Tools and comparisons

Manual trading vs automated trading: should you keep trading by hand?

Trading by hand gives you judgment and full control. It also costs you sleep, consistency and, the research says, money. Here is what automation fixes, what it breaks, which tasks each one wins, and a middle path that keeps you in charge.

[

November 15, 2026

]
•
[

14

 

min to read

]

The short answer

Neither wins everywhere. Trading by hand keeps your judgment and full control, but decades of research on individual traders show the cost: trading too often, selling winners early and holding losers. Automation runs a rule the same way every time, even while you sleep, yet it can’t fix a bad rule. The middle path: you think, software watches.

What does trading by hand really involve?

Manual trading means you make every decision and place every order yourself. On a chart it looks simple. Over a full day it is a job with many small steps, and every one of them depends on you being there and being sharp.

A day of manual trading, step by step
WhenWhat you doWhat it takes
Before you tradeRead the trend on the daily chart, the structure on the 4-hour and the entry on the 1-hour or 15-minute. Check the news and what is due that day.Time, and the same routine every morning
Finding a setupWait for price to reach your level or for an indicator to signal, often on several markets at onceHours of attention, or alerts you set and trust
EnteringWork out the size from your stop distance, place the order, set the stop and the targetA calm head at the moment the setup appears
ManagingMove the stop as price moves, take part of the profit, decide whether a pullback is noise or a warningSticking to your rules while your money moves
After the sessionLog the trade in your journal: why you took it, how you felt, what you would changeHonesty, on the losing days too
OvernightHold, close, or leave orders and hope they are enoughSleep, which the market doesn’t respect

Crypto never closes, and stock perps on Hyperliquid can trade around the clock too (our guide to trading stocks on Hyperliquid explains how). When you are asleep, at work or on a plane, your positions stay open and your next setup may come and go without you. Whatever edge a manual trader has only works while the trader is at the screen.

Where does manual trading break down?

Every manual trader knows the missed move and the stop moved “just this once.” Research on individual traders turns those moments into numbers. Most of it comes from brokerage records, where researchers can see what real people did with real money.

Trading too much

Brad Barber and Terrance Odean studied 66,465 US households at a large discount broker from 1991 to 1996. The households that traded most earned 11.4% a year while the market returned 17.9%, and the average household turned over 75% of its portfolio each year. Before trading costs, frequent and infrequent traders did about the same; the cost of all that trading made the difference. The authors point to overconfidence, and their title is the conclusion: trading is hazardous to your wealth.

Selling winners, holding losers

The disposition effect is the habit of selling winning positions too soon and holding losing ones too long. Odean tested it on 10,000 brokerage accounts from 1987 to 1993. On days they sold, investors cashed in 14.8% of their winning positions but only 9.8% of their losing ones. It didn’t pay: over the next year, the winners they sold beat the losers they kept by 3.4 percentage points, measured against the market.

Chasing losses

Professionals do it too. Joshua Coval and Tyler Shumway studied proprietary traders at the Chicago Board of Trade, people who trade for a living, and found them highly loss-averse: after a losing morning, they regularly took above-average risk in the afternoon to win it back.

Trading what’s in the news

Attention is limited, so people buy what catches it. Barber and Odean found that individual investors are net buyers of attention-grabbing stocks: stocks in the news, stocks with unusually high trading volume and stocks with extreme one-day moves.

Day trading for a living

The hardest version is day trading, opening and closing positions within the same day. In Taiwan, where researchers could follow every day trader from 1992 to 2006, fewer than 1% beat the market predictably and reliably once fees were paid. The SEC’s guide to day trading warns that day traders “typically suffer severe financial losses in their first months of trading, and many never graduate to profit-making status,” and that they “must watch the market continuously.”

None of this says you can’t trade well by hand. It says the usual problem isn’t a lack of ideas. It is execution: too many trades, exits driven by feelings, decisions that change with how tired you are, and a plan that only exists while you are watching.

What does automated trading fix, and what can’t it fix?

Automated trading means software places the orders when conditions you defined are met. It can be a bot that runs one fixed rule, or an AI trading agent that also reads the market and explains what it sees. Either way, it goes after the execution problems above.

What it fixes

  • Watching. The rule is checked while you sleep, work or travel, on every market you gave it.
  • Consistency. The same setup gets the same entry, size and exit every time. That is also what makes a rule testable.
  • Exits. A stop loss or take profit set in advance fires whether or not you feel like selling, so the disposition effect gets no say.
  • Freezing and revenge. A rule doesn’t hesitate after three losses or double its size to win them back.

There is evidence that taking the person out of the moment helps. A study of a robo-advising tool, software that builds a portfolio around an investor’s holdings and preferences, found that it reduced, but did not remove, behaviors such as the disposition effect and trend chasing among investors who adopted it. It carries a warning too: investors who were already well diversified traded more after adopting it.

What it can’t fix

  • A bad idea. Automation carries out a rule; it doesn’t make the rule good. A losing strategy run perfectly loses money on schedule.
  • A mistake in the code. On August 1, 2012, after new code was deployed incorrectly, a routing system at the brokerage firm Knight Capital sent more than 4 million orders in the first 45 minutes of trading while trying to fill 212 customer orders. Knight lost more than $460 million, according to the SEC. Software does exactly what it is told, at machine speed.
  • A backtest that fooled you. A backtest replays a rule on past prices. Tune a rule until it looks great on the past and it often fails on new data; our guide to backtesting a strategy covers the traps.
  • A market that changed. A rule built for a sideways market keeps trading it after a trend starts. The CFTC, the US regulator for futures and other derivatives, says in its advisory on trading bots that “SI technology,” its short form for super intelligence, “can’t predict the future or sudden market changes.”
  • New things to break. Connections drop, keys leak, a setting gets mistyped. You swap one set of human errors for a smaller, different one.

Manual or automated: which wins at each trading task?

Here is the comparison job by job. “Automated” covers both plain bots and AI agents; where they differ, the row says so.

Manual vs automated trading, task by task
TaskBy handAutomatedVerdict
Reading the marketThe charts you have time for, on the timeframes you checkA plain bot sees only the indicators it was built with; an AI agent can read several timeframes and markets in secondsTie: an agent covers more ground, you decide what matters
Finding why it movedSearch the news, check related markets, form a viewPlain bots don’t; an AI agent can search the web and show its sourcesAgent, if it cites what it found
Testing an ideaScroll back through charts, or skip itA backtest replays the exact rule on past dataAutomated
SizingYou work out the size from the stop, under pressureThe size you set is applied every timeAutomated, once your size rule is right
EntriesEarly, late, or missed while you were awayWhen the rule fires, every timeAutomated for written rules; by hand for setups you can’t write down
ExitsOpen to the disposition effect: sell winners early, hold losersStops and targets fire as setAutomated
Watching while you sleepYou can’tIt keeps goingAutomated
DisciplineDepends on the day and the last tradeThe rule doesn’t change unless you change itAutomated, though you can still switch it off at the worst moment
Judgment in unusual marketsYou notice that today is different: a halt, a shock, an outageA rule can’t tell; it needs limits that pause itBy hand
LearningEvery trade teaches you about the market and about yourselfYou learn less from trades you didn’t place, unless you review themBy hand
ControlEvery order is yoursOrders go out on their own once you switch it onBy hand
SetupNothing beyond your broker or exchangeRules, tests, limits and connections to set upBy hand
CostTrading costs and your timeTrading costs plus the software, though many tools have free tiersBy hand on money, automated on time

The pattern is plain. Automation wins the mechanical jobs: testing, entering, exiting, watching and sticking to the plan. Trading by hand wins the human ones: judgment when the market does something new, learning, control and simplicity. And the studies above found the costliest mistakes in the mechanical jobs: trading too often and exiting on feelings.

If you are weighing a plain rule bot against an AI agent rather than against your own hands, that is a different question, covered in trading bots vs AI agents. If you already use ChatGPT or Claude for trading ideas, see Ethy vs ChatGPT and Ethy vs Claude.

Is there a middle path between manual and automated trading?

Yes, and for anyone who trades part-time it is often the sensible one: you do the thinking, and software does the watching, the testing and the rule-following. You decide what to trade and why. The machine makes sure the plan runs the same way at 3 a.m. as at 3 p.m.

That split is what Ethy is built for. Ethy is an AI trading agent you ask before you trade, and the work divides like this:

  1. You form the idea, with a second pair of eyes

    Ask Ethy to read a chart and it looks at the daily, 4-hour, 1-hour and 15-minute charts together. When a market jumps, it searches the web for the reason and shows you the sources.

  2. You check it against your own rules

    Ethy keeps your trading plan, your rules on risk, markets, process and goals, and reads it first in every chat. If an idea breaks the plan, it says so. It never keeps a risky habit, such as trading without a stop, as one of your rules.

  3. Ethy tests it

    Describe the rule in plain English. Ethy shows you the rule as it understood it, then backtests that exact rule on up to 365 days of past data, with every entry and exit marked on the chart.

  4. Ethy watches

    Price and indicator alerts, such as a level, an RSI reading or a moving-average cross, reach you in the app and on Telegram.

  5. Ethy runs the rule, inside your limits

    An automation can trade the rule with a take profit, a stop loss and a trailing stop, a stop that follows the price as it moves your way. You can add a maximum daily loss and a maximum drawdown, the biggest drop from a peak you will accept; a pause after a run of losing trades is on by default. Before you activate it, Ethy shows what one stop-out would cost.

  6. You review and decide

    Every live trade sends you a notification. Start in paper mode, which uses live prices and moves no money, then go live when you are ready. To stop, archive the automation and close its positions in one step.

Automations run on Ethy’s servers, so they keep going while you are offline, and the stop loss and take profit sit on Hyperliquid itself as native orders, so they fire even when Ethy is closed. On Hyperliquid they trade perpetual futures (perps, contracts with no expiry date that track a price): crypto on its main exchange, and stocks, indices, commodities and FX on trade.xyz.

Ethy is software for research and analysis, and it doesn’t give investment advice. Its backtests are estimates on past data, not a promise of future results, and the decision to switch a rule on stays yours.

How do you move from manual to automated trading?

Don’t automate everything at once. Move one rule, watch it closely, and expand only what earns your trust.

  1. Pick one rule you already follow

    The best first candidate is a rule you trade often and can write in one sentence: the entry, the exit, the stop and the size. If you can’t write it down, keep trading it by hand for now.

  2. Write it so nothing can misread it

    Name the market, the timeframe, the indicator settings, whether you enter at the close or the next open, how much per trade, and what ends the trade. Our guide to algorithmic trading for beginners shows one rule written three ways.

  3. Test it on past data

    Look at the worst losing run and the deepest drawdown, not just the total. Then test a different period or market to see if it holds.

  4. Run it on paper

    Paper trading runs the rule on live prices with pretend money, so you see how it behaves in today’s market before anything is at stake.

  5. Go live small

    Use a size you could lose without changing your plans, and set the limits first: a stop on every trade, a daily loss cap and a pause after a losing streak.

  6. Review it every week

    Compare what the rule did with what you would have done, and keep a journal for the automated trades too. Change one setting at a time, and only for a reason you can write down.

Keep trading by hand what you can’t write down: the news trade, the unusual market, the setup you read from years of screen time. Over time, the rules that hold up move to the machine, and your attention goes to the decisions only you can make.

Summary

  • Trading by hand gives you judgment and full control, but it only works while you are at the screen, rested and calm.
  • Research points at execution as the leak: the most active households earned 11.4% a year against the market’s 17.9% (1991 to 1996), and investors sold winners far more readily than losers.
  • Automation fixes watching, consistency and exits. It can’t fix a bad idea, and it follows a bad rule just as faithfully as a good one.
  • Trading by hand still wins on judgment in unusual markets, learning, control and setup.
  • A middle path suits many part-time traders: you do the thinking, an agent watches, tests and runs the rule inside limits you set.
  • Start with one rule: write it down, backtest it, run it on paper, go live small and review it every week.

Questions people ask

Is automated trading better than manual trading?

For the mechanical jobs, usually yes: testing a rule, entering and exiting on time, and watching while you sleep. For judgment, learning and control, trading by hand still wins. Automation doesn’t make a strategy profitable; it makes it consistent. A rule that loses money by hand will lose money automated too.

Do automated traders make more money than manual traders?

We found no reliable public study that compares the two head to head. What research does show is that individual traders lose money by trading too often and exiting on emotion, which automation can reduce. Be wary of pages that quote average returns for each without a source, and judge any rule by its own tests and its worst drawdown.

What are the pros and cons of automated trading?

Pros: it watches markets you can’t, follows the rule the same way every time, enters and exits on schedule and makes the rule testable. Cons: it can’t judge an unusual market, it repeats a bad rule perfectly, a backtest can mislead you, and it adds new things to break, such as connections, keys and settings. The losses are still yours.

Can AI trade better than a human?

At some jobs. An AI agent reads more charts, faster, and doesn’t get tired or scared. It is not a reliable price predictor, and the CFTC’s advisory on trading bots warns that super intelligence (SI) technology can’t predict the future or sudden market changes. Use AI to read, test and watch, and keep the judgment for yourself.

Should beginners trade manually before automating?

Placing some small trades by hand first helps you understand orders, stops, slippage and your own reactions. You don’t need years of it, though. Writing your rules down and backtesting them early teaches you faster than trading on feel, and paper trading lets you try automation without risking money.

Can I automate only part of my trading?

Yes, and it is a sensible way to start. Automate the exits first: a stop loss and a take profit on every position, or a trailing stop, so the disposition effect has less room. Next come alerts for the setups you watch, then one fully automated rule. Discretionary trades can stay manual for as long as you like.

Sources

  1. Barber and Odean, Trading Is Hazardous to Your Wealth (The Journal of Finance, April 2000)faculty.haas.berkeley.edu
  2. Odean, Are Investors Reluctant to Realize Their Losses? (The Journal of Finance, October 1998)faculty.haas.berkeley.edu
  3. Coval and Shumway, Do Behavioral Biases Affect Prices? (The Journal of Finance, February 2005)doi.org
  4. Barber and Odean, All That Glitters: The Effect of Attention and News on the Buying Behavior of Individual and Institutional Investors (The Review of Financial Studies, 2008)faculty.haas.berkeley.edu
  5. Barber, Lee, Liu and Odean, The Cross-Section of Speculator Skill: Evidence from Day Trading (Journal of Financial Markets, 2014)faculty.haas.berkeley.edu
  6. D’Acunto, Prabhala and Rossi, The Promises and Pitfalls of Robo-Advising (The Review of Financial Studies, May 2019)doi.org
  7. SEC: Day Trading: Your Dollars at Risksec.gov
  8. SEC: SEC Charges Knight Capital With Violations of Market Access Rule (press release 2013-222, October 16, 2013)sec.gov
  9. CFTC: Customer advisory on trading bots (“SI Won’t Turn Trading Bots into Money Machines”)cftc.gov

Trading involves risk, including the loss of your money. Ethy AI is a research and trading tool and does not give investment advice or recommendations. Backtests are simulations on historical data and do not predict future results.

Ethy blog

Read more

The $ETHY ecosystem

Stop watching charts.
Start living.

Deploy your AI trading agent in one click.