MidasIn short: A DCA bot with a volume scale above 1 puts more of its wins at stake: in the example one deep fall erases 1,013 of them, more than 22 times the 45 of an equal-order grid, although at equal money it loses no more dollars than the grid.
This applies to the grid bots and martingale or DCA-style bots of exchanges and platforms such as Binance, Bybit, OKX, 3Commas, Veles and Pionex, and to grid expert advisors on MT4 and MT5. The inputs are round made-up numbers, not a forecast and not advice.
On this page (7)
What are a grid bot and a DCA bot?
A grid bot trades a price range with many buy and sell orders; a DCA bot averages down a falling position with safety orders and sells above its average price.
- Grid bot: orders at fixed steps between a lower and an upper price. Each filled buy has a sell one step higher, so every swing banks one step. Order sizes are usually equal.
- DCA bot: one base order, then safety orders when price falls by set percentages. Common settings are base order size, number of safety orders, price deviation, volume scale and step scale. The whole position closes at a profit target above its average entry.
The line between them is the size rule. A DCA bot with volume scale 1 behaves like an equal-order grid that closes as one basket. A grid with a multiplier above 1 is a DCA bot with another name; see grid vs martingale vs averaging.
How does each one build exposure?
An equal-order grid adds exposure in a straight line, N × q; a DCA bot with volume scale c multiplies it at every safety order.
The DCA total is q × (c^N − 1) / (c − 1). With c = 2 and ten orders it is 1,023 times the first order. The settings screen shows only the first order, which is why two bots with the same "order size" can differ a hundredfold in position.
Worked example: same settings, then same money
Ladder: price 100, ten levels 100 … 82, first order 0.1 coin, a win means $2 per coin, fees ignored.
| Grid bot (equal orders) | DCA bot (scale ×2) | |
|---|---|---|
| Order sizes (coins) | 0.1 each | 0.1, 0.2, 0.4 … 51.2 |
| Total volume | 1.0 | 102.3 |
| Cost of all 10 orders | $91 | $8,591 |
| Average entry | 91.00 | 83.98 |
| Climb from the last level to break-even | +9.00 | +1.98 |
| Loss when the last level (82) fills | $9.00 | $202.60 |
| Extra loss per $1 further down | $1.00 | $102.30 |
| Win when only the first order fills (0.1 × $2) | $0.20 | $0.20 |
| Wins that one full run erases | 9.00 / 0.20 = 45 | 202.60 / 0.20 = 1,013 |
That compares equal settings, not equal money. Now give the grid the same $8,591: 8,591 / 910 (the sum of the ten prices) = 9.44 coins per level.
| Price reached | 94 | 90 | 86 | 82 | 70 |
|---|---|---|---|---|---|
| Grid, 9.44 coins per level: loss | $113 | $283 | $529 | $850 | $1,983 |
| DCA ×2 from 0.1 coin: loss | $2 | $11 | $49 | $203 | $1,430 |
At equal money the doubling ladder loses less at every depth above zero, because it buys more at lower prices. Its costs are elsewhere. A single-order win is $0.20 against $18.88 for the grid (9.44 × $2), so one full run still erases 1,013 wins against 45. Nearly half of its money, 4,198.4 / 8,591.2 = 49%, goes into the last order, which only fills when the fall is deep. Both lose the whole $8,591 if price goes to zero.
Where does liquidation come from?
Liquidation comes from leverage on a position that is large relative to the deposit, whichever bot built it: the exchange closes the position when the loss uses up the margin.
- Grid bot: the position grows steadily as price slides through the range. A long grid below its lower bound just sits on the full position.
- DCA bot: the position stays small through the first fall, then multiplies. The margin needed jumps with each safety order, and once the last safety order is used the bot waits for price to return to the profit target.
- Without leverage (spot) nothing is liquidated; the loss is the fall in the value of what is held, up to its whole cost.
Grid bot vs DCA bot: comparison
| Grid bot (equal orders) | DCA bot (volume scale above 1) | |
|---|---|---|
| Exposure growth | Straight line | Multiplies each safety order |
| What the settings screen shows | The real size, roughly | A small first order hiding a large total |
| Typical win | Small, repeated | Small, relative to the total |
| Loss shape | Steady from the first fall | Quiet at first, steep at the end |
| Failure mode | Price leaves the range | Last safety order used, price keeps falling |
| What to compute | N × q, range, leverage | q × (c^N − 1) / (c − 1), average entry, leverage |
This is arithmetic of the settings, not a forecast. Trading with leverage carries risk: if the deposit is too small, a stop-out is possible. Size your deposit with a safety margin. Nothing here is individual investment advice.
Check your own numbers
- Grid bot risk calculator for exchange and platform grid bots.
- DCA bot calculator (safety orders) for base and safety orders.
- Gold grid calculator for MT5 gold grid EAs.
- Midas Grid emulator replays a gold grid EA's settings on real XAUUSD ticks since January 2025. It does not test crypto bots.
Related: is grid trading safe, how much deposit a grid bot needs, martingale blow-up arithmetic.
FAQ
Is a grid bot safer than a DCA bot?
Per setting, usually yes: equal orders add position in a straight line, and a DCA bot with volume scale above 1 multiplies it. At equal money in the example the doubling ladder loses less above zero but earns far less per win, so one deep fall erases 1,013 wins against 45.
Is a DCA bot the same as a martingale?
When the safety orders grow in size, yes in effect: a volume scale above 1 is a martingale size rule. With a scale of 1 the bot averages down with equal orders and total exposure grows in a straight line.
Which bot is better for a small deposit?
Neither by name. Compute the total position of your settings: q × (c^N − 1) / (c − 1) for a DCA bot, N × q for a grid. A small first order can hide a large total; compare the total with your deposit and the minimum order of your exchange.
Can a grid bot be liquidated like a DCA bot?
Yes, when it uses futures with leverage. A long grid is closed out when the floating loss uses up the margin, usually after price falls below the range. Spot bots are not liquidated, but they can hold a falling asset down to a large loss.
Which has the bigger maximum loss?
Both can lose their whole cost, which is the same $8,591 in the equal-money example. The difference is in how quickly the loss builds and how many small wins it takes to repay.
Can the Midas Grid emulator test a DCA or grid bot on an exchange?
No. It replays the settings of a gold (XAUUSD) grid expert advisor on real ticks. For exchange bots use the grid bot and DCA bot calculators.