OFFLINE REPORT VIEWER
Brass Band Rankings
A statistically defensible set of international brass band rankings.
RANKING RUN
dynamic-bayesian-plackett-luce as of 2026-11-13
How to read this: annual PL estimates one strength for each band-year. Higher values mean a greater predicted chance of finishing ahead of another band in the same observed field. After a band’s final observed year, its as-of strength decays toward the population mean according to the configured half-life. Uncertainty is a diagonal Laplace approximation around the MAP estimate; it also propagates through inactive years, but does not yet represent posterior correlations between bands and years.
| Rank | Band | Mean strength | Approx. standard deviation | Conservative strength | Rank change |
|---|
ANNUAL STRENGTH HISTORY
Fitted strength history
The chart follows the ten highest-ranked bands at the selected year. Hover over a point to see the band, observed year, and fitted strength. Points exist only in years with an included result; lines connect consecutive observed states and do not mean a band competed every intervening year. Historical points are retrospective estimates and may be informed by later results through the ranking snapshot’s as-of date. The separate uncertainty chart shows approximate marginal distributions; it omits correlations between band-year states.
Approximate uncertainty at the selected year
The box plot shows the ten highest mean-strength bands. The centre line is the estimated mean, the box spans the middle half of the approximate marginal distribution, and whiskers span an approximate 95% interval. These are not joint uncertainty intervals.
ROLLING OUT-OF-SAMPLE BACKTEST
How do the fitted models predict later contests?
Each score comes from a contest that was excluded from that fold’s training fit. Higher log likelihood is better. The comparison is paired: both models score the same contest occurrence.
static-plackett-luce
-10.244mean full-order log likelihood1941 held-out contests · winner log loss 1.702annual-dynamic-plackett-luce
-11.784mean full-order log likelihood1941 held-out contests · winner log loss 1.847annual-mean-reverting-plackett-luce
-10.487mean full-order log likelihood1941 held-out contests · winner log loss 1.673Static PL advantage
+0.472mean log-likelihood difference95% contest bootstrap interval 0.42 to 0.527Dynamic PL advantage
-1.541over static PL95% contest bootstrap interval -1.716 to -1.393A bootstrap interval above zero supports the named first model for this retrospective set. It resamples whole contests, so a large field does not count as many independent contests.
PAIRWISE CALIBRATION
Do stated pairwise probabilities match observed results?
Calibration shown for annual-dynamic-plackett-luce.
For each strict pair of bands, the model predicts a probability that a fixed band-ID orientation finishes ahead. A calibrated model’s average predicted probability should be close to the observed win rate. Tied pairs are omitted; pairs from the same contest are shown for diagnostics, not treated as independent evidence.
| Predicted range | Pairs | Mean prediction | Observed win rate | Volume |
|---|---|---|---|---|
| 0.0-0.1 | 9909 | 4.4 % | 22.2 % | |
| 0.1-0.2 | 6091 | 14.7 % | 34.0 % | |
| 0.2-0.3 | 4984 | 25.0 % | 40.3 % | |
| 0.3-0.4 | 4805 | 34.9 % | 42.8 % | |
| 0.4-0.5 | 5304 | 45.0 % | 47.0 % | |
| 0.5-0.6 | 8759 | 53.1 % | 53.0 % | |
| 0.6-0.7 | 4730 | 64.9 % | 56.3 % | |
| 0.7-0.8 | 5378 | 75.0 % | 60.8 % | |
| 0.8-0.9 | 6337 | 85.3 % | 67.1 % | |
| 0.9-1.0 | 11555 | 95.8 % | 79.3 % |
WHERE THE MODEL HELPS
Performance slices
These break the same held-out scoring down by reviewed contest geography, observed field size, and whether a field contains a band absent from its training fold. Sparse buckets need cautious interpretation; use the interval and occurrence count together.
| Dimension | Bucket | Contests | Static − equal | Dynamic − static | 95% interval (dynamic) |
|---|---|---|---|---|---|
| contest-geography | AU | 629 | +0.088 | -1.043 | -1.264 to -0.811 |
| contest-geography | BE | 39 | +1.177 | +0.167 | -0.292 to 0.64 |
| contest-geography | CH | 80 | +1.38 | -0.284 | -0.821 to 0.212 |
| contest-geography | DE | 7 | +0.165 | +0.607 | 0.282 to 0.896 |
| contest-geography | GB | 463 | +0.836 | -2.571 | -3.034 to -2.139 |
| contest-geography | GB-SCT | 11 | +0.368 | -0.574 | -1.961 to 0.787 |
| contest-geography | GB-WLS | 80 | +0.579 | -0.986 | -1.583 to -0.438 |
| contest-geography | NL | 52 | +0.349 | -1.339 | -2.033 to -0.729 |
| contest-geography | NO | 64 | +1.605 | -0.483 | -1.007 to 0.048 |
| contest-geography | NZ | 347 | +0.03 | -2.294 | -2.731 to -1.855 |
| contest-geography | SE | 16 | +0.567 | +0.188 | -0.113 to 0.519 |
| contest-geography | ZZ | 67 | +0.192 | -1.5 | -2.255 to -0.892 |
| contest-geography | international | 86 | +1.319 | -0.611 | -1.077 to -0.169 |
| field-size | 11+ | 372 | +1.234 | -3.231 | -3.785 to -2.657 |
| field-size | 2 | 124 | +0.007 | -0.078 | -0.195 to 0.028 |
| field-size | 3-5 | 612 | +0.068 | -0.67 | -0.824 to -0.524 |
| field-size | 6-10 | 833 | +0.499 | -1.643 | -1.888 to -1.423 |
| new-band-status | contains-new-band | 1046 | +0.464 | -1.717 | -1.963 to -1.475 |
| new-band-status | known-bands-only | 895 | +0.482 | -1.334 | -1.547 to -1.128 |
DATA COVERAGE
What entered the model?
1991 eligible partial rankings from 2545 imported occurrences, containing 15025 ranked rows. An eligible observation is an ordering of its reported subset, not a complete contest field.
| Contest geography | Eligible occurrences |
|---|---|
| AU | 629 |
| GB | 503 |
| NZ | 357 |
| international | 86 |
| GB-WLS | 80 |
| CH | 80 |
| ZZ | 67 |
| NO | 64 |
| NL | 52 |
| BE | 39 |
| SE | 16 |
| GB-SCT | 11 |
| DE | 7 |