Docs

All docs

Concepts

Guides

Draft team rows

  • Concept
  • Engine v0.18.0
  • Facts schema 0.1.0 and later
  • commishdesk.com/docs/draft-team-rows/

Definition

Each team gets one row that joins three results: the board (what the team drafted), the consensus comparison and the grade (see draft-grade).

  • pick_count and pick_nos are the number of picks and their overall pick numbers, in order;
  • positional_counts counts picks by position, always listing QB, RB, WR and TE in that order with zeros filled in, followed by any other position in sorted order;
  • back_to_back lists each pair of consecutive overall picks the team made;
  • best_value_pick is the team's pick with the largest delta, and biggest_reach_pick the one with the smallest, each as pick number, board label, player and delta.

See Draft team rows in a full league

Supported: Sleeper · See supported apps

See a full league while we open sign-ups.

Worked example

Fixture expected-draft-recap-facts.json

Fixture: tests/fixtures/facts/expected-draft-recap-facts.json (the anonymized league "Trench Warfare").

{
  "teams": [
    {
      "roster_id": "2",
      "manager": "Pull-Guard Pumas",
      "pick_count": 12,
      "pick_nos": [
        1,
        30,
        37,
        42,
        43,
        44,
        45,
        46,
        47,
        54,
        55,
        66
      ],
      "positional_counts": {
        "QB": 1,
        "RB": 4,
        "WR": 6,
        "TE": 1
      },
      "back_to_back": [
        [
          42,
          43
        ],
        [
          43,
          44
        ],
        [
          44,
          45
        ],
        [
          45,
          46
        ],
        [
          46,
          47
        ],
        [
          54,
          55
        ]
      ],
      "best_value_pick": {
        "pick_no": 54,
        "board_label": "5.06",
        "player": "Tai Felton",
        "delta": 10
      },
      "biggest_reach_pick": {
        "pick_no": 1,
        "board_label": "1.01",
        "player": "Ashton Jeanty",
        "delta": -3
      }
    }
  ]
}

Pull-Guard Pumas (roster 2) made 12 picks: 1, 30, 37, 42, 43, 44, 45, 46, 47, 54, 55 and 66. Their counts are 1 QB, 4 RB, 6 WR and 1 TE. They picked at 42 through 47 in a row, giving five adjacent pairs (42 and 43, 43 and 44, 44 and 45, 45 and 46, 46 and 47), plus 54 and 55. Their best value was pick 54 (5.06, Tai Felton, +10) and their biggest reach was pick 1 (1.01, Ashton Jeanty, -3).

See Draft team rows in a full league

Put in your Sleeper username or league ID, or look around a fictional league first.

Supported: Sleeper · See supported apps

See a full league while we open sign-ups.

Code

Show the code commishdesk/facts/build.py::_merge_teams

Source: commishdesk/facts/build.py::_merge_teams

def _merge_teams(
    league: LeagueModel,
    board: BoardMetrics,
    consensus: ConsensusMetrics,
    grades: DraftGrades,
) -> list[TeamRow]:
    board_by_roster = {t.roster_id: t for t in board.teams}
    cons_by_roster = {t.roster_id: t for t in consensus.teams}
    grade_by_roster = {t.roster_id: t for t in grades.teams}
    label_by_pick = {pick.pick_no: pick.board_label for pick in league.picks}
    rows: list[TeamRow] = []
    for team in league.teams:
        tb = board_by_roster.get(team.roster_id)
        tc = cons_by_roster.get(team.roster_id)
        tg = grade_by_roster.get(team.roster_id)
        rows.append(
            TeamRow(
                roster_id=team.roster_id,
                manager=team.manager,
                pick_count=tb.pick_count if tb else 0,
                pick_nos=list(tb.pick_nos) if tb else [],
                positional_counts=_canon_counts(tb.positional_counts if tb else {}),
                back_to_back=list(tb.back_to_back) if tb else [],
                best_value_pick=_extreme(tc.best_value_pick if tc else None, label_by_pick),
                biggest_reach_pick=_extreme(tc.biggest_reach_pick if tc else None, label_by_pick),
                # ``tg`` is expected to always be present (grades are computed for
                # every roster in ``league.teams``); when it is not, the ``TeamRow(...)``
                # construction below raises pydantic's ``ValidationError`` (``None`` is not
                # a valid ``float``/``int``/``str``) -- caught by the outer try/except in
                # ``build_draft_recap_facts`` and re-raised as ``SchemaValidationError``,
                # just as it did before these casts -- cast() is a no-op at runtime, it
                # only tells mypy the same thing the surrounding merge already assumes.
                draft_score=cast(float, tg.draft_score if tg else None),
                premium_picks=cast(int, tg.premium_picks if tg else None),
                format_fit=cast(int, tg.format_fit if tg else None),
                grade_input=cast(float, tg.grade_input if tg else None),
                grade=GradeRef(
                    letter=cast(str, tg.letter if tg else None),
                    driving_picks=list(tg.driving_picks) if tg else [],
                ),
            )
        )
    return rows

View this file on GitHub at v0.18.0

Edge cases

  • Ties for best or worst. The earlier pick number wins.
  • No ranked pick. If none of a team's players has a consensus slot, both extremes are null.
  • Unknown position. A pick with no position counts as UNK. It still adds to pick_count and appears in positional_counts after the four standard positions.
  • Team with no picks. pick_count is 0, the lists are empty and the counts are all zero.
  • Back to back means consecutive overall pick numbers, which in a snake draft happens at the turn between rounds.

See Draft team rows in a full league

Put in your Sleeper username or league ID, or look around a fictional league first.

Supported: Sleeper · See supported apps

See a full league while we open sign-ups.

On this page