The alternative is to encode the rules as data. Not because a schema is more truthful, but because it forces you to name the fields, and three of those fields do most of the work.

The schema

{
  "firm": "example",
  "account": {
    "size": 100000,
    "profit_target_pct": 8,
    "max_daily_loss_pct": 5,
    "max_total_loss_pct": 10,
    "drawdown_mode": "trailing_intraday" | "trailing_eod" | "static",
    "floor_basis": "balance" | "equity",
    "floor_locks_at_starting_balance": true,
    "reset_timezone": "America/New_York",
    "min_trading_days": 5,
    "consistency_cap_pct": null,
    "news_restriction": "none" | "window" | "banned",
    "weekend_holding": true,
    "payout_split_pct": 90
  }
}

Nine firms, one array. A comparison becomes a filter rather than an essay, and "which is easier" becomes a question about a specific field instead of a general impression.

The three fields that change the arithmetic

1. drawdown_mode

Static means the floor never moves. Trailing intraday means it follows the high-water mark inside the session. Trailing end-of-day means it only moves at the close.

Same trades, different outcome: a position that runs to plus two R and closes at plus half an R raises an intraday floor and leaves an end-of-day floor untouched. If your edge is a wide target with a small give-back, the mode decides whether you are trading a recovering account or a shrinking one.

2. floor_basis

Balance or equity. On a balance basis, an open loss does not move your floor. On an equity basis it does, which means the room available to a position that is already open is smaller than the room you measured before you entered it. This is the field that separates "my stop is 1%" from "my stop is 1% and it is 40% of what is left".

3. reset_timezone

The daily loss resets at a wall clock in a stated timezone. Two sessions of yours can sit inside one of theirs, and a firm that resets at 5pm New York is not giving a trader in Johannesburg a calendar day. The same rule, read twice by two people, produces two different daily budgets — which is precisely what prose does and a field does not.

What encoding buys you

function roomLeft(a, equity, highWater) {
  const base = a.floor_basis === 'equity' ? equity : highWater;
  const floor = a.drawdown_mode === 'static'
    ? a.size * (1 - a.max_total_loss_pct / 100)
    : base * (1 - a.max_total_loss_pct / 100);
  return equity - floor;
}

Once the rules are fields, three things get cheap. Sizing from the remaining room instead of from account size. Checking whether a strategy is viable on a firm at all, before paying a fee. And noticing when a term changes, because a change is a diff rather than a re-read.

The caveat is the point of the exercise: this encodes what a firm publishes today, and a term written differently is a different rule. Keep the fields separate from the marketing and you can re-encode without re-learning.

I keep the encoded comparison at the prop firm rules matrix, and the method for checking each firm against its own terms at methodology.