Why we publish our losses

Every trading service on the internet shows you the same thing: a green chart, a winning streak, a screenshot of someone's best month. Almost none of them show you the strategy that died, the week of stop-outs, or the backtest that fell apart in live trading.

That's not an accident. Showing only wins is the business model. If your revenue comes from people joining, you sell the dream and bury the misses. By the time the misses show up in a customer's account, the next cohort is already signing up.

Glasshouse is built on the opposite bet: in a market full of curated screenshots, verifiable honesty is the scarcest asset there is.

What "run in the open" actually means here

• The whole strategy book is public. Not just the flagship — every strategy, with its backtest and its live numbers side by side, including the ones currently losing.
• The failure log only grows. When a strategy is killed, the post-mortem is published and the entry is never deleted. What we chose not to trade is as informative as what we did.
• Trades are explained in one consistent format — market · side · entry · size · thesis · confidence · invalidation · exit — so a decision can be judged against what we said before the outcome, not rationalised after it.
• Backtests are labelled as backtests. A simulation is not a track record, and we never present one as if it were.

Why this is rational, not just virtuous

Because our economics only work one way: Glasshouse earns a share of profits, on a high-water mark, in accounts we can never withdraw from. There is no joining fee to harvest, so there's nothing to gain from overselling. The only way this desk makes money is the slow way — by actually being right over time, for people who stayed because they could verify what was happening.

If a service won't show you its losses, assume the losses are doing the talking somewhere you can't see.

What we don't publish

Honesty about results doesn't mean giving away the machine. Exact entry parameters, live order levels and unfilled targets stay private — publishing those would let the market trade against the desk and its investors. The line is simple: outcomes are public, the playbook is not.

Judge us on the record

The live book is early, and parts of it are not winning yet — you can see exactly which parts on the research page. That page will get better or worse in public, in real time. That's the point of a glasshouse.