Gold Digger · An Essay · Screwcap Games
Earned Conviction
Why a trader should learn the past before betting the present — and why the obvious shortcut quietly manufactures confident fools.
Most people form firm opinions about markets years before they assemble a single piece of evidence that those opinions are any good. Gold Digger is built to run that sequence in the correct order — which is to say, almost exactly backwards from how everyone does it. This essay lays out the reasoning behind the product's central choice: that a player should not begin in the live present but should first earn a position through structured, experience-first training. We lean on three bodies of research — the description–experience gap, the difference between kind and wicked learning environments, and John Boyd's OODA loop — and we are candid about the trap: that naïve "just replay history" onboarding can reproduce the exact disease it claims to cure. Three safeguards convert the trap into a teacher. They are not optional, and we will explain why with the appropriate amount of menace.
§1 The unearned position
A newcomer dropped into a live market does not arrive empty-handed. They arrive armed — with opinions assembled from headlines, the collective mood of a forum, and whichever commentator sounded most certain at two in the morning. What they conspicuously lack is any record of whether those opinions survive contact with an outcome. The wager behind Gold Digger is that the gap between holding a view and having tested one is the most expensive thing in a young trader's life, and — happily — the only expensive thing on the list you can close for free, before a dollar is at risk.
The intuition is almost too simple to publish, so we have surrounded it with citations to feel respectable. Starting in the present feels natural; it is also, technically, cheating. Today's headlines are familiar, and familiarity is a con artist — it flatters you into believing you understand a situation you have merely recognised. Learn to read the past first, then act in the present, and you build a sturdier skill than either half produces alone. The literature below explains why that intuition is correct, and — the part worth the admission price — exactly where it betrays you if you build it lazily.
§2 Description versus experience
Decision researchers have known for decades that people make systematically different bets depending on how they learned the odds. Hand someone the probabilities as a description — a number, a forecast, a glossy brochure — and they overweight rare events. Make them learn the same odds the hard way, one lived outcome at a time, and they underweight those rare events instead. The founding demonstration is Hertwig, Barron, Weber & Erev (2004), with the broader synthesis in Hertwig & Erev (2009). It replicates well enough to have earned a blunt name: the description–experience gap.
A player reasoning from this morning's headlines is deciding from description. A player who has traded through a regime is deciding from experience. They are not the same trader, and only one of them is dangerous to bet against.
This is the academic skeleton inside the phrase "unearned position." The newcomer in the live present is, by definition, deciding from description — the brochure version of risk. The player who has sat inside a market regime, felt a thesis form, watched it get tested, and eaten the loss when it failed, has metabolised the same information into experience. So designing the onboarding such that a player's first brushes with risk are experiential is not a matter of taste. It changes the mental model they build of how markets actually behave — and the model, not the opinion, is the thing they keep.
§3 Kind and wicked environments
Here is the catch that ruins most "learn by doing" pitches: experience is not automatically educational. Whether practice yields genuine skill or merely well-upholstered error depends entirely on the structure of the feedback. Robin Hogarth's distinction between kind and wicked learning environments (Hogarth, 2001; Hogarth, Lejarraga & Soyer, 2015) is the load-bearing idea. In a kind environment, feedback is accurate, plentiful, and tied tightly to the action that produced it; chess is the patron saint. In a wicked one, the rules drift, the patterns refuse to repeat, and the feedback arrives late, noisy, or grinning and lying to your face.
Markets are the textbook wicked environment, which is precisely why experience in them so reliably teaches the wrong lesson. Hogarth's favourite example deserves to be read twice: a celebrated nineteenth-century physician who could "diagnose" typhoid by palpating patients' tongues — and was, of course, simply spreading it with his hands. Every fresh case "confirmed" his gift. His run of successes was teaching him, with perfect feedback, the single worst lesson available. Kahneman & Klein (2009) arrive at the same place from the opposite tradition: an intuition is only as trustworthy as the environment is predictable and the learner's chance to grasp its regularities. Hand someone repetitions in a wicked environment and you do not produce an expert. You produce a person who is wrong at volume, and proud.
Naïve historical replay can look kind while remaining thoroughly wicked. If the player recognises the period, they recall an answer instead of making a call. A single replayed path teaches one realised outcome, not the distribution of things that could have happened. And paper money never bleeds, so the emotional half of the lesson — the half that actually sticks — is simply missing. Build history-first onboarding carelessly and you have not made a teacher. You have built a confidence factory, and confidence is the one commodity markets are happy to short.
§4 Manufacturing kindness on purpose
The cheering news is that kindness can be engineered; it does not have to be hoped for. Hogarth & Soyer (2011) showed that letting people experience sequentially simulated outcomes turns an opaque, easily-misread description into kind experience — which reads less like a finding and more like a build spec for the mechanic Gold Digger is wrapped around. Their later book, The Myth of Experience (2020), supplies the bookend nobody wants on the poster: unstructured experience misleads at least as often as it instructs. So we treat three safeguards as non-negotiable, in the load-bearing sense, not the marketing one.
1 · Anonymise the scenario
While you play, the period's identity — its date, and where feasible its instrument labels — is withheld, and revealed only after the verdict is locked. A player who knows they are inside the 2008 unwind is testing their memory, not their judgment, and memory is a much easier exam. Revealing the market one observation at a time is the same trick that makes professional bar-replay tools work: it confiscates hindsight by forcing the past to unfold as if it were, inconveniently, the present.
2 · Score calibration, not luck
Because a single wicked path will cheerfully reward a terrible decision and punish a brilliant one, the headline number is not profit-and-loss but calibration: did your stated confidence actually match your hit-rate? We operationalise it with a Brier-style score (Brier, 1950), the long-standing measure for probabilistic forecasts. Rewarding the well-justified call over the lucky one is the entire trick by which a wicked domain is dragged, complaining, into kindness.
3 · Teach the distribution, not the path
Where we can, the same setup is shown with more than one plausible continuation, and every debrief is framed as "here is what happened — and here is what could just as reasonably have happened instead." It is the cheapest insurance policy available against the very human conviction that whatever did occur was always obviously going to.
§5 The OODA loop, borrowed from people who meant it
The moment-to-moment loop a player runs is lifted, without apology, from military aviation. The OODA loop — Observe, Orient, Decide, Act — was developed by USAF Colonel John Boyd to explain how a fighter pilot wins not through better hardware but by cycling decisions faster and more coherently than the other aircraft (Boyd, 1996; for the serious treatment, Osinga, 2007). The detail everyone skips: Boyd's own model is not a tidy four-box carousel. Orientation is the step that does the work — the instant where raw observation is weighed against everything you already carry. Without it, data is just weather.
The first scenario is paced for orientation, not speed. You are made to watch the market move before anyone lets you move with it.
This drives the onboarding directly. A player's first session is a synthetic, invented scenario — no real history to recognise, no capital to lose — paced deliberately slowly so the loop can be learned one stage at a time. You observe the feed and the price action; you orient with scaffolding that explains how a move in the dollar or in real yields actually bears on the metal; you decide a direction and, crucially, a confidence; you act, and the clock runs. Then the loop comes round again, tightening each pass as the scaffolding quietly falls away and the verdict window narrows. Only once the loop is in the hand does the player reach the fork: historical rounds, anonymised, or the live present, consequences included.
§6 The training ladder
The progression — synthetic → anonymised historical → live present — mirrors how serious practitioners actually build a craft, and for the same reason. Deliberate-practice theory (Ericsson, Krampe & Tesch-Römer, 1993) holds that skill is forged through focused repetition at the edge of current ability with immediate feedback, and historical replay is almost unfairly well-suited to this: it compresses the reps, letting a player run dozens of regime-snippets in an afternoon rather than waiting a decade for them to politely recur. We will note, against our own interest, that the literature is contested — a large replication found the effect of practice real but distinctly smaller than the original headline (Macnamara & Maitra, 2019) — which is itself a fine argument for pairing the reps with calibration scoring rather than rewarding sheer volume, the trader's favourite vanity metric.
The final rung we handle with conspicuous honesty. Graduating to the present does not just add recency; it adds consequence — the one ingredient no simulation can supply, no matter how good the felt looks. We say so out loud, because a training ladder that pretends the sim is the real thing has merely taught a more sophisticated species of unearned confidence, and we already covered why those keep losing.
§7 The record is the product
Every call a player makes is logged immutably — no edits, no quiet deletions, no flattering revisions after the fact — because the integrity of the record is the product, and an editable track record is just a diary with good PR. From that record we surface accuracy by market and by regime, confidence calibration over time, and the precise conditions under which a player's judgment holds or quietly falls apart. The leaderboard built on those records is, deliberately, public and free — not sealed behind the subscription. A visible, competitive, shareable record is the engine of return play and word-of-mouth; the paid tier sells analytical depth — the laboratory, not the scoreboard. We argue the full case for that inversion in the companion essay, The Free Leaderboard Paradox; the short version is that gating the social hook strangles the exact behaviour the product runs on.
The whole thing, on one card
| Principle | Why it's there |
|---|---|
| Experience before description | Closes the description–experience gap; builds a felt model of risk, not a brochure one. |
| Anonymise history | Confiscates hindsight, so the player makes a call instead of a recollection. |
| Calibration over P&L | Drags a wicked single-path domain into kindness by paying out for good decisions, not luck. |
| Orientation-first pacing | Honours Boyd's schwerpunkt; the first loop is slow on purpose, not by accident. |
| Honest graduation | The present adds consequence the sim cannot fake; saying so inoculates against false confidence. |
| Free public leaderboard | The record is the product; the social hook is what brings people back. |
§8 What this will not do, stated plainly
We are candid about the ceiling, partly from principle and partly because pretending otherwise is how people end up suing you. A simulation cannot reproduce slippage, real liquidity, fees, or — the big one — the way real money rearranges your nervous system; and a glittering record inside the trainer is emphatically not proof of a live edge. Gold Digger is an educational, simulated environment. It is not financial advice and it is not a brokerage. Market data shown in the live mode is delayed. Where the trainer eventually links out to real platforms, those links exist so a trained player can take an earned next step — not so we can manufacture one and call it graduation. The entire point of this apparatus is to make a player's conviction earned, and then to be unfashionably honest about precisely how far "earned" gets you.
§9 Data & community resources
The trainer stands on public data and public argument. Sources are listed in full and linked directly, in the spirit of attribution and the absence of pretence that any of this was conjured from a private genius.
Market & macro data
Community intelligence
From the Screwcap Games studio
§ References
- Boyd, J. R. (1996). The Essence of Winning and Losing [briefing].
- Brier, G. W. (1950). Verification of forecasts expressed in terms of probability. Monthly Weather Review, 78(1), 1–3.
- Ericsson, K. A., Krampe, R. T., & Tesch-Römer, C. (1993). The role of deliberate practice in the acquisition of expert performance. Psychological Review, 100(3), 363–406.
- Hertwig, R., Barron, G., Weber, E. U., & Erev, I. (2004). Decisions from experience and the effect of rare events in risky choice. Psychological Science, 15(8), 534–539.
- Hertwig, R., & Erev, I. (2009). The description–experience gap in risky choice. Trends in Cognitive Sciences, 13(12), 517–523.
- Hogarth, R. M. (2001). Educating Intuition. University of Chicago Press.
- Hogarth, R. M., & Soyer, E. (2011). Sequentially simulated outcomes: Kind experience versus nontransparent description. Journal of Experimental Psychology: General, 140(3), 434–463.
- Hogarth, R. M., Lejarraga, T., & Soyer, E. (2015). The two settings of kind and wicked learning environments. Current Directions in Psychological Science, 24(5), 379–385.
- Kahneman, D., & Klein, G. (2009). Conditions for intuitive expertise: A failure to disagree. American Psychologist, 64(6), 515–526.
- Macnamara, B. N., & Maitra, M. (2019). The role of deliberate practice in expert performance: revisiting Ericsson, Krampe & Tesch-Römer (1993). Royal Society Open Science, 6(8), 190327.
- Osinga, F. P. B. (2007). Science, Strategy and War: The Strategic Theory of John Boyd. Routledge.
- Soyer, E., & Hogarth, R. M. (2020). The Myth of Experience. PublicAffairs.