Table of Contents
This interview features Matt DeLong—entrepreneur-turned-full-time trader—breaking down how he went from manual swing trades to building automated systems. He’s founded and sold multiple software companies, trades stocks/options/ETFs, and is now developing systematic engines and a future fund structure. It’s a candid live conversation about why he favors bullish trend-following, when he sells naked puts, and how he balances real trading with automation work.
You’ll learn how Matt validates a strategy from first backtest to walk-forward testing, why clean historical data matters, and how he defines risk using fixed-dollar “R” units. He explains portfolio construction for overnight gaps, switching strategies by market regime, avoiding curve-fit traps, and when to hire a programmer or use tools like TradingView, MetaTrader, and lightweight ML prototypes. The goal: a simple blueprint you can copy—design rules, test honestly, size sanely, and automate only what survives.
Matt DeLong Playbook & Strategy: How He Actually Trades
Core approach: bullish swing, pullbacks, end-of-day signals
Matt DeLong trades like a builder: simple rules, clean data, and repeatable execution. He focuses on bullish trend-following swing trades using end-of-day signals, preferring pullbacks over breakouts for better risk/reward and fewer fakeouts.
- Trade direction: long bias only unless a separate short strategy is explicitly enabled.
- Timeframe: swing; entries and exits are placed after the daily close and executed next session.
- Entry style: buy pullbacks within established uptrends, not breakouts.
- Minimum expectancy: target setups that show ≥2:1, ideally 3:1 reward-to-risk.
- No intraday tinkering: once orders are staged, let the system run.
Finding the trade: define the uptrend and the pullback
You don’t need exotic indicators—just a consistent, testable definition. Matt wants higher highs/higher lows with a controlled retrace that lets him size the trade properly.
- Universe filter: U.S. stocks/ETFs priced roughly $5–$200.
- Uptrend definition: price above a medium-term baseline and printing higher swing highs/lows.
- Pullback window: buy only after a measured retracement into value (e.g., prior swing area or mean).
- Confirmation: enter on the next session if price holds the pullback area; skip if momentum collapses.
- Disqualifiers: thin liquidity, major corporate events pending, or structurally broken charts.
Entries & exits that compound the edge
The edge is earned by entering when risk is cheap and exiting where risk stops paying. This section turns the pullback idea into rules you can code or follow manually.
- Entry trigger: next-day buy stop or limit at/near the pullback zone; no chasing beyond planned price.
- Initial stop: just beyond the logical invalidation (below swing/structure), placed when the order is placed.
- Profit taking: scale or exit into predefined targets (T1 ≈ 1R, T2 ≈ 2–3R); trail only if rules say so.
- Time stop: if price goes nowhere after N bars (your tested value), exit flat and recycle capital.
- Earnings rule: no new entries within the earnings risk window; flatten or reduce before the print.
Position sizing: fixed-dollar “R” and portfolio limits
Matt sizes risk in dollars, not vibes. He caps open positions and adjusts aggression by account heat, so one bad day can’t nuke the month.
- Risk unit: 1R = fixed dollar loss per trade (commonly aligns with ~1–2% of equity).
- Share/contract calc: position_size = R / (entry − stop). Round down; never exceed.
- Max concurrent trades: ~8–10 per strategy; tighten when overall account heat is high.
- Monthly dial: if the account is up meaningfully for the month/quarter, you can step slightly heavier; if down, cut risk to half-R until back above water.
- Catastrophe guard: hard daily or weekly loss caps that pause new entries system-wide.
Options overlay: get paid to wait with naked puts
When he’s happy to own the stock lower, Matt sells puts instead of placing far-off buy limits. The premium cushions entry and forces discipline around assignment.
- Underlyings: only liquid names you’d own if assigned.
- Strike: near a price you’d gladly buy shares (often near pullback support).
- Tenor: short-dated enough to keep premium decay brisk, but always liquid.
- Risk rule: if price closes decisively through your assignment pain line, buy back/roll—don’t hope.
- Assignment plan: predefine stock position sizing and stop once assigned; options don’t erase risk rules.
Testing pipeline: from backtest to “live-like” simulation
Before money touches a rule set, it must survive history and a current-market sandbox. Matt runs long historical tests, then live-environment simulations to catch regime issues and data problems.
- Backtest span: test across many years and different regimes; require stable equity curves, not one golden decade.
- Walk-forward: validate on unseen periods, then simulate in current markets without capital.
- Checklist gate: no strategy goes live until it passes data integrity, slippage, and robustness checks.
- Challenger vs. contender: A/B test new variants against a proven baseline; promote only if statistically better.
- Kill switch: auto-disable on bad data flags or if drawdown exceeds predefined thresholds.
Data hygiene & automation guardrails
Automation only works if your inputs are right. Matt treats data quality as a risk factor and builds alerts that keep the system honest.
- Data verification: block trading when feeds show gaps/dupes/missing fields; alert and investigate.
- One source of truth: standardize corporate actions (splits/dividends) before calculations.
- Order staging: generate and stage EOD orders; human reviews exceptions only.
- Monitoring: real-time status panel for fills, slippage outliers, and rule violations.
- Change control: version every strategy; log parameter changes with timestamps and reasons.
Portfolio rules: capacity, diversification, and maintenance
Trade small, in many places, with independent edges. Matt spreads exposure by ticker and by strategy, and re-tunes non-price parameters on a schedule—not every night.
- Capacity: cap total exposure per ticker/sector; avoid correlated pile-ups.
- Strategy mix: separate long-only swing, options overlay, and any specialty systems; each has its own heat cap.
- Re-optimization: review non-directional variables (e.g., max positions) monthly; never curve-fit entries/exits to the last few weeks.
- Cash buffer: keep dry powder for sudden pullback clusters; don’t run 100% margin.
- Review cadence: weekly performance review; monthly parameter audit; quarterly deprecation of weak systems.
Business mindset: track record first, capital later
Matt treats trading as an operating business. Build a real, auditable track record before seeking scale; let the numbers invite the capital.
- Track record: let live strategies accrue 6–18 months of results before scaling.
- Segregation: keep “R&D” systems small and ring-fenced from production capital.
- Investor readiness: hard docs—playbooks, risk limits, and operational checklists—prepared in advance.
- Communication: performance is reported by strategy and consolidated; no black-box promises.
- Scale rule: increase allocation only after new equity highs and within predefined capacity limits.
Size fixed-dollar risk, let winners run, cap total account head.t
Matt DeLong keeps it simple: define 1R in dollars and never violate it. He calculates position size from that fixed-dollar risk and the distance to the stop, then rounds down so slippage can’t push him over the line. Because the risk is fixed per trade, he doesn’t need to guess conviction or feel; the math sets the size. That frees him to focus on execution and post-trade management instead of tinkering with entries.
Once in, Matt DeLong lets profitable trades breathe while respecting account-level heat limits. Winners are allowed to stretch toward predefined multiples, but new risk pauses if total open exposure crosses the heat cap. If a trade stalls or violates structure, he exits without debate to recycle capital into cleaner setups. The result is a repeatable loop: define loss first, scale gains second, and keep the entire book within a temperature you can actually sleep with.
Trade bullish pullbacks, end-of-day entries, skip earnings landmines.
Matt DeLong prefers buying strength on sale, not chasing breakouts. He defines an uptrend first, then waits for a clean pullback into prior structure or a mean zone where risk is cheap. Signals are reviewed after the close, so entries can be staged calmly for the next session. This keeps emotions out and turns each trade into a scheduled decision, not a midday impulse.
End-of-day execution also makes risk controls straightforward around volatile events. Matt DeLong avoids opening new positions into earnings, and he’ll flatten or reduce if a report sneaks up on him. By sidestepping those landmines and entering only when pullbacks hold, he keeps losses small and gives winners room to move. The combo—trend first, discount entry, no earnings roulette—adds up to cleaner trades and steadier equity curves.
Diversify by ticker, strategy, and timeframe to cut correlation spikes.
Correlation clusters can wreck a good system, so spread bets across independent names. Cap exposure per ticker and sector, and avoid piling into symbols that move together on the same macro theme. Stagger entries across days so one headline can’t slam every fill at once. Mix individual stocks with broad ETFs to reduce single-name risk without losing market participation.
Layer multiple strategies that win for different reasons, not just the same signal with new colors. Combine a long-only pullback system with an options income overlay and a separate trend-rider to smooth the curve. Matt DeLong tracks heat for each bucket and trims the most correlated slice first when markets tighten up. Review rolling correlations monthly, prune the laggards, and keep a cash buffer so you can rotate without forced sells.
Systemize mechanics over predictions, test honestly, promote only robust edge.s
Matt DeLong treats rules like code: clear inputs, deterministic outputs, no “feel.” He builds mechanics that answer when to enter, where to place the stop, and how to exit—then follows them verbatim. Indicators are servants, not oracles; if a rule can’t be stated in one line, it’s probably curve-fit. The goal is repeatability under pressure, not brilliance on a perfect chart.
Testing is equally strict. Matt DeLong demands out-of-sample results, walk-forward checks, and parameter ranges that work, not just a single magic number. Any strategy facing data gaps, slippage spikes, or fat drawdowns gets throttled or killed automatically. Only edges that stay profitable across regimes get promoted to real capital, and even then, they run with version control and a prewritten kill switch.
Use options smartly: sell puts to enter quality names cheaply. er
Matt DeLong uses options as a disciplined entry tool, not a lottery ticket. When he wants stock exposure but prefers a discount, he sells cash-secured puts at strikes where he’d be happy to own shares. The premium lowers rs effective entry price and pays him for waiting, which beats tossing in lowball limit orders that may never fill. He chooses liquid names and expirations with tight spreads so slippage doesn’t eat the edge.
Risk rules still run the show. If price closes through his pain line, Matt DeLong buys back or rolls—he doesn’t “hope” assignment turns around. Assignment is planned, not a surprise: he sizes for the potential shares and applies the same stop and target logic as his swing system. Earnings roulette is off limits; no fresh short puts into a binary event. The result is a simple, repeatable overlay: collect premium when conditions are right, accept assignment only by design, and keep total portfolio heat within strict limits.
Matt DeLong’s core lesson is to make trading boring on purpose: define loss first, automate the mechanics, and let the numbers—not feelings—decide. Fixed-dollar R keeps every position honest, while end-of-day execution turns entries into scheduled decisions instead of impulses. He buys strength on sale through bullish pullbacks, steers clear of earnings roulette, and lets winners stretch toward predefined multiples. When premium makes more sense than a resting limit, he sells puts only in names he wants to own, with assignments sized and planned.
Just as important is how he runs the operation. Strategies must pass long-history tests, walk-forward checks, and live-like simulations before they touch real capital—and each one carries its own kill switch. Data hygiene, version control, and account-level heat caps prevent one bad feed, one tweak, or one headline from wrecking the month. Diversification is practical, not decorative: spread by ticker, sector, and strategy so correlation spikes don’t blindside the book. Above all, Matt DeLong treats trading like a business—documented rules, measured risk, consistent reporting—so scale comes from a durable track record, not a lucky week.
