Skip to main content
Glama

Backtesting Arena

by io.github.Schoasch
OverviewDiscussionsAdmin

Get Regime-Aware Strategy Performance

arena_get_strategy_performance_by_regime

In which macro regime has this strategy worked? Historical backtest performance for ONE (strategy, asset, interval) combination SPLIT BY macro market regime (sweet_spot / late_cycle_warning / crisis / recovery — classified at each trade's entry date), PLUS the CURRENT live regime so you can align the buckets yourself. You get the per-regime numbers to weigh directly (per-bucket verdicts live in the per-cell tools, where the pool is stable). Each regime bucket returns trades, trades_per_config (trade counts pool ALL parameter-variant configs — see config_count), win_rate, avg_pnl_pct (per-trade return, not annualized), reward_risk_ratio (per-trade mean/stddev, NOT annualized Sharpe), share_of_time_pct (calendar-day-weighted — each regime observation counts the days until the next one, so the mixed weekly/daily cadence of the regime history does not skew the share) and a rating. The benchmark block anchors the payload with the combination's buy-and-hold CAGR (identical to arena_get_strategy_performance vs_buy_hold — without that anchor, regime avg_pnl_pct is a trajectory, not an excess). For a decision-grade view compose with arena_get_strategy_filter_effect and arena_is_distinguishable. [Free tier]

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
assetYesCrypto pair / symbol (e.g. BTCUSDT, ETHUSDT). Case-insensitive.
intervalNoDefault '1w'. Candle interval: '1d' daily, '2d'/'3d' multi-day, '1w' weekly, '1M' monthly. Multi-day candles (2d/3d) are anchored to the Unix epoch, so one of n possible alignments is used. Measured on our own corpus, the choice of alignment alone moves CAGR by 6.66 pp on average (max 12.30). Treat differences below that as not distinguishable — 1d/2d/3d behaved as one block in our tests, not a ranking.
strategyYesStrategy key (e.g. rsi_sma, golden_cross). See arena_get_strategies.
asset_typeNoOptional asset class filter to disambiguate identical pair-names.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed3 schema fields changed
    • addedInput schema / additionalProperties
      Added value: +false
    • removedInput schema / properties / context
      Removed value: -{
-  "description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\"",
-  "type": "string"
-}
    • changedInput schema / required
      Previous value: -[
-  "strategy",
-  "asset",
-  "context"
-]New value: +[
+  "strategy",
+  "asset"
+]
  2. Changed1 schema field changed
    • changedInput schema / properties / interval / description
      Previous value: -"Default '1w'."New value: +"Default '1w'. Candle interval: '1d' daily, '2d'/'3d' multi-day, '1w' weekly, '1M' monthly. Multi-day candles (2d/3d) are anchored to the Unix epoch, so one of n possible alignments is used. Measured on our own corpus, the choice of alignment alone moves CAGR by 6.66 pp on average (max 12.30). Treat differences below that as not distinguishable — 1d/2d/3d behaved as one block in our tests, not a ranking."
  3. Changed1 schema field changed
    • changedInput schema / properties / interval / enum
      Previous value: -[
-  "1d",
-  "1w",
-  "1M"
-]New value: +[
+  "1d",
+  "2d",
+  "3d",
+  "1w",
+  "1M"
+]
  4. Changed1 schema field changed
    • changedInput schema / properties / asset_type / enum
      Previous value: -[
-  "crypto"
-]New value: +[
+  "crypto",
+  "tokenized_equity",
+  "tokenized_etf",
+  "commodities"
+]
  5. Changed2 schema fields changed
    • changedInput schema / properties / asset / description
      Previous value: -"Pair / symbol (e.g. BTCUSDT, AAPL.US). Case-insensitive."New value: +"Crypto pair / symbol (e.g. BTCUSDT, ETHUSDT). Case-insensitive."
    • changedInput schema / properties / asset_type / enum
      Previous value: -[
-  "crypto",
-  "stock",
-  "etf",
-  "commodities",
-  "forex"
-]New value: +[
+  "crypto"
+]
  6. First observed

TDQS

A4.2/5.0
Behavior4/5

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

With no annotations, the description carries the full burden and does so well. It discloses that regime classification happens at trade entry date, that trades_per_config counts all parameter-variant configs, that avg_pnl_pct is not annualized, reward_risk_ratio is not annualized Sharpe, and that share_of_time_pct is calendar-day-weighted to avoid cadence skew. It also explains the benchmark anchor's purpose. Minor gaps remain around the rating scale and live regime presentation.

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 description is dense and informative, but it is long and structured as a single block, which reduces scannability. Many parentheticals and qualifiers are valuable, but some could be trimmed or front-loaded more effectively. It earns a mid-range score because every sentence carries meaningful content, but formatting and pacing are not optimal.

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?

Despite having no output schema, the description compensates by enumerating most returned fields and their semantics, explaining the benchmark block, and addressing the large interval caveat. It also references sibling tools for composition. Missing details such as rating scale or live regime format are minor. Overall, the description is robust enough for an agent to call and interpret results correctly.

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

Parameters4/5

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

Schema description coverage is 100%, so the baseline is 3. The description adds meaningful context beyond the schema, especially for the interval parameter: it warns that multi-day candle alignment can move CAGR by 6.66 pp on average and that 1d/2d/3d behaved as one block. This enriches interpretation of the interval parameter. The description doesn't add much for asset or asset_type, but the schema already covers them.

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

Purpose5/5

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

The description states a specific verb and resource: it retrieves historical backtest performance split by macro regime for ONE (strategy, asset, interval) combination. It also distinguishes itself from related tools by clarifying that per-bucket verdicts live elsewhere and that the benchmark block matches arena_get_strategy_performance. This provides clear differentiation.

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

Usage Guidelines4/5

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

The description gives clear context for when to use the tool: when you need per-regime numbers to weigh directly and align with the current live regime. It also suggests composing with arena_get_strategy_filter_effect and arena_is_distinguishable for a decision-grade view. It lacks explicit when-not-to-use instructions, but the guidance is adequate.

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

All Backtesting Arena toolsInspect
Try in Browser
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources