The screen says “grid profit: $5.” The account started at $500 and is now worth $455. Those numbers can both be correct.
A DCA bot versus grid bot comparison should track remaining inventory and total equity, not just completed trades. A price-triggered DCA cycle combines entries under its exit rule; a grid pairs orders across configured levels. Either can retain losing holdings after showing profitable completed sales. The exact order and funding rules determine the exposure.
The constructed path below makes that distinction visible without forecasting which approach performs better. Every number is an educational example, not a replay of market prices or a product performance claim.
Describe the rules before comparing the labels
For this article, “DCA” means price-triggered averaging, not equal purchases on a calendar. There are variants of both DCA and grid trading. Some use adaptive levels, partial exits, signal filters, trailing logic or automatic restarts. A comparison is meaningful only after those details are fixed.
Binance's Spot DCA explanation contrasts multiple entries with a common round exit against individually paired grid entries and exits. Its Spot Grid explanation describes orders arranged within a configured range. Those descriptions establish the distinction, not identical behavior across every implementation.
Our two invented rules are:
- DCA rule: buy 1 unit at $100 and another at $90. After both fill, offer the entire 2-unit position at 2% above its $95 average, or $96.90. No further additions or restart occur before a full exit.
- Paired-level grid rule: buy 1 unit at $100 for an associated sale at $105, and buy 1 at $90 for an associated sale at $95. When the $90-to-$95 pair completes, replace its $90 purchase order. There are no other levels.
This miniature two-pair construction illustrates accounting. It does not reproduce a venue's complete grid setup, initial inventory allocation or order-normalization rules. KLINEO's DCA product is the context for the comparison; this article makes no claim that KLINEO offers this particular grid implementation.
Follow the same prices through both rules
Both begin with $500 in quote cash and no asset units. Assume a price path that moves from $100 down to $90, recovers to $95, then falls through $90 to $70. Each eligible order fills completely at the stated price before the path continues.
There are no fees, spread, slippage, partial fills, leverage, borrowing, taxes or quote-currency changes. Cash in the ledger means total quote cash, including any amount locked for accepted orders; locks are not deducted again.
| Price event | DCA state | Grid state |
|---|---|---|
| First purchase at $100 | $400 cash, 1 unit | $400 cash, 1 unit |
| Purchase at $90 | $310 cash, 2 units | $310 cash, 2 units |
| Recovery to $95 | No exit: target is $96.90 | Sell the $90 unit at $95; $405 cash, 1 unit |
| Fall back through $90 | No additional purchase | Replacement $90 buy fills; $315 cash, 2 units |
| Final mark at $70 | $310 cash + $140 holdings | $315 cash + $140 holdings |
The DCA construction ends at $450, a $50 decrease from $500. The grid ends at $455, a $45 decrease. Its completed $90-to-$95 pair earned $5 before costs, but its remaining two units cost $190 and are marked at $140: a $50 open loss.
Grid total change = $5 realized − $50 unrealized = −$45
Grid equity = $315 quote cash + 2 × $70 = $455
This is the central accounting lesson. A positive matched-pair number does not establish a profitable account. The example also does not prove the grid is superior because it lost $5 less on one deliberately chosen path.
Why the small recovery produces different actions
At $95, the DCA construction's entire position has recovered to its pre-fee average. Its specified 2% target remains above that level, so it does not close. The grid instead sells one paired unit, which changes cash and inventory before the next decline.
If the path had continued to $96.90 before falling, the DCA rule would have closed under the exact-fill assumption. If it had reached $105, another grid sale would have become eligible. If it had fallen directly to $70 without the $95 recovery, the completed $5 grid pair would not exist.
Those are different path conditions, not reasons to predict a favorable one. A comparison needs to examine the paths that fail each rule as well as the one that makes its reporting look attractive.
The DCA take-profit worksheet explains how target basis and fees change the cash result. The same caution applies to grid pairs: a nominal spread is not automatically the net matched profit after actual costs.
A range boundary does not remove holdings
A grid with its lowest purchase at $90 can still own units when price reaches $70. The lack of additional orders below the range does not make the assets stop losing value. Conversely, a rising market can sell available inventory and leave a strategy with cash while the asset continues upward.
A DCA ladder also has an exhaustion point. After the final addition, it holds its resulting quantity under the remaining exit rules. Calling one approach suitable for a range and another for a decline is too coarse to establish a funding or loss decision.
The persistent-trend example follows the point at which a finite averaging ladder becomes an ordinary holding. Use a comparable continuation beyond a grid boundary, rather than ending an evaluation exactly where the configured orders end.
This is also why expanding the range or adding another averaging level mid-deal is a new capital decision. It changes the exposure you originally approved. A lower revised average is not proof that the extra commitment was warranted.
Compare budgets and units as carefully as exits
The miniature example uses equal unit purchases at the same prices, so the initial inventory is directly comparable. A real comparison often does not. One bot may deploy most of its allocation at startup while another retains substantial quote cash for later levels.
Record the initial quote balance, initial asset inventory, maximum intended purchases, exit quantities and what happens to proceeds after a completed pair or cycle. If one rule reinvests profits and the other does not, identify that difference. A starting “$500 investment” does not necessarily mean both hold the same market exposure.
The capital-requirements guide supplies a ledger for committed and remaining purchase cash. For a grid, add the inventory assigned to sales and the cash that replacement buys can reuse. Do not add every future theoretical transaction as though it requires permanently separate cash; do not assume reuse before a sale has actually produced funds either.
A fair report should show allocated cash and deployed purchase cost separately. Return on deployed capital can look high while much of the allocation remains idle. Return on starting equity answers a different question.
Test the execution assumptions that could change the ledger
The exact-price path assumes the orders exist and fill in sequence. A candle that touches $90 and $95 does not prove that a purchase preceded a sale, especially if both lie inside one interval. Queue position, timing and partial fills can change the quantities available at the next event.
A replacement purchase must also be accepted after its paired sale. If it is delayed or rejected, the grid ends the example with a different inventory. A DCA exit submitted for an outdated quantity can likewise leave residual exposure or require reconciliation.
Use the backtest methodology guide to distinguish observed candles from modeled order sequences. A strategy ranking that depends on an undisclosed order inside an ambiguous candle is less informative than a clearly labeled uncertainty range.
Finally, specify what happens when you stop either strategy. Canceling pending orders and selling existing holdings are separate actions. The pause-versus-close guide gives a practical way to inspect both sides.
Choose a comparison question you can actually answer
Rather than asking which label makes more money, ask: under these funded rules, how do cash, inventory and equity evolve on the same paths? Then include costs, unsuccessful fills and a continuation beyond the configured range. Keep any uncertainty in the report.
The worked example is spot without borrowing. Futures grids and futures averaging introduce leverage, collateral, liquidation and funding that this ledger does not model. Stops and software limits also cannot guarantee a maximum loss through gaps or outages. Read the risk disclosure before connecting a real account.
Editorial method: This AI-assisted comparison uses current primary venue documentation and an independently checked synthetic inventory ledger. It reports no historical test, live trading result or evidence that either strategy has a return advantage.