Understand the failure.
Verify the repair.
Small, reproducible software failures. The broken implementation, the fix that didn’t work, and the one that passed—preserved together.
Explore the cases ↓How results are verified ↗WHAT THE ARCHIVE CONTAINS
100840 executable cases. 20168 are open.
Every case records the implementation that fails, the fix that did not work, and the repair that passed its checks—with recorded outputs and source hashes. This release adds 100840 cases across 20168 failure mechanisms and 254 domains.
The open tier gives you the failure and the unsuccessful fix for one case in every mechanism. The remaining 80672 cases, 5 variants per mechanism, are member-only: the verified repair, its recorded checks, and the full fixture suite are held in the member archive. Read the methodology ↗
A RECORD OF WHAT WENT WRONG
Browse the archive / 100840
Points computed from the uncorrected stat line · case 01
The stats table shows the corrected value but the points total still reflects the original.
Points computed from the uncorrected stat line · case 02
The stats table shows the corrected value but the points total still reflects the original.
Points computed from the uncorrected stat line · case 03
The stats table shows the corrected value but the points total still reflects the original.
Points computed from the uncorrected stat line · case 04
The stats table shows the corrected value but the points total still reflects the original.
Points computed from the uncorrected stat line · case 05
The stats table shows the corrected value but the points total still reflects the original.
Upper middle score used as the median in even leagues · case 01
In an eight-team league the fourth-best team is marked below the median every week.
Upper middle score used as the median in even leagues · case 02
In an eight-team league the fourth-best team is marked below the median every week.
Upper middle score used as the median in even leagues · case 03
In an eight-team league the fourth-best team is marked below the median every week.
Upper middle score used as the median in even leagues · case 04
In an eight-team league the fourth-best team is marked below the median every week.
Upper middle score used as the median in even leagues · case 05
In an eight-team league the fourth-best team is marked below the median every week.
Median ending in .005 truncated before comparison · case 01
A team exactly 0.005 below a half-hundredth median is recorded as a median tie.
Median ending in .005 truncated before comparison · case 02
A team exactly 0.005 below a half-hundredth median is recorded as a median tie.
Median ending in .005 truncated before comparison · case 03
A team exactly 0.005 below a half-hundredth median is recorded as a median tie.
Median ending in .005 truncated before comparison · case 04
A team exactly 0.005 below a half-hundredth median is recorded as a median tie.
Median ending in .005 truncated before comparison · case 05
A team exactly 0.005 below a half-hundredth median is recorded as a median tie.
Median taken from unsorted score order · case 01
The median changes when teams are listed in a different order.
Median taken from unsorted score order · case 02
The median changes when teams are listed in a different order.
Median taken from unsorted score order · case 03
The median changes when teams are listed in a different order.
Median taken from unsorted score order · case 04
The median changes when teams are listed in a different order.
Median taken from unsorted score order · case 05
The median changes when teams are listed in a different order.
Scoring exactly the median recorded as a loss · case 01
In an odd league the median team always takes a median loss.
Scoring exactly the median recorded as a loss · case 02
In an odd league the median team always takes a median loss.
Scoring exactly the median recorded as a loss · case 03
In an odd league the median team always takes a median loss.
Scoring exactly the median recorded as a loss · case 04
In an odd league the median team always takes a median loss.
Scoring exactly the median recorded as a loss · case 05
In an odd league the median team always takes a median loss.
Equal matchup scores awarded to the listed home team · case 01
Two teams tied at 90.25 produce a win and a loss.
Equal matchup scores awarded to the listed home team · case 02
Two teams tied at 90.25 produce a win and a loss.
Equal matchup scores awarded to the listed home team · case 03
Two teams tied at 90.25 produce a win and a loss.
Equal matchup scores awarded to the listed home team · case 04
Two teams tied at 90.25 produce a win and a loss.
Equal matchup scores awarded to the listed home team · case 05
Two teams tied at 90.25 produce a win and a loss.
Every tied entry paid the highest shared prize · case 01
Two entries tied for first each receive the first-place prize.
Every tied entry paid the highest shared prize · case 02
Two entries tied for first each receive the first-place prize.
Every tied entry paid the highest shared prize · case 03
Two entries tied for first each receive the first-place prize.
Every tied entry paid the highest shared prize · case 04
Two entries tied for first each receive the first-place prize.
Every tied entry paid the highest shared prize · case 05
Two entries tied for first each receive the first-place prize.
Leftover cents from a tie split disappear · case 01
Three entries sharing 1000 cents receive 333 each and one cent is never paid.
Leftover cents from a tie split disappear · case 02
Three entries sharing 1000 cents receive 333 each and one cent is never paid.
Leftover cents from a tie split disappear · case 03
Three entries sharing 1000 cents receive 333 each and one cent is never paid.
Leftover cents from a tie split disappear · case 04
Three entries sharing 1000 cents receive 333 each and one cent is never paid.
Leftover cents from a tie split disappear · case 05
Three entries sharing 1000 cents receive 333 each and one cent is never paid.
Ties decided on the displayed tenths · case 01
Entries scoring 149.95 and 149.93 are paid as if tied.
Ties decided on the displayed tenths · case 02
Entries scoring 149.95 and 149.93 are paid as if tied.
Ties decided on the displayed tenths · case 03
Entries scoring 149.95 and 149.93 are paid as if tied.
Ties decided on the displayed tenths · case 04
Entries scoring 149.95 and 149.93 are paid as if tied.
Ties decided on the displayed tenths · case 05
Entries scoring 149.95 and 149.93 are paid as if tied.
Zero-cent entries listed as winners · case 01
Every non-cashing entry appears in the payout report with 0 cents.
Zero-cent entries listed as winners · case 02
Every non-cashing entry appears in the payout report with 0 cents.
Zero-cent entries listed as winners · case 03
Every non-cashing entry appears in the payout report with 0 cents.
Zero-cent entries listed as winners · case 04
Every non-cashing entry appears in the payout report with 0 cents.
Zero-cent entries listed as winners · case 05
Every non-cashing entry appears in the payout report with 0 cents.
Captain salary not rounded up to the salary grid · case 01
A 5300 captain costs 7950 instead of 8000, letting over-cap lineups through.
Captain salary not rounded up to the salary grid · case 02
A 5300 captain costs 7950 instead of 8000, letting over-cap lineups through.
Captain salary not rounded up to the salary grid · case 03
A 5300 captain costs 7950 instead of 8000, letting over-cap lineups through.
Captain salary not rounded up to the salary grid · case 04
A 5300 captain costs 7950 instead of 8000, letting over-cap lineups through.
Captain salary not rounded up to the salary grid · case 05
A 5300 captain costs 7950 instead of 8000, letting over-cap lineups through.
Captain half-hundredths floored · case 01
A captain with 13.33 points scores 19.99 instead of 20.00; negative captains lose extra.
Captain half-hundredths floored · case 02
A captain with 13.33 points scores 19.99 instead of 20.00; negative captains lose extra.
Captain half-hundredths floored · case 03
A captain with 13.33 points scores 19.99 instead of 20.00; negative captains lose extra.
Captain half-hundredths floored · case 04
A captain with 13.33 points scores 19.99 instead of 20.00; negative captains lose extra.
Captain half-hundredths floored · case 05
A captain with 13.33 points scores 19.99 instead of 20.00; negative captains lose extra.
Lineup spending exactly the cap rejected · case 01
A lineup using every dollar of the cap is refused as over budget.
Lineup spending exactly the cap rejected · case 02
A lineup using every dollar of the cap is refused as over budget.
Lineup spending exactly the cap rejected · case 03
A lineup using every dollar of the cap is refused as over budget.
Lineup spending exactly the cap rejected · case 04
A lineup using every dollar of the cap is refused as over budget.
Lineup spending exactly the cap rejected · case 05
A lineup using every dollar of the cap is refused as over budget.
Captain also rostered as a flex player · case 01
A lineup lists the same quarterback as captain and flex and earns 2.5x his points.
Captain also rostered as a flex player · case 02
A lineup lists the same quarterback as captain and flex and earns 2.5x his points.
Captain also rostered as a flex player · case 03
A lineup lists the same quarterback as captain and flex and earns 2.5x his points.
Captain also rostered as a flex player · case 04
A lineup lists the same quarterback as captain and flex and earns 2.5x his points.
Captain also rostered as a flex player · case 05
A lineup lists the same quarterback as captain and flex and earns 2.5x his points.
Captain team ignored in the two-team rule · case 01
Five flex players from one team plus an opposing captain are rejected.
Captain team ignored in the two-team rule · case 02
Five flex players from one team plus an opposing captain are rejected.
Captain team ignored in the two-team rule · case 03
Five flex players from one team plus an opposing captain are rejected.
Captain team ignored in the two-team rule · case 04
Five flex players from one team plus an opposing captain are rejected.
Captain team ignored in the two-team rule · case 05
Five flex players from one team plus an opposing captain are rejected.
Innings pitched parsed as a decimal number · case 01
A 6.2 inning start is credited with 19 outs instead of 20.
Innings pitched parsed as a decimal number · case 02
A 6.2 inning start is credited with 19 outs instead of 20.
Innings pitched parsed as a decimal number · case 03
A 6.2 inning start is credited with 19 outs instead of 20.
Innings pitched parsed as a decimal number · case 04
A 6.2 inning start is credited with 19 outs instead of 20.
Innings pitched parsed as a decimal number · case 05
A 6.2 inning start is credited with 19 outs instead of 20.
Exactly six innings denied a quality start · case 01
A 6.0 inning, 3 earned run start earns no quality-start bonus.
Exactly six innings denied a quality start · case 02
A 6.0 inning, 3 earned run start earns no quality-start bonus.
Exactly six innings denied a quality start · case 03
A 6.0 inning, 3 earned run start earns no quality-start bonus.
Exactly six innings denied a quality start · case 04
A 6.0 inning, 3 earned run start earns no quality-start bonus.
Exactly six innings denied a quality start · case 05
A 6.0 inning, 3 earned run start earns no quality-start bonus.
Hold stacked on a save · case 01
A reliever credited with both a hold flag and a save earns 8 points.
Hold stacked on a save · case 02
A reliever credited with both a hold flag and a save earns 8 points.
Hold stacked on a save · case 03
A reliever credited with both a hold flag and a save earns 8 points.
Hold stacked on a save · case 04
A reliever credited with both a hold flag and a save earns 8 points.
Hold stacked on a save · case 05
A reliever credited with both a hold flag and a save earns 8 points.
ERA truncated to whole runs before ranking · case 01
Teams with ERAs of 3.10 and 3.90 are ranked as tied.
ERA truncated to whole runs before ranking · case 02
Teams with ERAs of 3.10 and 3.90 are ranked as tied.
ERA truncated to whole runs before ranking · case 03
Teams with ERAs of 3.10 and 3.90 are ranked as tied.
ERA truncated to whole runs before ranking · case 04
Teams with ERAs of 3.10 and 3.90 are ranked as tied.
ERA truncated to whole runs before ranking · case 05
Teams with ERAs of 3.10 and 3.90 are ranked as tied.
WHIP ranked high-to-low · case 01
The staff allowing the most baserunners collects the most WHIP points.
WHIP ranked high-to-low · case 02
The staff allowing the most baserunners collects the most WHIP points.
WHIP ranked high-to-low · case 03
The staff allowing the most baserunners collects the most WHIP points.
WHIP ranked high-to-low · case 04
The staff allowing the most baserunners collects the most WHIP points.
WHIP ranked high-to-low · case 05
The staff allowing the most baserunners collects the most WHIP points.
INSPECTABLE BY DESIGN
Every result has a runnable source.
Runnable implementations with recorded outputs, source hashes, and explicit contracts. Related variants share a failure mechanism and belong together in evaluation splits.
Read the methodology ↗