IV
How reliable is this?
To measure how often the pipeline misses a real eclipsing binary, known ones were run blind through the unmodified pipeline - in an isolated copy that cannot touch the real results - and counted at every stage.
I
The result
A known eclipsing binary counts as recovered if it passes every stage and the pixel check returns ON_TARGET. The period the pipeline finds is not yet compared with the catalogue period.
10 of 60
known eclipsing binaries came out the other end (16.7%)
The 60 were drawn from Kostov et al. 2025, table 3 (newly catalogued eclipsing binaries). Drawn at random, stratified by TESS magnitude quartile so the sample spans the real brightness range. A low number here means the pipeline misses many real binaries; the stages below show where. It does not tell you how many of the site's candidates are false - that needs a different test, such as injecting artificial eclipses with known answers.
II
Where they were lost
Each stage, and how many known binaries it stopped.
Known eclipsing binaries sampled60
Light curve fetched47-13
Vetted (reliable data)42-5
Passed every vetting gate22-20
Novel (no catalog match)21-1
On-target at the pixel check10-11
Fetch outcomes
- NO_DATA
- 8
- OK
- 47
- PROCESSING_ERROR
- 5
Vetting gates tripped
A star can trip more than one gate.
- < 10 in-transit points
- 10
- SNR < 10
- 0
- noisy photometry (>2% above 1.2x baseline)
- 2
- inhomogeneous photometry segments
- 6
- depth < 10 ppm
- 0
- non-finite BLS result
- 0
- < 3 eclipses observed
- 16
- median/BLS depth mismatch
- 6
Pixel-check verdicts
- INCONCLUSIVE
- 9
- OFF_TARGET
- 1
- ON_TARGET
- 10
- UNRESOLVED
- 1
III
Brightness matters
Recovery by TESS magnitude - bigger numbers are fainter stars.
Brighter than 121 of 3 (33.3%)
12 to 132 of 11 (18.2%)
13 to 144 of 23 (17.4%)
Fainter than 143 of 23 (13.0%)
IV
The eclipse-count gate
What the fewer-than-3-eclipses gate is really stopping.
Known binaries stopped by the fewer-than-3-eclipses gate have a median catalogue period of 4.27 days (16 stars), against 3.29 days for the rest (26). One TESS sector covers about 27 days, so a 4.3-day binary would show around 6 eclipses in a single sector. Coverage alone does not explain these losses - and this gate stops more vetted binaries than any other, which makes it the first thing to investigate.
V
Calibration controls
Two calibration stars with known answers are run alongside to validate the pixel check; they are excluded from every count.
- TIC 427332229 - control
- Verdict: ON_TARGET
- TIC 158329671 - control
- Verdict: OFF_TARGET
VI
Every star
Where each known binary stopped, and why.
Show all 60
| TIC | TESS mag | Catalogue period | Stopped at | Outcome |
|---|---|---|---|---|
| 207148552 | 12.92 | 1.990 d | - | Recovered (ON_TARGET) |
| 257736428 | 14.72 | 1.613 d | - | Recovered (ON_TARGET) |
| 233188334 | 12.91 | 2.637 d | - | Recovered (ON_TARGET) |
| 123890398 | 10.01 | 5.026 d | - | Recovered (ON_TARGET) |
| 284729033 | 14.02 | 3.190 d | - | Recovered (ON_TARGET) |
| 34784455 | 14.32 | 1.687 d | - | Recovered (ON_TARGET) |
| 454113205 | 13.26 | 3.888 d | - | Recovered (ON_TARGET) |
| 301278870 | 13.89 | 2.011 d | - | Recovered (ON_TARGET) |
| 263811213 | 13.19 | 2.032 d | - | Recovered (ON_TARGET) |
| 242968405 | 13.14 | 10.232 d | - | Recovered (ON_TARGET) |
| 149696773 | 12.84 | 7.896 d | Vetting | failed a vetting gate< 10 in-transit points; < 3 eclipses observed |
| 17124100 | 12.25 | 9.761 d | Vetting | failed a vetting gate< 10 in-transit points; < 3 eclipses observed |
| 355449770 | 14.24 | 1.472 d | Vetting | failed a vetting gate< 3 eclipses observed |
| 176760623 | 14.70 | 1.219 d | Vetting | failed a vetting gate< 3 eclipses observed |
| 265674312 | 14.51 | 3.568 d | Fetch | NO_DATA |
| 22623237 | 14.38 | 1.634 d | Fetch | NO_DATA |
| 289960699 | 14.88 | 3.293 d | Fetch | NO_DATA |
| 115347208 | 12.43 | 5.252 d | Catalog check | KNOWN_OTHER_TYPE |
| 178717799 | 14.75 | 5.321 d | Vetting | failed a vetting gate< 3 eclipses observed |
| 165171867 | 14.71 | 2.541 d | Fetch | PROCESSING_ERROR |
| 388788212 | 13.92 | 2.438 d | Fetch | NO_DATA |
| 266215601 | 11.47 | 3.019 d | Vetting | failed a vetting gateinhomogeneous photometry segments; < 3 eclipses observed |
| 332534999 | 14.94 | 3.552 d | Vetting | failed a vetting gate< 10 in-transit points; < 3 eclipses observed |
| 116248810 | 13.48 | 6.123 d | Vetting | failed a vetting gateinhomogeneous photometry segments |
| 21509909 | 14.28 | 1.518 d | Vetting | failed a vetting gatenoisy photometry (>2% above 1.2x baseline); median/BLS depth mismatch |
| 233660802 | 12.99 | 4.223 d | Fetch | PROCESSING_ERROR |
| 147515022 | 13.94 | 6.478 d | Vetting | UNRELIABLE_DATA |
| 266136973 | 13.36 | 1.979 d | Vetting | failed a vetting gatenoisy photometry (>2% above 1.2x baseline); inhomogeneous photometry segments; median/BLS depth mismatch |
| 177273977 | 13.92 | 2.232 d | Pixel check | INCONCLUSIVE |
| 274053153 | 14.48 | 1.444 d | Fetch | PROCESSING_ERROR |
| 124103438 | 13.53 | 4.687 d | Vetting | failed a vetting gate< 10 in-transit points; < 3 eclipses observed; median/BLS depth mismatch |
| 280216031 | 14.50 | 2.900 d | Pixel check | INCONCLUSIVE |
| 220129408 | 13.73 | 3.876 d | Vetting | failed a vetting gate< 10 in-transit points; < 3 eclipses observed |
| 470120373 | 13.82 | 2.264 d | Vetting | UNRELIABLE_DATA |
| 401851394 | 12.65 | 17.012 d | Vetting | failed a vetting gate< 3 eclipses observed |
| 130927029 | 13.29 | 3.544 d | Vetting | failed a vetting gate< 10 in-transit points; < 3 eclipses observed; median/BLS depth mismatch |
| 395811975 | 13.37 | 1.304 d | Pixel check | UNRESOLVED |
| 373616761 | 13.40 | 3.396 d | Vetting | failed a vetting gateinhomogeneous photometry segments |
| 359756865 | 14.62 | 13.893 d | Pixel check | INCONCLUSIVE |
| 447428113 | 12.80 | 1.145 d | Pixel check | OFF_TARGET |
| 206699788 | 14.21 | 4.656 d | Pixel check | INCONCLUSIVE |
| 14497682 | 14.72 | 5.324 d | Pixel check | INCONCLUSIVE |
| 262401356 | 13.94 | 4.498 d | Pixel check | INCONCLUSIVE |
| 396605560 | 13.99 | 1.445 d | Fetch | PROCESSING_ERROR |
| 352610727 | 13.76 | 6.180 d | Vetting | UNRELIABLE_DATA |
| 413076401 | 13.01 | 7.209 d | Pixel check | INCONCLUSIVE |
| 262751441 | 12.09 | 1.061 d | Vetting | failed a vetting gate< 10 in-transit points; < 3 eclipses observed |
| 393454264 | 14.64 | 2.055 d | Vetting | failed a vetting gate< 10 in-transit points; < 3 eclipses observed; median/BLS depth mismatch |
| 219174532 | 14.26 | 10.632 d | Vetting | failed a vetting gate< 10 in-transit points; < 3 eclipses observed |
| 188729963 | 13.61 | 1.997 d | Vetting | UNRELIABLE_DATA |
| 273543575 | 14.86 | 8.804 d | Fetch | NO_DATA |
| 301376615 | 11.26 | 7.494 d | Vetting | failed a vetting gateinhomogeneous photometry segments; < 3 eclipses observed |
| 468533471 | 13.70 | 15.482 d | Fetch | NO_DATA |
| 401536004 | 14.58 | 12.367 d | Fetch | NO_DATA |
| 231854767 | 12.25 | 14.013 d | Pixel check | INCONCLUSIVE |
| 266588882 | 14.06 | 1.824 d | Fetch | PROCESSING_ERROR |
| 189884150 | 12.07 | 4.670 d | Vetting | failed a vetting gate< 10 in-transit points; inhomogeneous photometry segments; < 3 eclipses observed; median/BLS depth mismatch |
| 334555521 | 13.40 | 7.478 d | Pixel check | INCONCLUSIVE |
| 18431420 | 14.83 | 1.394 d | Vetting | UNRELIABLE_DATA |
| 428863492 | 13.90 | 2.864 d | Fetch | NO_DATA |