Table of Contents
Adam Grimes sits down for a candid chat on the Desire To Trade podcast, and it’s exactly the kind of no-nonsense perspective traders crave. Grimes is a veteran discretionary-quant trader, CIO at Waverly Advisors, and author of The Art & Science of Technical Analysis. In this interview, he explains why most “chart magic” has no measurable edge, why simplicity beats indicator soup, and how disciplined execution matters only after you’ve proven your method actually works.
You’ll learn Adam’s swing-trading playbook: find robust, testable patterns (like pullbacks), size positions sanely, cut trades that don’t develop, and judge yourself by rule-following—not P&L—while your system proves its edge in forward tests. He shows how to blend quantitative validation with structured discretion, avoid over-leveraging traps, and ignore seductive but untested ideas so you can spend your energy on what compounds: process, statistics, and survivability.
Adam Grimes Playbook & Strategy: How He Actually Trades
Core Philosophy: Edge First, Then Execution
Adam Grimes builds everything on a validated statistical edge, and only then worries about discipline and mindset. If a pattern can’t be shown to beat randomness over many samples, it’s out. This section spells out how to think about edge the way he does, so your plan isn’t just “beliefs,” it’s math with teeth.
- Define your pattern’s hypothesis in one sentence (“Buy pullbacks in uptrends after momentum thrusts”).
- Backtest or forward-test a minimum of 200 trades; the expectancy isn’t positive after costs, discard it.
- Track distribution of outcomes (median win/loss, fat tails, outliers) and design rules around the distribution, not anecdotes.
- Separate detection (setup) from execution (entry/exit): if detection requires judgment, entry should be mechanical.
- Never “optimize” to perfection; prefer simple, robust rules that survive parameter nudges.
Market & Timeframe Selection
He trades liquid markets where slippage is small and behavior is stable. Swing is the default because it captures meaningful moves without the noise of tick-by-tick chop. Use this section to narrow your hunting grounds and avoid edges that die in illiquid names.
- Trade highly liquid futures, FX majors, or top-tier equities/ETFs with tight spreads.
- Default to daily/4h/1h for swing; use 5–15m only if your fills and discipline are excellent.
- Skip assets with irregular gaps, thin books, or event-driven spikes that overwhelm technical edges.
- Keep a “tradable universe” list and review quarterly; remove symbols that degrade your fills or edge.
- Avoid over-diversification: 10–20 symbols that truly move are better than 100 you can’t watch.
Regime & Context Filters
Context beats signals. Grimes emphasizes identifying trend, volatility, and mean-reversion regimes so you’re not forcing a tool into the wrong market. These rules tell you when a setup is valid and when it’s noise.
- Define trend with a simple rule: price above a rising 50MA = uptrend; below a falling 50MA = downtrend; else neutral.
- Define volatility state with ATR percentile (e.g., 20-day ATR vs 1-year history): <30% = quiet, >70% = hot.
- Only run trend-following entries in trending regimes and mean-reversion entries in neutral/quiet regimes.
- Stand down after outsized range days in quiet regimes; edge often mean-reverts first.
- If regime shifts mid-trade (e.g., volatility explodes against a mean-reversion position), cut size or exit.
Set up Taxonomy (What He Actually Trades)
He keeps a small menu: pullbacks in trends, breakouts with fresh momentum, and selective mean reversion in ranges. The point is repetition—fewer patterns, more reps, cleaner data.
- Trend Pullback: A strong impulse (range expansion) → 2–5 bar pause toward the 10–20MA → buy the first sign of momentum resumption.
- Breakout Continuation: Identify multi-week balance; enter on range expansion day closing beyond the balance with above-average true range.
- Mean Reversion (Selective): Faded extremes only in non-trending regimes; look for exhaustion bars into prior VWAP/MA “value.”
- Ignore indicator mashups; if a naked chart can’t show the edge, adding lines won’t fix it.
- Cap the active setup list at 3; add a new one only after you can prove it out of sample.
Entry Triggers (Simple, Testable, Repeatable)
Entries confirm the setup and time risk efficiently. Grimes prefers triggers you can code or at least measure, so you don’t guess.
- For pullbacks: buy stop 1 tick above the signal bar high that closes back in trend direction.
- For breakouts: enter on close beyond the structure or on next open if slippage is tolerable; avoid chasing mid-bar.
- For mean reversion: limit orders at predefined value zones; if not filled on the touch + confirmation bar, skip.
- Use a “two-strike” rule: if the same setup fails twice in a row on the same symbol/regime, pause that symbol for the day.
- Never average down; one entry per thesis unless a pre-planned scale plan exists and is tested.
Stop Loss & Trade Invalidation
Stops live where your idea is wrong, not where the P&L hurts least. Grimes leans on volatility and structure to avoid death by whipsaw.
- Initial stop for pullbacks/breakouts: beyond the opposing swing point or 1.5–2.0× ATR(14), whichever is farther.
- For mean reversion: hard stop outside the extreme (e.g., prior day high/low ± ATR buffer).
- Time stop: if no favorable progress after N bars (e.g., 5–8 on 1h, 2–4 on daily), exit at market.
- Move to breakeven only after price travels at least 1R; earlier moves cut off winners.
- If a large shock-gap invalidates the structure at the open, exit immediately—no heroic holds.
Exits & Targets (Harvest the Edge)
He wants exits that reflect the pattern’s payoff shape, not vibes. Let winners breathe while ensuring you actually bank gains.
- Pullbacks: first scale at +1R, trail remainder under higher lows (swing) or a short MA that stair-steps the trend.
- Breakouts: partial at measured move (height of base) or +1.5–2R; trail with ATR stop that widens in quiet regimes.
- Mean reversion: target the return to value or mid-range; rarely trail—take the snapback and be done.
- If a bar closes against your position with range expansion twice the recent average, consider an early exit (failed pattern).
- Always predefine exit logic before entry; no mid-trade reinvention.
Position Sizing & Risk
Survival > returns. Grimes suggests that so drawdowns are tolerable and compounding can work. The goal is staying in the game long enough for the edge to show up.
- Risk a fixed fraction per trade: 0.25%–0.75% of equity for swing; cap daily risk at 1–2× that amount.
- Size shares/contracts from the stop distance: Position = (Account × Risk%) ÷ Stop $/point value.
- Correlation cap: if multiple positions share the same macro driver, halve per-trade risk or take fewer at once.
- Hard max open risk: total portfolio risk ≤ 3% of equity across all trades.
- Reduce risk after a losing streak (e.g., -5R in a week → cut size by 50% until back to high-water mark).
Playbook for Daily Routine
Consistency comes from a boring, repeatable routine. Here’s how he structures the day so the plan gets executed the same way every time.
- Pre-market: mark regimes (trend/vol), key levels, and candidate lists; no predictions—just scenarios.
- During session: alerts at trigger levels; only act on preplanned signals; log each decision in real time.
- Post-close: grade trades on process (followed rules Y/N), not outcome; screenshot charts and tag by setup.
- Weekly: update stats (win rate, avg R, expectancy by setup/regime) and retire underperforming tactics.
- Monthly: review drawdown curves and adjust portfolio risk if heat is climbing.
Data Hygiene & Rule Maintenance
Edges decay if you don’t measure them. Grimes treats his rules as living code: stable by default, updated when the data demands it.
- Maintain a simple database of trades with columns for setup, regime, R outcome, and notes; sample size matters.
- Re-test parameters quarterly with small nudges (±10–20%) to confirm robustness; avoid wholesale overhauls.
- Separate “innovation account” at any size to trial new ideas without contaminating core performance.
- If a setup loses edge across multiple markets and regimes for 2+ cycles, deprecate it and free up focus.
- Keep your checklist printed and visible; if you can’t point to a rule on the sheet, don’t do it.
Psychological Execution (After the Edge Exists)
Mindset matters only in the service of the system. The job is to execute boring rules flawlessly, not to feel like a hero.
- Define “maximum boredom”: if you feel excited/scared, you’re probably off plan—pause and re-check the checklist.
- Pre-commit to a daily “no discretionary override” rule unless a predefined kill-switch condition is met.
- Use a streak protocol: after two losses, take a 15-minute reset before any new entry; after three, stop for the day.
- Score each session 0–5 on process adherence; only increase size after four consecutive sessions scoring ≥4.
- Celebrate rule-following with small, non-P&L rewards; wire your brain to crave execution, not outcomes.
Risk Events & Newsflow
He doesn’t trade headlines; he manages exposure around them. If a scheduled event can overwhelm technicals, he respects the blast radius.
- Flat or half-size into major scheduled events that historically spike ATR (policy decisions, key data).
- No fresh entries in the 15–30 minutes before high-impact releases on short timeframes.
- For positions held through events, widen stops via ATR only if that rule is in the plan; otherwise, exit and re-enter post-event.
- After shock gaps, reassess the regime; if the structure changed, treat it as a new market, not the same trade.
- Record event outcomes vs setup performance; if your edge dies around a specific release, add a “no-trade window.”
Quality Control & Continuous Improvement
This is how he keeps the machine sharp without chasing every shiny object. Iterate deliberately and protect the core.
- One change at a time; A/B new rules over 30–50 trades before promoting.
- Kill any rule that adds complexity without improving expectancy or reducing drawdown.
- Maintain a “stop doing” list (revenge trades, intraday tinkering, indicator adds) and read it before the open.
- Quarterly “edge audit”: verify each setup’s expectancy by regime and symbol; prune ruthlessly.
- Keep your playbook to a single page; if it doesn’t fit, you’re not trading it—you’re collecting hobbies.
Prove the edge first: mechanics beat predictions and opinions.
Adam Grimes is blunt: if you can’t show a repeatable edge, you’re just guessing with confidence. He wants traders to define a simple hypothesis, then test it across markets and time, looking for positive expectancy after costs. Mechanics—clear entries, stops, and exits—are what translate that edge into results, not a clever macro take. Prediction can be entertaining; mechanics are what pay.
Grimes suggests you forward-test for a few hundred trades, log every decision, and judge success by rule-following before P&L. If the idea can’t survive small parameter nudges or shifting volatility, it isn’t robust enough to trade size. Build a checklist that separates setup detection from trigger and exit so there’s no room for mid-trade invention. When the data says the edge is gone, retire it without ceremony and move on.
Size risk sanely using ATR; survive volatility, compound steadily.y
Adam Grimes stresses that risk comes first, returns second, and ATR is the cleanest way to translate chart distance into dollars. He sizes positions from the stop, not the entry, so one trade never outweighs another just because the instrument is “fast.” Pick a small fixed risk per trade—say 0.25%–0.75%—and let ATR determine how many shares or contracts you can hold. When volatility expands, size shrinks automatically; when markets calm, size scales up without guesswork.
Grimes also caps total portfolio heat, so several correlated trades don’t sink the account together. He uses volatility filters to cut exposure during stormy regimes instead of widening stops until the trade loses meaning. The math stays simple: Position = (Account × Risk%) ÷ Stop distance (in $ or points × value). If a losing streak hits, he reduces risk by half until process scores and equity stabilize. The goal isn’t hero trades—it’s steady compounding that survives every regime.
Diversify by market, setup, and timeframe to smooth the equity curve.s
Adam Grimes pushes diversification beyond just “more tickers.” He wants you to spread risk across uncorrelated markets, a small menu of proven setups, and at least two timeframes. If your entire book moves with the same macro driver, you’re not diversified—you’re levered to one story. Diversifying by setup matters too: a trend pullback and a mean-reversion fade don’t win and lose on the same days, which smooths the ride.
Grimes suggests building a core universe of liquid symbols from different asset classes, then assigning each setup to the timeframes where it performs best. A pullback on the daily can coexist with an intraday breakout in another symbol without stepping on each other. Cap position count when correlations spike, and rotate to the best-behaving instruments rather than forcing trades everywhere. Track expectancy by market, setup, and timeframe separately so you know what to add—or cut—when conditions change. Over time, that mix lowers drawdown depth and keeps your equity curve climbing with fewer gut punches.
Simple rules, repeatable routines: predefine entries, exits, and invalidation
Adam Grimes says your trading should run like a checklist, not a mood. Before the bell, define what counts as a valid setup, the exact trigger to get in, and the line that proves you wrong. If the market doesn’t give your trigger, you don’t “improvise” a different one—no signal, no trade. Exits are preplanned too: partials, full, and a time stop if price stalls. This keeps decisions consistent so you can trust your stats, not your adrenaline.
Grimes also builds a daily rhythm: scan, shortlist, alert levels, execute, log, review. He grades each session on rule-following, not P&L, so size only increases when the process stays tight. A simple “two-strike” pause—two quick failed attempts in one symbol or setup—prevents spiral behavior. When a trade violates the invalidation rule, he’s out immediately, no storytelling, and the routine starts again.
Avoid undefined risk; cap exposure, limit correlations, respect regime shifts.
Adam Grimes is clear: undefined risk is where traders blow up, not where they get rich. If a position can theoretically lose multiples of your account on a gap or volatility spike, it doesn’t belong in a repeatable playbook. Cap per-trade risk, cap total portfolio heat, and refuse positions whose downside you cannot bracket with a hard stop or defined structure. When correlations jump, treat a basket like one big trade and cut size accordingly. Survival is the edge multiplier; undefined risk is the edge killer.
Grimes also adapts to regime shifts instead of hoping they pass. Rising volatility, changing trend structure, or news-driven markets demand smaller size, faster trade invalidation, and fewer open bets. If a model’s edge degrades in the new regime, he parks it and redistributes risk to tactics that still work. He avoids “insurance selling” without protection, and when he uses options, he prefers defined-risk structures and exits early if volatility compresses the edge away. Respect the tape, respect the math, and make sure every position can be taken off quickly and cleanly when the market proves you wrong.
Adam Grimes leaves zero room for magical thinking. His core message is simple and demanding: build on a real edge you can articulate and observe across time, then let clear rules do the heavy lifting. He’s brutally honest about how easy it is to fool yourself with pretty charts, moving-average lore, or a lucky streak, and he repeatedly returns to the only test that matters—does the pattern actually behave differently from randomness after many samples, costs included? The moment you accept that standard, your toolbox shrinks to a few robust setups, your leverage drops to sane levels, and your decisions start serving the data instead of the ego.
He also reframes psychology as an execution discipline, not a substitute for a bad method. Wear the two hats: developer and trader. The developer defines and audits the system; the trader follows it to the letter and is judged only on adherence. That separation keeps you from rewriting rules mid-trade or blaming emotions for what is really a weak edge. Add in a respect for market context—some conditions reward pullbacks, others don’t—and a sober view of variance where strings of wins and losses are normal, and you get a process that survives. In the end, Grimes’s lesson is to trade fewer ideas at a smaller size with sharper definitions, accept uncertainty without trying to predict it away, and let consistent execution expose whether your edge is real or make-believe.
