Most retail crypto traders only know one direction: buy low, sell high. That limits them to profiting only when markets rise. Professional traders and well-configured bots operate in both directions — going long when prices are expected to rise and going short when they're expected to fall.
This distinction is fundamental. A bot that can only go long sits idle during bear markets, or worse, keeps buying into a sustained downtrend. A bot that can go short capitalizes on those same drops. Combining both directions through hedge mode creates strategies that can extract profit in virtually any market condition.
This guide covers the mechanics of long and short bot strategies on crypto futures, the mathematics of leverage and liquidation, the hidden cost of funding rates, and the practical frameworks for deciding which direction to run — complete with real calculations using current market prices.
Key Takeaways
- Long bots profit from price increases by buying dips and selling recoveries; short bots profit from price decreases by selling rallies and covering on drops.
- Perpetual futures contracts enable short selling and leverage — but introduce liquidation risk, funding rate costs, and margin management that spot trading doesn't have.
- Leverage of 2-3x is the sweet spot for bot trading: enough amplification to matter, but liquidation prices remain far enough away to survive normal volatility.
- Funding rates of 0.01% per 8 hours compound to 10.95% annually — a significant hidden cost that erodes profits on directional positions held for weeks.
- Hedge mode (simultaneous long + short positions) enables delta-neutral strategies that profit from volatility regardless of market direction.
- Cross margin uses your entire account as collateral (higher liquidation buffer); isolated margin limits risk to the specific position's allocated margin.
The dollar amounts and percentages in this guide are worked examples, not recommendations. Crypto trading can lose money — size every bot with funds you can afford to lose, and read the full Risk Disclosure before going live.
Spot Trading vs Futures: The Foundation
Before discussing long and short strategies, you need to understand the two trading environments where bots operate. The differences aren't cosmetic — they fundamentally change your risk profile, capital efficiency, and strategy options.
Spot Trading
In spot trading, you buy and own the actual asset. If you buy 0.5 ETH at $3,500, you hold 0.5 ETH in your wallet. Your maximum loss is 100% (ETH goes to zero), and you can't lose more than you invested. There's no leverage, no liquidation, no expiry, and no funding fees.
Spot trading bots can only go long — you buy the asset expecting the price to rise, then sell it later at a profit. You cannot natively short on spot markets (some exchanges offer spot margin, but it's effectively borrowing).
Perpetual Futures Contracts
Perpetual futures (perps) are derivative contracts that track the price of the underlying asset without owning it. You don't hold BTC — you hold a contract that pays or charges you based on BTC's price movement. Key differences from spot:
- Leverage: You can control a $10,000 position with $1,000 of your own capital (10x leverage)
- Short selling: You can profit from falling prices by opening a short position
- Liquidation: If your position moves far enough against you, the exchange closes it automatically and you lose your margin
- Funding rates: Periodic payments between longs and shorts that anchor the futures price to the spot price
- No ownership: You never hold the underlying asset
| Feature | Spot Trading | Perpetual Futures |
|---|---|---|
| Asset Ownership | Yes — you hold the actual crypto | No — you hold a price-tracking contract |
| Direction | Long only (buy → sell) | Long or Short (profit in either direction) |
| Leverage | 1x (no leverage) | 1x to 125x (exchange dependent) |
| Maximum Loss | 100% of invested capital | 100% of margin (liquidation), potentially more with cross margin |
| Liquidation Risk | None | Yes — position auto-closed at liquidation price |
| Funding Fees | None | Every 8 hours (positive or negative) |
| Capital Efficiency | Low — full asset cost required | High — only margin required |
| Expiry Date | None — hold indefinitely | None (perpetual) — but funding costs accumulate |
| Best For Bots | Conservative DCA, accumulation | Directional trading, hedging, leverage strategies |
| Complexity | Simple | Significantly more complex |
Futures trading adds multiple layers of risk that don't exist in spot: liquidation, funding rate drag, and the temptation of excessive leverage. If you're new to bot trading, start with spot strategies on established pairs before moving to futures. Read our Complete Guide to Crypto Trading Bots first.
Long Bot Mechanics
A long bot profits when prices go up. Its core logic: buy the asset (or open a long futures position), accumulate more on dips via DCA safety orders, and sell when the price recovers above your average entry.
How Long DCA Bots Work
- Entry: Bot opens a long position at the current price (base order)
- Averaging on dips: As the price drops, safety orders buy more at lower prices, pulling down the VWAP (Volume-Weighted Average Price)
- Take profit on recovery: When the price rises above the VWAP by the take-profit percentage, the bot closes the entire position
- Restart: Bot opens a new deal and repeats
Where Long Bots Excel
Long bots thrive in two market conditions:
Uptrends with pullbacks: BTC trending from $90,000 to $110,000 over 3 months, but with periodic 5-10% dips along the way. Each dip triggers safety orders, each recovery closes a profitable deal. The upward bias means deals close quickly and frequently.
Sideways/range-bound markets: BTC oscillating between $95,000 and $105,000. The bot buys dips near support, sells recoveries as price returns to the upper range. Predictable oscillation creates reliable deal cycles.
Where Long Bots Struggle
Sustained downtrends: If BTC drops from $100,000 to $60,000 over several months without meaningful bounces, a long bot exhausts all safety orders and sits in a deep unrealized loss. The DCA averaging reduces the pain, but it can't prevent losses when the market moves persistently against you.
Example: Long DCA Bot on ETH
- Entry: ETH at $3,500
- Safety orders: 5 orders, 2% step, 1.5 step scale, 1.5 volume scale
- Take profit: 1.8% above VWAP
If ETH drops 8% to $3,220 and then recovers:
| Order | Price | Size | Cumulative Cost | VWAP |
|---|---|---|---|---|
| Base | $3,500 | $100 | $100 | $3,500 |
| SO1 (-2%) | $3,430 | $150 | $250 | $3,458 |
| SO2 (-5%) | $3,325 | $225 | $475 | $3,385 |
| SO3 (-9.5%) | $3,168 | $337.50 | $812.50 | $3,282 |
VWAP after SO3: $3,282. Take profit at $3,282 × 1.018 = $3,341. ETH only needs to recover from $3,168 to $3,341 — a 5.5% bounce, not the full 9.5% back to entry. The bot captures 1.8% profit on $812.50 = $14.63.
Short Bot Mechanics
A short bot profits when prices fall. Its core logic is the mirror image of a long bot: sell (open a short position) when prices rally, add to the short on further rallies (averaging up the entry price), and cover (close the short) when the price drops.
How Short DCA Bots Work
- Entry: Bot opens a short position at the current price (base order sell)
- Averaging on rallies: As the price rises against the short, safety orders add to the short position at higher prices, pulling the average entry price up
- Take profit on drops: When the price falls below the average entry by the take-profit percentage, the bot closes the entire position (buys back)
- Restart: Bot opens a new short deal and repeats
Where Short Bots Excel
Downtrends with bounces: ETH dropping from $3,500 to $2,000 over 3 months, with periodic 5-10% relief rallies. Each rally triggers safety orders, each subsequent drop closes a profitable short deal.
Overbought altcoin corrections: After a parabolic rally, altcoins frequently retrace 20-40%. A short bot on an overextended altcoin captures these corrections systematically.
Where Short Bots Struggle
Bull markets: In a sustained uptrend, a short bot fights the trend. Each safety order adds to a losing short position at increasingly bad prices. If the asset keeps rising, losses compound rapidly — especially with leverage.
Short selling has theoretically unlimited loss potential. When you go long, the worst case is the asset drops to $0 (100% loss). When you go short, there's no ceiling — an asset can rise 200%, 500%, or more. With leverage, even modest adverse moves can trigger liquidation. This isn't just a theoretical risk: BTC has rallied 50%+ in single months multiple times in its history.
Why Short Bots Are Psychologically Harder
Beyond the mathematical asymmetry of unlimited losses, shorting is psychologically challenging for three reasons:
-
Counter-intuitive: Humans naturally think in terms of "buy low, sell high." Profiting from decline feels unnatural and creates cognitive dissonance that leads to premature exits.
-
Bullish bias in crypto: Crypto markets have a long-term upward bias. Shorting means fighting this structural trend, which means your margin for error is smaller and your windows of opportunity are shorter.
-
Rally speed exceeds crash speed: While crashes are often cited as violent, in crypto, short squeezes and relief rallies can be equally explosive. A 15% short squeeze in hours can trigger safety orders and potentially liquidation before you can react.
Leverage for Bots: Why 2-3x Is the Sweet Spot
Leverage amplifies both your profits and your losses proportionally. A 10% move at 3x leverage is a 30% gain or loss on your margin. The critical question for bot trading: how much leverage before your liquidation price is too close to normal market volatility?
Liquidation Price Calculation
For a simplified calculation (ignoring maintenance margin, which varies by exchange):
Long position liquidation price:
Liquidation Price = Entry Price × (1 − 1/Leverage)
Short position liquidation price:
Liquidation Price = Entry Price × (1 + 1/Leverage)
Concrete Examples: BTC at $100,000
| Leverage | Long Liquidation | Distance to Liquidation | Short Liquidation | Distance |
|---|---|---|---|---|
| 2x | $50,000 | -50% | $150,000 | +50% |
| 3x | $66,667 | -33.3% | $133,333 | +33.3% |
| 5x | $80,000 | -20% | $120,000 | +20% |
| 10x | $90,000 | -10% | $110,000 | +10% |
| 20x | $95,000 | -5% | $105,000 | +5% |
| 50x | $98,000 | -2% | $102,000 | +2% |
The numbers tell a clear story:
-
At 3x leverage (long): Liquidation at $66,667, requiring a 33.3% drop. BTC has dropped 33%+ only during major bear markets and black swan events. Your bot has wide room to operate and place safety orders.
-
At 5x leverage (long): Liquidation at $80,000, requiring a 20% drop. BTC drops 20% several times per year. A DCA bot with safety orders spanning 15% of price action has minimal buffer before liquidation.
-
At 10x leverage (long): Liquidation at $90,000, requiring just a 10% drop. BTC experiences 10% corrections almost monthly. Your bot will likely be liquidated within weeks.
-
At 20x+ leverage: You're not trading — you're gambling. A normal day's volatility can liquidate you.
The 2-3x leverage sweet spot gives you enough amplification to meaningfully boost returns (2-3x higher profit per deal) while keeping the liquidation price far enough away that normal market corrections won't trigger it. For DCA bots with safety orders spanning 10-15% of price, 3x leverage means your liquidation is still 18-23% below your deepest safety order.
Accounting for Maintenance Margin
The simplified formula above ignores maintenance margin — the minimum margin ratio the exchange requires to keep your position open. Maintenance margin is typically 0.4-1.0% of position value for BTC on major exchanges. The actual liquidation price is slightly closer than the simplified formula suggests.
For BTC at $100,000 with 3x leverage and 0.5% maintenance margin:
- Simplified liquidation: $66,667
- Actual liquidation (accounting for maintenance margin): approximately $67,200
The difference is small at low leverage but becomes more significant at higher leverage. At 10x, maintenance margin can move your liquidation price 1-2% closer than expected.
Funding Rates: The Hidden Cost of Holding
Funding rates are periodic payments exchanged between long and short traders on perpetual futures. They exist to keep the futures price anchored to the spot price. If the futures price drifts above spot (bullish market), longs pay shorts. If it drifts below spot (bearish market), shorts pay longs.
How Funding Rates Work
- Frequency: Every 8 hours (3 times per day) on most exchanges
- Direction: Positive rate = longs pay shorts; Negative rate = shorts pay longs
- Calculation: Funding payment = Position size × Funding rate
The Real Cost
A typical funding rate of 0.01% per 8 hours seems tiny. But compound it:
| Period | Cumulative Funding Cost |
|---|---|
| 8 hours | 0.01% |
| 24 hours (3 payments) | 0.03% |
| 1 week | 0.21% |
| 1 month | 0.90% |
| 3 months | 2.70% |
| 1 year | 10.95% |
On a $10,000 position at 0.01% funding rate:
- Daily cost: $3.00
- Monthly cost: $90.00
- Annual cost: $1,095
Funding rates aren't fixed — they fluctuate with market sentiment. During extreme bull markets, funding can spike to 0.1% per 8 hours (0.3%/day, 109.5%/year). During crashes, funding goes deeply negative — longs get paid to hold. Bot strategies must account for this variable cost. During high-funding periods, the cost of holding a leveraged long position can exceed the bot's take-profit percentage.
Strategic Implications for Bots
Long bots in bull markets: You typically pay funding (longs pay shorts when market is bullish). If your DCA bot's deals last 2-3 days on average and your take profit is 1.5%, funding cost is roughly 0.06-0.09% — a minor drag. But if a deal stays open for 2 weeks during a stagnant period, funding eats 0.21% of your profit.
Short bots in bear markets: You typically receive funding (shorts get paid when market is bearish). This is a bonus — you profit from the price drop AND get paid funding. This makes short bots more cost-efficient in bear markets than long bots are in bull markets.
Practical rule: If your average deal duration exceeds 5-7 days and you're paying funding, increase your take-profit target by 0.1-0.15% to compensate. If your deals average under 48 hours, funding is negligible.
Cross Margin vs Isolated Margin
How you allocate margin to your futures positions fundamentally changes your risk exposure. Both modes have legitimate use cases for bot trading.
Isolated Margin
Each position has its own dedicated margin. If the position is liquidated, you lose only the margin assigned to that position — the rest of your account is safe.
Example: You have $5,000 in your futures account. You open a BTC long with $1,000 isolated margin at 3x leverage (controlling a $3,000 position). If BTC drops 33% and you get liquidated, you lose the $1,000 margin. Your remaining $4,000 is untouched.
Pros for bots:
- Risk per position is capped and predictable
- Multiple bots can run independently without affecting each other
- One bad trade can't wipe your entire account
Cons for bots:
- Liquidation price is determined only by the assigned margin — no buffer from other funds
- You need to pre-allocate margin for each bot, reducing capital efficiency
Cross Margin
All positions share your entire account balance as collateral. The benefit is that unrealized profits from one position can help support a losing position, effectively widening the gap to liquidation.
Example: You have $5,000 in your futures account. You open a BTC long at 3x leverage worth $3,000. In isolated margin, your liquidation buffer is based only on the $1,000 margin. In cross margin, your entire $5,000 account supports the position — your liquidation price is significantly lower.
Pros for bots:
- Lower effective liquidation price (more room to maneuver)
- Positions automatically use available margin, improving capital efficiency
- Better for running multiple correlated bots that rarely all go wrong simultaneously
Cons for bots:
- A single catastrophic position can drain your entire account
- No firewall between positions — all risk is shared
- Harder to reason about total risk exposure across multiple bots
| Aspect | Isolated Margin | Cross Margin |
|---|---|---|
| Risk Scope | Limited to allocated margin per position | Entire account balance at risk |
| Liquidation Buffer | Only assigned margin | Full account balance |
| Best For | Running multiple independent bots | Single-pair focus or correlated strategies |
| Capital Efficiency | Lower (pre-allocated per bot) | Higher (shared pool) |
| Risk of Ruin | Limited per position | Account-level if extreme move occurs |
| Recommended For | Most bot traders | Experienced traders with position size discipline |
For most bot traders, isolated margin is the safer choice. It turns each bot into a contained risk unit. If one bot gets liquidated, the others keep running. Cross margin is appropriate only when you thoroughly understand your total exposure and have strict position-sizing rules that prevent any single bot from drawing excessive margin.
Hedge Mode: Running Long and Short Simultaneously
Hedge mode allows you to hold both a long and a short position on the same trading pair at the same time. On exchanges without hedge mode, opening a short while holding a long would close (or reduce) the long. Hedge mode keeps them separate.
Why Hedge Mode Matters for Bots
Consider running a long DCA bot and a short DCA bot on BTC/USDT simultaneously. Without hedge mode, their orders would cancel each other out. With hedge mode, each bot operates independently:
- Long bot: Buys dips, takes profit on recoveries
- Short bot: Sells rallies, takes profit on drops
Delta-Neutral Concept
When your long and short positions are equal in size, your net exposure (delta) is zero — you're delta-neutral. You don't care whether BTC goes up or down; you profit from the volatility itself.
How it works in practice:
- Long bot opens $1,000 long at $100,000
- Short bot opens $1,000 short at $100,000
- BTC drops to $95,000: Long is -$50, Short is +$50. Net P&L = $0
- Long bot's safety orders fire, short bot takes profit on the short
- BTC bounces to $98,000: Long bot takes profit on recovery, short bot's safety orders fire on the rally
The beauty: each bot captures its directional moves independently. The long bot profits from bounces; the short bot profits from drops. Over a full oscillation cycle, both bots can close profitable deals.
Limitations of Hedge Mode
- Double the capital: You need margin for both positions, effectively halving your position size per direction
- Double the fees: More trades = more trading fees and funding payments
- Net funding exposure: If you're equally long and short, funding payments roughly cancel out — longs pay and shorts receive (or vice versa)
- Not truly risk-free: If the price trends strongly in one direction, one bot accumulates a large losing position that may exceed the other bot's profits
Real Scenario: BTC Drops 15%
Let's compare how the same $3,000 capital performs across two common setups when BTC drops from $100,000 to $85,000 (-15%) and then partially recovers to $92,000.
Setup A: Long DCA Bot on Spot
- Capital: $3,000
- No leverage
- 5 safety orders, 3% initial deviation, 1.5 step scale, 1.5 volume scale
- Take profit: 1.5%
| Order | Price | Size | Cumulative | VWAP |
|---|---|---|---|---|
| Base | $100,000 | $200 | $200 | $100,000 |
| SO1 (-3%) | $97,000 | $300 | $500 | $98,200 |
| SO2 (-7.5%) | $92,500 | $450 | $950 | $95,620 |
| SO3 (-14.25%) | $85,750 | $675 | $1,625 | $92,070 |
| SO4 (-24.38%) | $75,625 | $1,012 | $2,637 | $88,520 |
At $85,000, SO3 has filled. Position: $1,625 invested, VWAP $92,070. SO4 trigger is at $75,625 — hasn't filled yet.
When BTC reaches $92,000:
- Current P&L: ($92,000 - $92,070) / $92,070 = -0.076% — still slightly underwater
- TP target: $92,070 × 1.015 = $93,451
- Status: Deal still open. Bot waits for further recovery.
When BTC reaches $93,451:
- Bot closes. Profit: $1,625 × 0.015 = $24.38
- Capital was locked for the full drawdown and recovery period
Setup B: Long DCA Bot on 3x Futures (Isolated Margin)
- Capital: $3,000 allocated as margin
- 3x leverage = controls up to $9,000 in position value
- Same parameters adjusted for futures: smaller margin allocation per order, same deviation/scale
- Take profit: 1.5%
| Order | Price | Margin | Position Value (3x) | Cumulative Margin | VWAP |
|---|---|---|---|---|---|
| Base | $100,000 | $150 | $450 | $150 | $100,000 |
| SO1 (-3%) | $97,000 | $225 | $675 | $375 | $98,200 |
| SO2 (-7.5%) | $92,500 | $337.50 | $1,012 | $712.50 | $95,620 |
| SO3 (-14.25%) | $85,750 | $506.25 | $1,519 | $1,218.75 | $92,070 |
Liquidation check: With $1,218.75 margin supporting a $3,656 position at VWAP $92,070, the effective leverage is ~3x. Simplified liquidation at ~$61,380 — well below current price. Safe.
When BTC reaches $92,000:
- Same P&L ratio as spot: -0.076%. Still underwater.
- But position size is 3x larger: unrealized loss is 3x larger too.
When BTC reaches $93,451:
- Bot closes. Profit on leveraged position: $3,656 × 0.015 = $54.84
- Minus funding cost (assume deal open 10 days, 0.01% per 8h): $3,656 × 0.03% × 10 = -$10.97
- Net profit: ~$43.87 vs $24.38 on spot
Key difference: The 3x futures bot earned ~80% more profit ($43.87 vs $24.38), but the position carried liquidation risk the spot bot didn't have. If BTC had continued below $61,380, the futures bot would have been liquidated — total loss of $1,218.75 margin. The spot bot would still hold its BTC, waiting for eventual recovery.
This comparison illustrates the fundamental leverage trade-off: higher returns in normal scenarios, but catastrophic risk in extreme scenarios. The spot bot can always wait for recovery (assuming BTC doesn't go to zero). The futures bot has a point of no return — the liquidation price. For bot trading, this is why moderate leverage (2-3x) is recommended: it provides meaningful return amplification while keeping the liquidation price far from normal market action.
Market Regime Detection: When to Run Which Direction
Choosing the right direction is more important than any parameter optimization. A perfectly configured long bot loses money in a bear market; a perfectly configured short bot loses money in a bull market.
Simple Regime Framework
| Regime | Characteristics | Bot Direction | Confidence Level |
|---|---|---|---|
| Strong Uptrend | Price above 50-day & 200-day MA, higher highs/lows | Long only | High |
| Weak Uptrend | Price above 200-day MA but below 50-day MA | Long with caution | Medium |
| Sideways | Price oscillating around MAs, no clear trend | Long + Short (hedge) | Medium |
| Weak Downtrend | Price below 50-day MA but above 200-day MA | Reduce long, add short | Medium |
| Strong Downtrend | Price below both MAs, lower highs/lows | Short only (or off) | High |
Practical Indicators for Regime Detection
Moving Average Position:
- BTC above 200-day SMA: structurally bullish — favor long bots
- BTC below 200-day SMA: structurally bearish — favor short bots or stay in spot
RSI (14-period on daily):
- RSI > 70: overbought — reduce long exposure, consider short on altcoins
- RSI 40-60: neutral — run both directions
- RSI < 30: oversold — reduce short exposure, favor long entries
Funding Rate Trend:
- Sustained high positive funding (>0.05%): market overheated, long-crowded — short setups become favorable
- Sustained negative funding: market oversold, short-crowded — long setups become favorable
Practical Strategy: "Always Long BTC + Conditional Short ALT"
This paired approach is one of the most practical frameworks for bot traders who want exposure to both directions without the complexity of constant regime analysis.
The Framework
Permanent position — Long BTC DCA bot (spot or low-leverage futures):
- Always running on BTC/USDT
- Conservative configuration: wide safety orders, moderate take profit
- Rationale: BTC has recovered from every drawdown in its history. Long-term structural uptrend makes "always long" viable for BTC specifically
Conditional position — Short Altcoin DCA bot (futures):
- Activated only when specific conditions are met:
- BTC dominance rising (capital flowing from alts to BTC)
- Altcoin has rallied 40%+ in 30 days (overextended)
- RSI (daily) above 75 on the altcoin
- Funding rate on the altcoin is significantly positive (>0.05%)
- Uses isolated margin at 2x leverage
- Strict stop loss at 15-20% above entry
Why This Combination Works
BTC long provides baseline returns. During bull markets, the long BTC bot captures the uptrend. During corrections, it averages down. Historical data shows that holding BTC through any 4-year cycle has always been profitable.
Conditional altcoin shorts capture corrections. Altcoins are more volatile than BTC and tend to give back gains sharply. After parabolic rallies, mean reversion is powerful. The short bot captures these corrections without fighting the structural uptrend.
Negative correlation during corrections. When BTC dominance rises (BTC outperforming), alts usually drop harder. Your BTC long holds steady or dips moderately, while your altcoin short profits from the sharper decline. This natural hedge reduces portfolio-level drawdowns.
Example Capital Allocation
| Component | Capital | Leverage | Effective Exposure |
|---|---|---|---|
| BTC Long DCA (spot) | $2,000 | 1x | $2,000 |
| ALT Short DCA (futures) | $500 margin | 2x | $1,000 |
| Reserve (unused) | $500 | — | — |
| Total Account | $3,000 | — | $3,000 |
The reserve ensures you have buffer capital for unexpected safety order fills or to add to either position if conditions change.
Frequently Asked Questions
Can I run a short bot on spot markets?
No. Short selling requires futures or margin trading. In spot, you can only sell assets you already own. To profit from falling prices, you need to open a short position on a futures contract (perpetual futures on exchanges like Binance, Bybit, or OKX). Some exchanges offer spot margin trading where you borrow the asset to sell it, but this is effectively a futures-like mechanism with borrowing costs.
What happens if funding rates change direction while my bot is running?
Your bot's profitability per deal shifts. If you're long and funding flips from negative (you receive) to positive (you pay), your effective take-profit is reduced by the funding cost. Most well-configured bots with deal durations under 3-5 days are minimally affected by funding rate changes. For longer-duration deals, monitor funding rates and consider adjusting your take-profit target upward to compensate. A 0.01% funding rate adds approximately 0.03% daily cost.
Is hedge mode (long + short simultaneously) risk-free?
Absolutely not. Hedge mode reduces directional risk but introduces other costs and risks. You pay double the trading fees, double the funding exposure (though longs and shorts partially cancel), and use double the capital. In a strong trend, the losing side accumulates a position that may exceed the winning side's profits — especially if one side's safety orders fill heavily while the other side's deals close quickly with small profits.
How do I decide between cross and isolated margin for my bots?
Use isolated margin as the default for bot trading. It contains the blast radius of any single bot's failure. Only consider cross margin if: (a) you're running a single bot on one pair, (b) you thoroughly understand your worst-case exposure, and (c) you want the extra liquidation buffer from your full account balance. Never use cross margin when running multiple independent bots — one bot's liquidation can cascade and drain capital meant for others.
What leverage should I use for a DCA bot on altcoins?
Lower than BTC — typically 2x maximum. Altcoins are more volatile than BTC: a 20-30% drop in an altcoin is common even in neutral markets. At 2x leverage, your liquidation price is 50% below entry, giving your safety orders room to work. At 3x, liquidation is at -33%, which altcoins can reach during moderate corrections. At 5x, liquidation at -20% is practically guaranteed during any significant pullback.
Should I always have both a long and short bot running?
No. Running both directions constantly is only appropriate in sideways markets. In a clear uptrend, your short bot will lose money faster than your long bot makes it. In a clear downtrend, the reverse. Use the regime detection framework above: strong trends = single direction; uncertain/sideways = both directions. The "Always Long BTC + Conditional Short ALT" strategy is a practical compromise that doesn't require constant regime switching.
How do liquidation fees work?
When you get liquidated, the exchange doesn't just close your position at the liquidation price. They typically charge a liquidation fee (0.5-1.5% of position value) on top of your losses. This means your actual loss exceeds what the simplified liquidation formula suggests. On a $3,000 position, a 1% liquidation fee costs an additional $30. This fee incentivizes traders to close losing positions manually before liquidation — which is exactly what a well-configured bot's stop loss should do.