components
lightning/src/routing/scoring.rsProbabilisticScoringFeeParameters default valuesProbabilisticScorer routing penalty calculationsInspect captured patch +30 / −27
diff --git a/lightning/src/routing/scoring.rs b/lightning/src/routing/scoring.rs
index c1efde3..c2d3e51 100644
--- a/lightning/src/routing/scoring.rs
+++ b/lightning/src/routing/scoring.rs
@@ -587,7 +587,7 @@ pub struct ProbabilisticScoringFeeParameters {
/// (implying scaling all estimated probabilities down by a factor of ~79%) resulted in the
/// most accurate total success probabilities.
///
- /// Default value: 1,024 msat (i.e. we're willing to pay 1 sat to avoid each additional hop).
+ /// Default value: 5,120 msat (i.e. we're willing to pay 5.12 sats to avoid each additional hop).
///
/// [`historical_liquidity_penalty_multiplier_msat`]: Self::historical_liquidity_penalty_multiplier_msat
pub base_penalty_msat: u64,
@@ -606,8 +606,8 @@ pub struct ProbabilisticScoringFeeParameters {
/// probabilities down by a factor of ~79%) resulted in the most accurate total success
/// probabilities.
///
- /// Default value: 131,072 msat (i.e. we're willing to pay 0.125bps to avoid each additional
- /// hop).
+ /// Default value: 655,360 msat (i.e. we're willing to pay roughly 6.1 basis points to avoid
+ /// each additional hop).
///
/// [`base_penalty_msat`]: Self::base_penalty_msat
/// [`historical_liquidity_penalty_amount_multiplier_msat`]: Self::historical_liquidity_penalty_amount_multiplier_msat
@@ -673,8 +673,8 @@ pub struct ProbabilisticScoringFeeParameters {
/// track which of several buckets those bounds fall into, exponentially decaying the
/// probability of each bucket as new samples are added.
///
- /// Default value: 10,000 msat (i.e. willing to pay 1 sat to avoid an 80% probability channel,
- /// or 6 sats to avoid a 25% probability channel).
+ /// Default value: 50,000 msat (i.e. willing to pay 5 sats to avoid an 80% probability channel,
+ /// or 30 sats to avoid a 25% probability channel).
///
/// [`liquidity_penalty_multiplier_msat`]: Self::liquidity_penalty_multiplier_msat
pub historical_liquidity_penalty_multiplier_msat: u64,
@@ -695,8 +695,8 @@ pub struct ProbabilisticScoringFeeParameters {
/// channel, we track which of several buckets those bounds fall into, exponentially decaying
/// the probability of each bucket as new samples are added.
///
- /// Default value: 1,250 msat (i.e. willing to pay about 0.125 bps per hop to avoid 78%
- /// probability channels, or 0.5bps to avoid a 38% probability
+ /// Default value: 6,250 msat (i.e. willing to pay about 6.4 bps per hop to avoid 78%
+ /// probability channels, or 25bps to avoid a 38% probability
/// channel).
///
/// [`liquidity_penalty_amount_multiplier_msat`]: Self::liquidity_penalty_amount_multiplier_msat
@@ -715,7 +715,7 @@ pub struct ProbabilisticScoringFeeParameters {
/// as this makes balance discovery attacks harder to execute, thereby creating an incentive
/// to restrict `htlc_maximum_msat` and improve privacy.
///
- /// Default value: 250 msat
+ /// Default value: 1,250 msat
pub anti_probing_penalty_msat: u64,
/// This penalty is applied when the total amount flowing over a channel exceeds our current
@@ -787,15 +787,15 @@ pub struct ProbabilisticScoringFeeParameters {
impl Default for ProbabilisticScoringFeeParameters {
fn default() -> Self {
Self {
- base_penalty_msat: 1024,
- base_penalty_amount_multiplier_msat: 131_072,
+ base_penalty_msat: 5_120,
+ base_penalty_amount_multiplier_msat: 655_360,
liquidity_penalty_multiplier_msat: 0,
liquidity_penalty_amount_multiplier_msat: 0,
manual_node_penalties: new_hash_map(),
- anti_probing_penalty_msat: 250,
+ anti_probing_penalty_msat: 1_250,
considered_impossible_penalty_msat: 1_0000_0000_000,
- historical_liquidity_penalty_multiplier_msat: 10_000,
- historical_liquidity_penalty_amount_multiplier_msat: 1_250,
+ historical_liquidity_penalty_multiplier_msat: 50_000,
+ historical_liquidity_penalty_amount_multiplier_msat: 6_250,
linear_success_probability: false,
probing_diversity_penalty_msat: 0,
}
@@ -3630,47 +3630,47 @@ mod tests {
info,
short_channel_id: 42,
});
- assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 42_252);
+ assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 211_262);
let usage = ChannelUsage {
effective_capacity: EffectiveCapacity::Total { capacity_msat: 1_950_000_000, htlc_maximum_msat: 1_000 }, ..usage
};
- assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 36_005);
+ assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 180_032);
let usage = ChannelUsage {
effective_capacity: EffectiveCapacity::Total { capacity_msat: 2_950_000_000, htlc_maximum_msat: 1_000 }, ..usage
};
- assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 32_851);
+ assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 164_259);
let usage = ChannelUsage {
effective_capacity: EffectiveCapacity::Total { capacity_msat: 3_950_000_000, htlc_maximum_msat: 1_000 }, ..usage
};
- assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 30_832);
+ assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 154_165);
let usage = ChannelUsage {
effective_capacity: EffectiveCapacity::Total { capacity_msat: 4_950_000_000, htlc_maximum_msat: 1_000 }, ..usage
};
- assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 29_886);
+ assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 149_434);
let usage = ChannelUsage {
effective_capacity: EffectiveCapacity::Total { capacity_msat: 5_950_000_000, htlc_maximum_msat: 1_000 }, ..usage
};
- assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 28_939);
+ assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 144_702);
let usage = ChannelUsage {
effective_capacity: EffectiveCapacity::Total { capacity_msat: 6_950_000_000, htlc_maximum_msat: 1_000 }, ..usage
};
- assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 28_435);
+ assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 142_178);
let usage = ChannelUsage {
effective_capacity: EffectiveCapacity::Total { capacity_msat: 7_450_000_000, htlc_maximum_msat: 1_000 }, ..usage
};
- assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 27_993);
+ assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 139_969);
let usage = ChannelUsage {
effective_capacity: EffectiveCapacity::Total { capacity_msat: 7_950_000_000, htlc_maximum_msat: 1_000 }, ..usage
};
- assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 27_993);
+ assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 139_969);
let usage = ChannelUsage {
effective_capacity: EffectiveCapacity::Total { capacity_msat: 8_950_000_000, htlc_maximum_msat: 1_000 }, ..usage
};
- assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 27_488);
+ assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 137_446);
let usage = ChannelUsage {
effective_capacity: EffectiveCapacity::Total { capacity_msat: 9_950_000_000, htlc_maximum_msat: 1_000 }, ..usage
};
- assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 27_047);
+ assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 135_238);
}
#[test]
@@ -4005,8 +4005,11 @@ mod tests {
let logger = TestLogger::new();
let network_graph = network_graph(&logger);
let source = source_node_id();
+ let anti_probing_penalty_msat =
+ ProbabilisticScoringFeeParameters::default().anti_probing_penalty_msat;
+ assert_eq!(anti_probing_penalty_msat, 1_250);
let params = ProbabilisticScoringFeeParameters {
- anti_probing_penalty_msat: 500,
+ anti_probing_penalty_msat,
..ProbabilisticScoringFeeParameters::zero_penalty()
};
let scorer = ProbabilisticScorer::new(ProbabilisticScoringDecayParameters::default(), &network_graph, &logger);
@@ -4032,7 +4035,7 @@ mod tests {
inflight_htlc_msat: 0,
effective_capacity: EffectiveCapacity::Total { capacity_msat: 1_024_000, htlc_maximum_msat: 1_024_000 },
};
- assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 500);
+ assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 1_250);
// Check we receive anti-probing penalty for htlc_maximum_msat == channel_capacity/2.
let usage = ChannelUsage {
@@ -4040,7 +4043,7 @@ mod tests {
inflight_htlc_msat: 0,
effective_capacity: EffectiveCapacity::Total { capacity_msat: 1_024_000, htlc_maximum_msat: 512_000 },
};
- assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 500);
+ assert_eq!(scorer.channel_penalty_msat(&candidate, usage, ¶ms), 1_250);
// Check we receive no anti-probing penalty for htlc_maximum_msat == channel_capacity/2 - 1.
let usage = ChannelUsage {
Risk score
Our methodology →Why this scored 21/100
Human-validated context
Community notes
Notes can correct, qualify, or add evidence to the AI analysis. Every note shown here has been validated by a human moderator.
No validated notes yet.
The AI analysis stands alone for now. Submit a note if you can add evidence or important context.