✓ The short version
A bad score is not the same as a mistake. You can bid and play a hand perfectly and still get 20% because the field did something odd, and you can get 80% off a misdefence by the opponents. Sorting those apart is the point — the score tells you what happened, this tells you why.
The board table
Each row is one board you played.
| Column | What it means |
|---|---|
| Board | Board number |
| Dir | Which way you sat |
| Contract | What was played at your table, and by whom |
| Score | Your side's score |
| MP% | What percentage of the field you beat on this board |
| Field | The contract most tables played. = means you were in it too |
| Cause | The verdict — see below |
| Notes | The short reason behind the verdict |
MP% is recalculated from scratch rather than taken from the club's scoresheet, so it means the same thing whatever the source. 50% is exactly average: you scored the same as a typical table.
The six verdicts
The notes column
Short and specific:
par (9)— you took exactly the number of tricks the hand allows on perfect play by everybody.field par (9)— you were in the field's contract and took the same tricks it allows.2 tricks fewer— compared to the average at other tables in the same strain.below DD (10)— the hand was worth 10 tricks on perfect play; you took fewer.defense slip (DD 8)— you took more tricks than perfect defence allows.underbid (4S)/overbid (2S)— you stopped below, or pushed past, the contract the field chose. Only flagged when it cost a game or slam bonus; a partscore difference is not treated as an error.S vs H— you and the field picked different strains.competed too high/competed successfully— a contested auction that did, or did not, work out.
Your history
Tracking a player keeps a running record across sessions. Each row is one event: boards played, average MP%, an Error% (the share of boards attributed to auction, play, or defence), and a count in each verdict column.
The chart plots error rate against matchpoint percentage over time. What to look for is not a single session but the shape: whether the error mix is shifting, and which column is the tall one.
The most useful reading is usually the balance between columns, not the totals. A player with lots of Auction and few Play entries has a bidding problem to work on, and vice versa — a much more actionable finding than an overall percentage.
What this cannot tell you
Being clear about the limits, because they matter for how much weight to put on a verdict:
- It only compares you to the field that played. The comparison is against the other tables in your own event. A weak field makes a good contract look ordinary, and a strong one makes an ordinary result look bad.
- It does not read your auction. With rare exceptions the app sees the final contract, not the bidding that reached it. “Auction” means the contract differed from the field's in a way that cost something — not that a specific bid was wrong.
- It does not watch the play. “Play” means the trick count came up short against the field or against perfect play. It cannot tell you which trick went astray.
- Single-hand replays have no field at all. If you send one hand from a hand viewer, there is nothing to compare it against, so it goes to the double-dummy solver instead.
- Ties and near-misses are blunt. The good/bad line sits at 55% and 45%. A board at 54% is not treated as a success even though it very nearly was.
When a verdict looks wrong to you, it usually is worth trusting your own judgement over the label. The categories are a way of sorting 24 boards into “look at these six” — they are not a ruling.