Scott Malkinson On execution
Scott Malkinson · Trading since 2005.
Here’s the game you’re actually playingNo guarantees.
No improvising.
The wrong game
The fantasy
You keep adding conditions because every loss feels like evidence that something is missing. Another indicator. Another rule to keep you out of the trade that just hurt.
The goal becomes a strategy that tells you exactly what will happen. That goal makes you abandon workable ideas in search of impossible certainty.
A losing trade can be perfectly executed. A winning trade can be a complete mistake.
Know exactly what must be true before you act.
Decide what the trade can cost before you enter.
When your conditions appear, execute without negotiation.
Manage the trade by the plan, not by how you feel.
The odds
Expectancy is only half the story
A strategy can make money on paper and still put you out of the game. The question is whether your account, your loss limits, and your discipline can survive the path to that return.
Win rate alone does not tell you how much a strategy earns, or how hard it is to stick with. These five illustrative setups all have positive expectancy before costs, but their average returns and the swings around those averages are very different.
1R is the amount risked on a trade. All ratios are risk:reward: 1:4 risks 1R to make 4R; 2:1 risks 1R to make 0.5R.
These are conservative assumptions, not typical or verified win rates. Risk:reward determines the break-even rate, not the rate a strategy will achieve. The payoffs must reflect realized wins and losses, including partial exits, rather than planned targets alone.
| Metric | 23% WR1:4 | 37% WR1:2 | 44% WR1:1.5 | 55% WR1:1 | 70% WR2:1 |
|---|---|---|---|---|---|
| Win / loss | +4R / −1R | +2R / −1R | +1.5R / −1R | +1R / −1R | +0.5R / −1R |
| Break-even win rate | 20% | 33.33% | 40% | 50% | 66.67% |
| Expectancy per trade | +0.15R | +0.11R | +0.10R | +0.10R | +0.05R |
| Expected result / 100 trades | +15R | +11R | +10R | +10R | +5R |
| Profit factor | 1.19 | 1.17 | 1.18 | 1.22 | 1.17 |
| Standard deviation per trade | 2.10R | 1.45R | 1.24R | 0.99R | 0.69R |
| Profitable after 100 trades | 71.9% | 76.4% | 75.9% | 81.7% | 77.9% |
| At least | 23% WR1:4 | 37% WR1:2 | 44% WR1:1.5 | 55% WR1:1 | 70% WR2:1 |
|---|---|---|---|---|---|
| 4 losses | >99.99% | 99.98% | 99.6% | 91.5% | 43.3% |
| 5 losses | >99.99% | 99.0% | 93.6% | 64.7% | 15.3% |
| 6 losses | 99.96% | 93.0% | 76.2% | 36.3% | 4.8% |
| 7 losses | 99.5% | 79.2% | 53.6% | 17.9% | 1.4% |
| 8 losses | 97.7% | 61.1% | 34.0% | 8.4% | 0.43% |
| 10 losses | 85.9% | 29.6% | 11.7% | 1.7% | 0.04% |
WR means win rate. Streak odds mean at least one run of that many consecutive losses anywhere in 100 trades, not just the next few trades.
That same streak hurts every setup equally at the same sizing. What changes is how often it shows up: about 85.9% for the 23% win-rate setup, versus 1.71% for the 55% setup over 100 trades. Risk a fixed 1% of starting equity instead, and ten losses cost exactly 10%, before fees and slippage.
The smallest trade-to-trade swings: 70% at 2:1. Its standard deviation is just 0.69R, and long losing streaks are least likely. But each win pays only 0.5R. Expectancy is a thin +0.05R per trade, so average costs of 0.05R per trade would erase the edge. Its chance of finishing 100 trades profitable is only 77.9%. Low volatility does not guarantee profit, especially once trading costs are included.
The highest expected return: 23% at 1:4. It averages +0.15R per trade, or +15R over 100 trades before costs. But its standard deviation is the largest at 2.10R, and its chance of finishing ahead is the lowest here at 71.9%. Larger winners come with more frequent losses in this example.
A steadier balance in this model: 55% at 1:1. It has the same +0.10R expectancy as 44% at 1:1.5, with smaller swings: 0.99R versus 1.24R. Its probability of profit after 100 trades is the highest here at 81.7%. The 37% at 1:2 example earns slightly more at +0.11R per trade, with greater volatility. These are comparisons of assumed outcomes, not evidence that one strategy is easier to find or trade.
Expectancy is win probability × reward minus loss probability × risk. For 44% at 1:1.5: (0.44 × 1.5R) − (0.56 × 1R) = +0.10R. This is an average across possible outcomes, not a promised result. Subtract average costs in R to get net expectancy: a 0.05R cost reduces +0.10R to +0.05R per trade. Break-even win rate is 1 ÷ (1 + reward in R). Profit factor divides expected gross wins by expected gross losses. Standard deviation measures how widely individual results spread around the average.
Ruin means hitting the point where you can no longer continue. That might be an account loss limit, a margin requirement, or your own capital floor. Its probability depends on that boundary, position sizing, and the sequence of all trades. A mix of wins and losses can also breach a limit. The streak table alone cannot tell you your chance of ruin.
These comparisons only matter if the stated win rates and payoffs hold up on unseen data. Choose a return profile you can execute, then size for the losses you may have to sit through, not just the return you hope to collect.
Model assumptions: 100 independent trades, unchanged win probabilities, exact stated payoffs, and no costs. Profitability uses a fixed cash value of 1R and means finishing above zero, not breaking even. These are calculated examples, not observed trading results. Real losses can cluster and edges can change. A smoother model does not prove a more robust strategy.
Method: binomial probabilities for final profit; exact run probabilities for losing streaks. References: NIST on the binomial distribution and CME on position sizing.
Here’s the part nobody wants to admit.
Live trade log
From the channel
Closed in Loss PYPL Entry: 53.22000 Exit: 52.68000 R: -0.06R
View post in TelegramClosed in Loss UBER Entry: 69.09000 Exit: 68.10000 R: -0.11R
View post in TelegramClosed in Profit PINS Entry: 18.60000 Exit: 19.19000 R: +0.09R
View post in TelegramThese posts document personal trades and are not trade signals, recommendations, or invitations to copy positions. The sole purpose is to demonstrate that my advice isn't BS.
Facts over hope
Scott’s advice
What no one tells you about risk:reward is that it doesn’t just affect how much you win or lose. It can also affect when those wins and losses show up.
Take a strategy with a 1:5 risk:reward ratio. You’re risking 1R to make 5R.
Assuming price moves at roughly the same pace in either direction, your stop is much closer than your profit target. A losing trade can therefore finish relatively quickly, while a winning trade may need considerably more time to reach its target.
That creates an interesting problem when you first start testing a strategy.
Imagine putting on a dozen trades. Some of the trades that will eventually become winners may still be open because they have much farther to travel. Meanwhile, the losers can hit their stops quickly and disappear from the book.
You look at your results and see: Loss. Loss. Loss. Loss.
The natural reaction is:
“This strategy doesn’t work.”
So you abandon it, change the rules, add another indicator, or start searching for an entirely new strategy.
But the strategy may not have been broken at all. You simply evaluated it before enough of the slower-moving winners had time to finish.
This is one reason I like using 1:1 risk:reward when initially testing an idea.
With the stop and target roughly the same distance from entry, you reduce some of the timing asymmetry between winners and losers. You’re not deliberately making the profit target five times farther away than the stop.
That gives you a cleaner starting point for answering the most important question:
Once you establish that an edge exists, then you can start experimenting with risk:reward.
Try 0.5R. Try 1R. Try 2R. Try 3R.
See how each version changes expectancy, win rate, drawdown, holding time, and the shape of the equity curve.
Risk:reward should be something you optimize after finding an edge, not something that disguises whether an edge exists in the first place.
Scott Malkinson
Get in touch
A direct line
Write to me on Telegram at @ScottMalkinson