Skip to main content
Glama

Backtest a draft target: player type vs a defense

invoke_backtest_draft_target
Read-onlyIdempotent

Added after you connected: the first call asks you to approve it. BankrollSZN Targets. Spot a defensive weakness and turn it into a target. Backtests how one kind of player did against one defense: every player-game this season (or a past season) where a player with the chosen traits faced that team, each graded against the player's OWN season average. Returns player-games, per-stat delta (for example +3.9 PTS/game), percent vs baseline and hit rate (share of games above his own average), the players behind it, the game list (newest first), and the players who qualify against that defense on today's slate. NBA traits: play-type share (Isolation >= 12%, PnR ball handler, spot up and the rest, from stats.nba.com Synergy), shot-zone share, size group, position, pull-up share, drop killer. NFL: position, target share, air yards share, rush share, WOPR. NHL: position, TOI, PP TOI, shots per game. MLB: bats, position, ISO, K rate, HR rate (hitters vs that team's pitching). After showing the result, ask the user if they want to save it as a target (my_targets action save with the same arguments); saved targets surface qualifying players on every slate and on https://bankrollszn.com/targets. Example: sport nba, defense ORL, play_types [{type: Isolation, min: 12}].

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "sport": "<sport>",
      "defense": "<defense>"
    },
    "name": "invoke_backtest_draft_target"
  }
}

Input Schema

NameRequiredDescriptionDefault
maxNoMaximum season trait values. Same keys as min.
minNoMinimum season trait values. NBA: pullup_share, catch_shoot_share, drop_killer, switch_killer, roll_threat, mins. NFL: target_share, air_yards_share, wopr, rush_share, targets_pg, carries_pg, pass_att_pg, ppr_pg. NHL: toi_pg, pp_toi_pg, shots_pg, points_pg, blocks_pg, dk_pg. MLB: iso, avg, slg, k_rate, bb_rate, hr_rate, sb_pg, dk_pg, pa. Shares as fraction or percent.
batsNoMLB only: L, R, S.
limitNoMax games in the game list. The plan limit applies.
sportYesSport.
zonesNoNBA only. Shot-zone share of the player's FGA: Restricted area, Paint (non-RA), Mid-range, Corner 3, Above the break 3.
offsetNoPaid plans: next page offset from page.next_offset.
seasonNoSeason. NBA and NHL 2025-26 style, NFL and MLB a year. Defaults: NBA 2025-26, NHL 2025-26, NFL 2025, MLB 2026 (the last full season). The current season works too.
defenseYesThe defense to test against: team abbreviation (NBA ORL, NY; NFL MIA, LA; NHL SJS; MLB COL means hitters vs that team's pitching).
playersNoLimit to these player names.
primaryNoStat the game list is graded on. NBA pts (default), pts36, fga, reb, ast, tpm, pra, ts, dk, min. NFL ppr (default), tgt, rec, recyd, car, rushyd, scrim, td, passyd. NHL dk (default), sog, pts, blk, sb, toi. MLB dk (default), h, tb, hr, rrbi, k.
min_gamesNoMinimum games in the season for a player to count (baseline sample). Defaults NBA 5, NFL 3, NHL 5, MLB 10.
positionsNoNBA G, F, C. NFL QB, RB, WR, TE. NHL C, L, R, D, or F for all forwards. MLB C, 1B, 2B, 3B, SS, LF, CF, RF, DH, OF, IF.
play_typesNoNBA only. Play-type share of the player's possessions (stats.nba.com Synergy). type: PnR ball handler, PnR roll man, Isolation, Transition, Spot up, Handoff, Post up, Off screen, Cut, Putback, Misc (short names like iso or pnr work). min and max as a fraction (0.12) or percent (12).
size_groupsNoNBA only: c7 (7-0 and up), big (6-9 to 6-10), wing, small (6-5 and under).
min_thresholdNoGame inclusion threshold. NBA minutes (default games over 24 min), NHL TOI minutes (default 10), NFL opportunities (default 1), MLB plate appearances (default 3).

Schema Changelog

Changes observed during successful MCP inspections.

  1. Added

TDQS

A3.6/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnly/idempotent/non-destructive, so safety is covered. The description still adds real behavioral context: an approval prompt on first call, plan-based limits, paid-plan pagination via next_offset, and the exact shape of the returned result (player-games, per-stat delta, percent vs baseline, hit rate, game list newest first, today's qualifiers). The approval sentence is awkwardly worded but discloses a genuine behavior.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The purpose and return-value content is front-loaded well, but the trait enumerations for NBA/NFL/NHL/MLB largely repeat schema descriptions, and the opening 'Added after you connected' fragment is a confusing lead. Dense but partly redundant for a 16-parameter tool.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 16-parameter tool with nested objects and no output schema, the description covers return values, season defaults, sample-size defaults, and the save-as-target follow-up. It is close to complete; only explicit sibling routing is missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the schema already documents every parameter including nested min/max, zones, and play_types. The description adds only an inline example call and a trait-to-field mapping (e.g. Isolation >= 12%), which is marginal beyond what the schema states. Baseline 3 applies when the schema carries the load.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The core sentence is precise: it backtests how one kind of player performed against one defense, graded against that player's own season average. That is a specific verb plus resource an agent can distinguish from the pool/matchup siblings. The opening sentence about 'added after you connected' is confusing noise that delays the real purpose statement.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It gives an implied usage context ('Spot a defensive weakness and turn it into a target') and a concrete follow-up path: after showing results, offer to save via my_targets with the same arguments. It never states when NOT to use this versus a per-player matchup tool, so an agent must infer the boundary from sibling names alone.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources