Dataset Viewer
Auto-converted to Parquet Duplicate
event_id
stringlengths
14
14
frame_id
int32
10k
70.5k
alignment_confidence
float64
0.1
1
alignment_error_seconds
float64
0
1.02
18226500000848
65,088
0.6066
0.591
18226500000849
65,926
0.6
0.5
18226500000850
66,033
0.6107
0.773
18226500000851
66,033
0.6069
0.85
18226500000853
66,222
0.6
0.699
18226500000855
66,287
0.6
0.63
18226500000862
70,275
0.6078
0.703
18226500000863
70,275
0.6095
0.152
18226500000864
70,315
0.6115
0.494
18226500000865
70,386
0.6557
0.94
18226500000866
70,364
0.6306
0.454
18226500000867
70,437
0.6143
0.669
18226500000868
70,500
0.6178
0.368
18226500000869
70,500
0.6118
0.19
18226500000871
70,500
0.6356
0.57
18226500000873
60,460
0.6172
0.45
18226500000159
10,001
0.6402
0.454
18226500000161
10,046
0.6105
0.727
18226500000162
10,126
0.6067
0.495
18226500000168
10,487
0.6196
0.956
18226500000169
10,581
0.6174
0.032
18226500000170
10,706
0.6089
0.623
18226500000171
10,916
0.9532
0.658
18226500000174
10,963
0.6266
0.435
18226500000176
11,072
0.6186
0.887
18226500000177
11,023
0.6118
0.5
18226500000178
11,072
0.6755
0.275
18226500000179
11,142
0.6259
0.151
18226500000180
11,142
0.6122
0.572
18226500000181
11,303
0.6102
0.259
18226500000182
11,277
0.6092
0.875
18226500000183
11,388
0.9573
0.433
18226500000184
11,450
0.7785
0.063
18226500000185
11,481
0.6219
0.4
18226500000186
11,615
0.6072
0.164
18226500000187
11,719
0.6077
1.01
18226500000189
13,001
0.6683
0.07
18226500000190
13,124
0.6435
0.258
18226500000191
13,124
0.6159
0.566
18226500000194
13,251
0.6135
0.871
18226500000195
13,246
0.613
0.587
18226500000196
13,315
0.6103
0.227
18226500000197
13,324
0.6208
0.856
18226500000199
13,449
0.6108
0.023
18226500000202
13,974
0.6085
0.503
18226500000203
14,015
0.6077
0.768
18226500000204
14,047
0.615
0.187
18226500000205
14,168
0.6101
0.449
18226500000206
14,105
0.6089
0.625
18226500000207
14,168
0.6104
0.953
18226500000208
14,259
0.6266
0.178
18226500000211
14,333
0.6076
0.996
18226500000213
14,333
0.6126
0.174
18226500000214
14,333
0.6126
0.36
18226500000215
14,379
0.9956
0.923
18226500000216
14,496
0.6273
0.674
18226500000217
14,514
0.8412
0.733
18226500000218
14,616
0.6113
0.116
18226500000219
14,659
0.61
0.573
18226500000220
14,722
0.6214
0.03
18226500000221
14,806
0.6713
0.317
18226500000222
14,864
0.6119
0.351
18226500000223
14,970
0.7415
0.628
18226500000224
15,075
0.6141
1.001
18226500000225
15,234
0.9049
0.432
18226500000226
15,246
0.6156
0.456
18226500000227
15,424
0.6078
0.721
18226500000228
15,539
0.6256
0.333
18226500000229
15,652
0.6082
0.164
18226500000230
15,695
0.6585
0.173
18226500000232
15,745
0.6109
0.823
18226500000233
15,781
0.6108
0.403
18226500000234
15,800
0.6198
0.421
18226500000235
15,862
0.6465
0.539
18226500000243
16,228
0.6348
0.226
18226500000244
16,302
0.6082
0.904
18226500000245
16,302
0.611
0.306
18226500000246
16,446
0.6094
0.82
18226500000247
16,375
0.6178
0.446
18226500000248
16,446
0.6147
0.439
18226500000249
16,496
0.611
0.34
18226500000250
16,514
0.6071
0.122
18226500000251
16,574
0.609
0.544
18226500000252
13,142
0.6141
0.85
18226500000255
13,164
0.6182
0.21
18226500000258
16,805
0.6186
0.21
18226500000259
16,880
0.619
0.194
18226500000260
16,916
0.7917
0.72
18226500000263
17,289
0.6209
0.749
18226500000264
13,220
0.6215
0.59
18226500000267
17,542
0.6119
0.017
18226500000268
17,597
0.6135
0.223
18226500000270
17,660
0.6096
0.725
18226500000271
13,317
0.6246
1.01
18226500000272
17,768
0.6136
0.068
18226500000273
13,324
0.6105
0.93
18226500000274
17,820
0.6211
0.581
18226500000275
17,910
0.6146
0.106
18226500000276
13,373
0.6084
0.87
18226500000277
17,985
0.609
0.508
End of preview. Expand in Data Studio

OBSO/PAUSA Input Data (IDSSE Events + ELASTIC Sync)

Prerequisite datasets for computing OBSO value surfaces and PAUSA pass timing scores. Contains IDSSE Bundesliga event data and ELASTIC event-tracking synchronization results for 7 matches.

Part of the (Right! Luxury!) Lakehouse soccer analytics platform.

Datasets

Events (data/events/)

DFL event data from 7 IDSSE Bundesliga matches. One row per event with position data. Coordinates are in DFL pitch-origin meters (x: 0-105, y: 0-68).

Column Type Description
match_id string Match identifier (idsse_J03...)
event_id string Unique event identifier within match
event_type string DFL event type (Play, KickOff, TacklingGame, etc.)
timestamp_seconds double Seconds from period start
period int Match half (1 or 2)
player_id string DFL PersonId of acting player
team string Team affiliation (home or away)
x double Event x-coordinate (DFL pitch-origin meters, 0-105)
y double Event y-coordinate (DFL pitch-origin meters, 0-68)

ELASTIC Sync Results (data/elastic_sync/)

Event-to-frame alignments produced by the ELASTIC algorithm (Kim et al. 2025). Maps each event to its best-matching tracking frame via ball acceleration and player-ball proximity features.

Column Type Description
match_id string Match identifier (idsse_J03...)
event_id string Event identifier (joins to events)
frame_id int Best-matching tracking frame number
alignment_confidence double Confidence score (0 to 1)
alignment_error_seconds double Time error between event and aligned frame

Coordinate Systems

  • Events: DFL pitch-origin meters (x: 0-105, y: 0-68). Transform to StatsBomb 120x80: x_sb = x / 105.0 * 120.0, y_sb = y / 68.0 * 80.0.
  • Tracking (separate dataset): Already in StatsBomb 120x80 at luxury-lakehouse/pitch-control-tracking (IDSSE partition).
  • Frame mapping: elastic_sync.frame_id maps to tracking frame column.

Usage with OBSO Pipeline

from datasets import load_dataset

events = load_dataset("luxury-lakehouse/obso-pausa-inputs", "events")["train"]
sync = load_dataset("luxury-lakehouse/obso-pausa-inputs", "elastic_sync")["train"]
tracking = load_dataset("luxury-lakehouse/pitch-control-tracking",
                        data_files="data/source_provider=idsse/**/*.parquet")["train"]

References

  • Bassek et al. (2025). "An integrated dataset of spatiotemporal and event data in elite soccer." Scientific Data, Nature. CC-BY 4.0.
  • Kim, H.S. et al. (2025). "ELASTIC: Event-Tracking Data Synchronization in Soccer Without Annotated Event Locations." ECML-PKDD MLSA 2025. arXiv:2508.09238.

License

MIT — computed from IDSSE open data (CC-BY 4.0).

Downloads last month
130

Models trained or fine-tuned on luxury-lakehouse/obso-pausa-inputs

Space using luxury-lakehouse/obso-pausa-inputs 1

Paper for luxury-lakehouse/obso-pausa-inputs