Skip to main content

NotAnotherSMAOffsetStrategyHO: Strategy Explained

1. What Does This Strategy Do?

1.1 One-Line Summary

This is a strategy specifically designed to find pullback buying opportunities in uptrends, like buying famous brands on sale — trend is good, just temporarily discounted, grab it!

1.2 What Does the Name Mean?

NotAnotherSMAOffsetStrategyHO broken down:

  • NotAnother: Don't underestimate me, I'm not just any strategy
  • SMA: Simple Moving Average (though actually uses EMA)
  • Offset: Offset — not only looking at whether price reaches MA, but by how much it deviates
  • HO: Likely "High Optimization" — heavily optimized version

Simply put, this is a "use MA to find bargains" strategy.


2. Core Principles in Plain English

2.1 Strategy's Core Idea

Imagine you're climbing a mountain. This strategy: wait for rest stops (price pullback), then you climb up to grab bargains.

It doesn't buy at the peak (too expensive), doesn't wait at the bottom (may catch a falling knife), but chooses to buy at the rest stop halfway up.

2.2 What is "Offset"?

This is the strategy's smartest feature!

Regular MA strategy: Buy when price crosses above MA

  • Problem: By the time it crosses, price has already risen; you bought expensive

Offset strategy: Wait below MA to buy

  • Set offset coefficient 0.966, meaning buy when price falls to 96.6% of MA
  • Equivalent to 3.4% off — like buying pants at 96.6 yuan when tagged 100 yuan

2.3 Why Three Buy Signals?

Like exams have multiple choice, judgment, and fill-in — different question types test different knowledge.

Three signals:

  • ewo1: Normal pullback, straightforward buy
  • ewo2: Deep pullback, brave bottom-fishing
  • ewolow: Market panic, contrarian

No matter how the market moves, there's a corresponding tactic.

2.4 What is EWO?

EWO (Elliott Wave Oscillator) is simply the difference between two MAs:

  • A fast MA (quick)
  • A slow MA (slow)

When fast MA is above slow MA: EWO is positive, market rising When fast MA is below slow MA: EWO is negative, market falling

The strategy uses EWO to judge: Is the major trend up or down


3. What Indicators Are Used?

3.1 Indicator Overview

IndicatorRoleAnalogy
EMATrend line, major directionHighway center line
HMAFast-reacting trend lineGas pedal, go when pressed
SMAShort-term trendSpeedometer, shows speed
RSIOverbought/oversold, sentimentThermometer, market fever or cold
EWOTrend strengthCompass, direction finding
VolumeTrading volumeFuel gauge, how much gas in tank

3.2 Three RSI Periods

RSI is a 0-100 number showing if market is "overheated" or "too cold."

The strategy uses three RSIs:

  • RSI_4 (fast): Like a stopwatch, reacts in seconds
  • RSI_14 (standard): Like a clock, normal time-reading
  • RSI_20 (slow): Like a calendar, sees long-term changes

Why three? Like weather — check current temperature (RSI_4), today's average (RSI_14), this week's trend (RSI_20).

3.3 EMA vs SMA

  • SMA: Equal weight to all prices, arithmetic average
  • EMA: Recent data weighted more, faster reaction

Strategy uses EMA as primary judgment because it's faster.

3.4 What is HMA?

HMA (Hull Moving Average) is both fast AND smooth.

Regular MAs are either fast but jittery, or smooth but slow. HMA is both. Strategy uses it to judge trend reversal because HMA reacts fast and catches problems early.


4. When to Buy?

4.1 Three Keys to Buy

SignalMarket SituationGeneral Meaning
ewo1Bull normal pullbackClimbing, rest for a bit
ewo2Deep pullbackDropped deep, bottom-fish
ewolowBear endFell to the bottom, wait for rebound

4.2 Key 1: ewo1 (Main Buy Signal)

Scenario: Climbing a mountain, was steadily going up, suddenly tired and rested. ewo1 says: Now resting, will climb again later, buy now!

Conditions:

  1. Fast RSI < 35: Market panicked, everyone selling
  2. Price < MA's 96.6%: 3.4% cheaper
  3. EWO > 3.422: Major trend still up, just temporarily fell
  4. RSI < 52: Not overheated yet, okay to buy
  5. Volume > 0: Not a dead market

Win rate: Highest, because following the trend

Risk: Medium, pullback may continue

4.3 Key 2: ewo2 (Deep Pullback)

Scenario: This climb not only tired but slipped, fell pretty deep. ewo2 says: Fallen this much, near the bottom, go!

Conditions:

  1. Fast RSI < 35: Market panicked
  2. Price < MA's 95.1%: ~5% cheaper
  3. EWO > -4.58: Fell but not catastrophically
  4. RSI < 52: Not overheated
  5. RSI < 25: Extremely oversold
  6. Volume > 0

Win rate: Medium

Risk: Higher, bottom-fishing can fail

4.4 Key 3: ewolow (Reversal Signal)

Scenario: Not climbing anymore, fell to the valley bottom. ewolow says: Can't fall further, prepare for reversal!

Conditions:

  1. Fast RSI < 35: Market panicked
  2. Price < MA's 96.6%: Relatively cheap
  3. EWO < -8.562: Fell terribly
  4. Volume > 0

Win rate: Lower

Risk: Highest, catching a falling knife

4.5 Signal Comparison

SignalWin RateExpected ReturnRiskBest Scenario
ewo1HighMediumMediumBull pullback
ewo2MediumHigherHigherDeep pullback
ewolowLowHighestHighestBear reversal

Advice: Beginners use ewo1 more; experts can try ewo2 and ewolow.


5. When to Sell?

5.1 Two Sell Signals

SignalMeaningAnalogy
Condition 1Rose to target, time to sellArrived at stop
Condition 2Can't rise anymore, running awayLeaking, patch it fast

5.2 Condition 1: Rose to Target, Sell

Conditions:

  1. Price > short-term MA (SMA_9): Short-term trend up
  2. Price > sell MA's 100.2%: Above baseline
  3. RSI > 50: Market still strong
  4. Fast RSI > Slow RSI: Momentum upward
  5. Volume > 0

Plain English: Rose to the planned position, although may rise more, strategy suggests taking the money and running, don't be greedy.

5.3 Condition 2: Trend May Reverse, Get Out

Conditions:

  1. Price < HMA (fast MA): Broke below fast trend line
  2. Price > sell MA's 101.4%: Still profitable
  3. Fast RSI > Slow RSI: Momentum still up (possibly fake)

Plain English: Rose to high ground but starting to fall, trend may change, protect your profits now.

5.4 Satisfy Either, Sell

The two conditions are OR — selling if either is met. Benefits:

  • Condition 1 catches upwardmarkets
  • Condition 2 catches reversal signals
  • Hard to miss sell opportunities

6. How is Risk Controlled?

6.1 Stop-Loss: 35% Hard Bottom Line

stoploss = -0.35

Bought at $100, drops to $65 → forced sell.

Why 35%? Crypto swings 10% in a day normally. Too tight gets stopped by normal fluctuation. 35% is balanced.

Newbie advice: If 35% feels harsh, change to 25% or 30% yourself.

6.2 Trailing Stop: Follow Price, Lock Profits

How it works:

  1. After buying, don't rush
  2. Wait for price to rise 3%, trailing stop activates
  3. Stop line follows highest price, maintaining 0.5% distance
  4. Price retraces to stop line → auto-sell

Example:

  • Bought at $100
  • Rose to $103 (3% gain), trailing stop activates
  • Rose to $110, stop line at $109.45 ($110 × 99.5%)
  • Dropped to $109.45 → auto-sell, locked 9.45% profit

Benefit: Whatever it rises, locks in more.

6.3 Tiered Take-Profit

Holding TimeProfit TargetMeaning
0 min28.9%Rose 28.9% immediately, sell now
25 min10.5%Held 25 min, 10.5% is fine
80 min3.8%Held 80 min, 3.8% is fine
140 min0%Held 140 min, sell even if breakeven

Design logic:

  • Just bought with a big bounce → run, don't be greedy
  • Held longer → lower expectations
  • Held over 140 min still not profitable → you're wrong, get out

6.4 Only Sell When Profitable

Strategy sets sell_profit_only = Truedon't sell losing trades!

Unless triggered 35% stop-loss, sell signals don't execute in loss. Like: bought stock at $100, down $10, sell signal rang but strategy says "wait, that'd be cutting losses."

6.5 Risk Control Summary

MeasureTriggerEffect
Fixed stop-lossLoss 35%Survival bottom line
Trailing stopProfit 3% activatesLocks profits
Tiered ROITime + profit ratioControls greed
Profit-only sellProfit >1%Avoid cutting losses
Slippage protectionSlip >2% rejectControls costs

7. What Do Parameters Mean?

7.1 Buy Parameters

ParameterValueMeaningAdjustment
base_nb_candles_buy16MA period for buyingLarger = smoother, smaller = more sensitive
low_offset0.966Main offsetLower = need bigger discount, fewer signals
low_offset_20.951Backup offsetFor bottom-fishing
ewo_high3.422ewo1 thresholdPositive means must be in uptrend
ewo_high_2-4.58ewo2 thresholdExpanded buy opportunities
ewo_low-8.562ewolow thresholdFor reversal signals
rsi_buy52RSI thresholdLower = more conservative

7.2 Sell Parameters

ParameterValueMeaning
base_nb_candles_sell10Sell MA period (faster than buying)
high_offset1.014Price 1.4% above MA to consider selling
high_offset_21.002Price 0.2% above MA to consider selling

Why sell faster than buy?

  • Buy MA: 16 candles → slow, more observation
  • Sell MA: 10 candles → fast, don't be greedy

8. Strategy Pros & Cons

8.1 Pros

1. Triple buy signals, many opportunities

  • Bull has ewo1
  • Deep pullback has ewo2
  • Bear reversal has ewolow
  • Whatever the market, there's a play

2. Offset design, better entry prices

  • Not chasing buys, waiting for discounts
  • 3-5% cheaper than crossing MA

3. Comprehensive risk control

  • Stop-loss, trailing stop, tiered take-profit
  • Slippage protection too
  • Multiple insurance policies

4. Good for optimization

  • Many adjustable parameters
  • Can optimize for different coins and markets

5. Clear labels

  • Every buy tagged (ewo1/ewo2/ewolow)
  • Backtesting shows which signal makes the most money

8.2 Cons

1. Stop-loss too wide

  • 35% loss for newbies is harsh
  • One bad trade loses a third

2. Too many parameters

  • Newcomers don't know how to adjust
  • Wrong parameters = poor results

3. Ranging market losses

  • If prices oscillate, MA useless
  • Frequent buys/sells, eats fees

4. Long only, no shorting

  • Makes money in bull, sits idle in bear
  • Misses bear market opportunities

5. No volume change analysis

  • Only checks volume > 0
  • Doesn't analyze if volume increasing or decreasing

8.3 How to Maximize Strengths, Minimize Weaknesses

Handle wide stop-loss:

  • Reduce stop-loss yourself, e.g., to 25%
  • Or reduce position size, invest less per trade

Handle parameter overload:

  • Start with default parameters in backtest
  • Then optimize slowly, one parameter at a time

Handle ranging markets:

  • Use strategy's built-in filters
  • Or manually judge — don't run during ranging

Handle long-only limitation:

  • Pair with another shorting strategy
  • Or only run strategy when major trend is up

9. Who Is This For?

9.1 Ideal User Profile

Suitable if you:

  • Have some quantitative trading basics
  • Can handle moderate risk
  • Have 2-4 hours daily to monitor
  • Want short-to-medium term trades
  • Interested in BTC, ETH, and major coins

Not suitable if you:

  • Know nothing about quantitative trading
  • Risk-averse (losing 10% keeps you up at night)
  • Don't have time to monitor
  • Want long-term investments
  • Only play small coins

9.2 Newbie Advice

  1. Paper trade for one month first — use virtual money, watch performance, don't rush live
  2. Start with low-risk parameters — only use ewo1 signal, 10% position, 25% stop-loss
  3. Daily review — analyze every trade's entry/exit, why profit, why loss, find patterns
  4. Control emotions — strategy says buy → buy, strategy says sell → sell, don't manually intervene

9.3 Capital Allocation

CapitalSuggested Per-Trade PositionReason
< $5000Not recommendedFee proportion too high
$5000-$2000020%Under 5 trades, controllable risk
$20000-$10000015%Diversify across 6-7 trades
> $10000010%Conservative priority

10. Common Questions

Q1: Why called "NotAnother"?

A: Developer's little joke — means "this is not just another ordinary MA strategy." And indeed, this strategy is much more complex and refined than simple MA strategies.

Q2: Is 35% stop-loss too high?

A: For crypto, 35% isn't really high. Bitcoin swings 10% in a day normally. Too tight gets stopped out by normal fluctuation.

Advice: Adjust to 20-25% if you can't handle it.

Q3: Why 5-minute candles?

A: 5 minutes is the balance between short and medium:

  • Shorter (1 min): Too noisy, too many false signals
  • Longer (1 hr): Too few signals, miss too many opportunities

Advice: Newcomers can start with 15 minutes.

Q4: Which of the three buy signals should I use?

A: Suggested order:

  1. Newcomers: only ewo1
  2. After familiar: add ewo2
  3. Experts only: ewolow

Q5: Can this be used on stocks?

A: In theory, but note:

  • Stocks less volatile than crypto
  • Need parameter adjustments
  • Stop-loss can be smaller (e.g., 10%)

Q6: How many trades per day?

A: Depends on volatility:

  • High volatility: 3-5 per day possible
  • Low volatility: may go all day without a signal
  • Average: 1-2 per day

Q7: How are fees calculated?

A: Per trade:

  • Buy: one fee
  • Sell: one fee
  • One round trip: two fees

Assuming 0.1% fee, buying $10,000:

  • Buy fee: $10
  • Sell fee: $10 (assuming still selling $10,000)
  • Total cost: $20

Advice: Use maker orders (limit orders) for lower fees.

Q8: Backtest great, live bad — why?

A: Common issue, reasons:

  • Backtest doesn't consider slippage
  • Backtest fees may be set too low
  • Past data, future may differ

Advice:

  • Discount backtest results by 30%
  • Verify with paper trading
  • Test live with small money first

11. Practical Tips

11.1 Parameter Optimization Tips

Step 1: Run backtest with default parameters

freqtrade backtesting --strategy NotAnotherSMAOffsetStrategyHO

Step 2: Optimize optimizable parameters

freqtrade hyperopt --hyperopt-loss SharpeHyperOptLoss --epochs 500 --spaces buy sell

Step 3: Re-backtest with optimal parameters, compare results

Warning: Don't over-optimize! Perfect historical fit may perform worse going forward.

11.2 Market Judgment Tips

Before running strategy, judge market state:

Market StateHow to JudgeStrategy Adjustment
BullPrice above long-term MACan increase position
RangingOscillating, no clear directionReduce position or don't run
BearPrice below long-term MAOnly use ewolow, reduce position

11.3 Dynamic Stop-Loss Adjustment

Adjust stop-loss by market volatility:

  • High volatility (wildmarkets): Widen stop to 40%
  • Low volatility (quiet consolidation): Tighten stop to 25%

11.4 Multi-Coin Portfolio

Don't run just one coin — run multiple:

CoinRiskExpected Return
BTCLowSteady
ETHMediumMedium
Other majorsHighPotentially high

Recommendation: 60% BTC/ETH, 40% other majors.


12. Comparisons with Other Strategies

12.1 vs Regular MA Strategy

ComparisonThis StrategyRegular MA Strategy
Entry timingOffset entry, position earlyCross entry, chase
Signal countThree signals, many opportunitiesUsually one signal
Risk controlMultiple protectionsUsually just stop-loss
ComplexityRelatively complexSimple

Conclusion: This strategy is more refined, but needs more understanding.

12.2 vs Grid Strategy

ComparisonThis StrategyGrid Strategy
Suitable marketTrendingRanging
Trading frequencyMediumHigh
FeesMediumHigh
Risk controlActivePassive

Conclusion: Use this strategy in trends, use grid in ranging markets.

12.3 vs High-Frequency Strategy

ComparisonThis StrategyHigh-Frequency
Time frame5 minutesSeconds
FrequencyLowExtremely high
Tech requirementMediumExtremely high
Capital thresholdLowHigh

Conclusion: Retail traders use this strategy; institutions use high-frequency.


13. Summary — One Line to Remember

"In an uptrend, wait for price to pull back below MA by a set percentage, buy in three batches, lock profits with trailing stop."

13.1 Key Points

  1. Only trade pullbacks: Not chasing, waiting for dips
  2. Triple signals: Different depths for different markets
  3. Offset entry: Find bargains below MA
  4. Multi-layer risk control: Stop-loss + trailing + ROI
  5. Adjustable parameters: Adapt to different markets

13.2 Pre-Launch Checklist

Before enabling strategy, confirm:

  • Understand strategy principles
  • Run backtest
  • Exchange API set up
  • Capital amount set
  • Stop-loss set
  • Notification methods set
  • Time to monitor available

13.3 Final Thoughts

This strategy is a tool, not a holy grail. Even the best strategy needs:

  • Understanding principles
  • Strict execution
  • Continuous learning
  • Emotion control

Good luck trading!


Document Version: 1.0 For: Quantitative trading beginners to intermediate players