CFD or futures — how to choose

What this lesson resolves

This lesson answers one question: through which of two wrappers for the same exposure — an exchange-traded contract or a CFD with a broker — will you implement your plan cheaper with your capital, horizon, and style. It does not decide whether CFDs are “worse,” because that is a question of paper ownership, supervision, and the compensation scheme; in the general dimension, it is sorted by the lesson Broker or CFD. Here remains the calculation: funding and commissions on one side, exchange and clearing fees and roll cost on the other, and at the end session hours, tick, and tax settlement.

On the chart, a CFD on an index and futures on the same index draw a similar shape. Underneath, they are two products with two cost structures, and only one of them includes funding charged for every day of holding.

Two cost baskets — what you are actually comparing

Each row of this table is a line item in a cost estimate, not a topic for an “on the alert” assessment. Two things follow from it immediately. First, on the CFD side, cost grows with holding time, while on the futures side it is mostly independent of the number of days — which is why horizon is more important in this comparison than the product itself. Second, the futures side has a component that is not visible in the daily accounting: infrastructure fees and roll. If you hold the position for less than one series, there is no roll, but spread and round-turn commission remain.

Formula: cost on your numbers in your chosen horizon

Count both sides separately, then reduce the result to two units. The first is a fraction of 1R, i.e. cost referenced to position risk. The second is a percentage of position value, i.e. cost referenced to capital engaged in the exposure. The first unit tells you whether the strategy has a chance after costs; the second tells you whether the exposure is worth taking with this product at all.

# position_value_USD = price × point_value_USD × number_of_contracts
# take commissions, rates, and fees from your broker's table

# CFD side
daily_funding_USD = position_value_USD × funding_rate_annual / 365
cost_CFD_fixed_USD      = spread_USD + commissions_USD
cost_CFD_USD            = cost_CFD_fixed_USD + daily_funding_USD × days
cost_CFD_R              = cost_CFD_USD / 1R
cost_CFD_percent        = cost_CFD_USD / position_value_USD × 100

# futures side
cost_round_turn_USD  = spread_pts × point_value_USD × contracts + commission_USD × 2
cost_roll_USD       = calendar spread + close/open commissions
cost_futures_USD     = cost_round_turn_USD + cost_roll_USD + exchange_fees_USD
cost_futures_R       = cost_futures_USD / 1R
cost_futures_percent = cost_futures_USD / position_value_USD × 100

Funding rates, commissions, exchange fees, and point value must be checked with your broker — they change together with the product, and some are charged only on trading days. The funding calculation method is just as important as the rate itself: the same annual rate gives a different cost with daily accrual than with accrual on trading days only. Before entering a number into the formula, establish the accrual basis. The roll cost formula is in the lesson Rolling contracts, and the calendar spread mechanics — in Contango and backwardation.

Example on illustrative numbers: twenty days

All numbers below are illustrative numbers. They show the sequence of actions, not the offer of any broker.

Position value: $40,000. Horizon: 20 days. Your 1R: $400.

CFD:     spread 0.05% of value = 20 USD
         commissions           = 0 USD
         funding rate 10% annual → 40,000 × 0.10 / 365 = 10.96 USD/day
         cost_CFD = 20 + 10.96 × 20 = 239.20 USD  →  0.60% of position  →  approx. 0.60R

Futures: spread and round-turn commission = 30 USD
         exchange and clearing fees       = 10 USD
         roll cost                        = 0 USD (horizon fits in one series)
         cost_futures = 30 + 10 = 40 USD  →  0.10% of position  →  0.10R

The conclusion is not “futures are cheaper.” It is: with this horizon and this funding rate, the CFD holding cost is several times higher. Change the funding rate, the horizon, or the position value, and the result may reverse — which is why you do the calculation on your own data, not on someone else’s set.

Breakeven horizon — the number that reverses the decision

Instead of asking which product is cheaper in general, calculate from which day the other product becomes cheaper. This number is independent of opinions, because it results exclusively from costs:

breakeven_days = (cost_futures_USD − cost_CFD_fixed_USD) / daily_funding_USD

For the numbers from the previous section: (40 − 20) / 10.96 ≈ 1.8 days. In other words, CFD comes out cheaper only if the position lives less than two days; with longer holding, the futures side takes the lead. This explains why the same product is sometimes suitable for a different trading style, and why the question “CFD or futures” without a horizon is incomplete. If the planned horizon exceeds the roll moment, add cost_roll_USD to the numerator — then breakeven_days shifts to the right, i.e. in CFD’s favour.

Session hours and tick: two microstructure parameters

Session hours decide whether your plan is feasible at all. An exchange-traded contract has a session set by the exchange, with breaks and holidays that change every year. A CFD product with a broker may be quoted longer, but outside the window in which the underlying market is liquid, spread rises and you pay for the broker’s quote. Therefore, record hours in one zone — Europe/Warsaw — and convert them twice a year, when the time changes. Confirm hours with your broker and in the exchange calendar, not in a screenshot from last year.

Tick works differently than intuition suggests. A finer tick improves stop precision, but by itself it does not lower cost. Cost is decided by the spread in points referenced to stop distance, i.e. spread_cost_R = spread_pts / stop_distance_pts, and by point value converting that distance to USD. A market with a fine tick and wide spread can be more expensive than a market with a coarse tick and tight spread. Check tick and point value with your broker, and the calculation method is covered in the lesson on indices and commodities.

Contract availability and product continuity

On the CFD side, availability depends on the broker’s product list and can change without warning: an instrument may be withdrawn, and quote hours shortened. That is convenience, but also operational risk — the decision to close a position can be forced by a change in the offering, not by your plan.

On the futures side, availability looks different. There is always a specific month with an expiry date, and liquidity transfers to the next series over time — hence roll and its cost. There is no “continuous” contract; there is a continuous chart that someone built from individual series. Before comparing, check two things: whether your broker supports the given series and whether the required deposit fits the buffer you maintain on the account. Check deposit, commissions, and fees with your broker — these are numbers assigned to the instrument and the account, not to the market name.

Taxes and formalities: who to ask about what

Settlement of the result differs on both sides. This is not about the rate itself, but about who is responsible for what. The rate does not replace a tax advisor — this fragment gives three questions whose answer you must know before the first profit, not after:

1. Who is the payer. Does the broker issue only PIT-8C, or does it settle for you — and whether a withholding tax occurs at all with your product. The difference is practical: with two accounts at two brokers, it is easy to miss one event.
2. What document do you receive after a year and whether it covers all transactions — including those on instruments that the broker withdrew during the year.
3. Where does the result go and whether income from both products is described by the same declaration. Ask your broker this question and check current official materials.

Demo account transactions are not tax events, so they do not go into tax documents or the return.

The settlement mechanics on a brokerage account are described in the lesson Withholding tax and brokerage costs, and the definition is collected in the entry Withholding tax. Do not transfer the rule from one product to another: a different product means a different way of documenting and a different person responsible for collection. Confirm current rules with your broker and in official sources.

Comparison protocol — six steps

Order matters: steps 1–2 set the framework, steps 3–5 give numbers, step 6 closes the decision.

1. Set one underlying instrument and one holding horizon in days. Without a horizon, the comparison has no development — it decides which basket wins.
2. Determine position value in USD for both variants. For futures, it is price × point_value_USD × number_of_contracts, for CFD — the nominal value indicated by the broker. Without this number, percentages do not exist.
3. Collect CFD-side costs: spread, commissions, and the annual funding rate together with the accrual basis. Check with your broker — these are three separate numbers, not one “commission.”
4. Collect futures-side costs: spread, per-contract commission, exchange and clearinghouse fees, and the planned roll on this horizon.
5. Calculate both costs and reduce them to R and to percentage of position value. Put both units side by side; one without the other is misleading.
6. Calculate breakeven_days and record the verdict together with the horizon at which it reverses. Only such an answer can be checked later in the journal.

Cost worksheet — fill it in yourself

After eight rows, calculate the median of the “Difference (% of position)” column and check whether its sign is constant. If one side wins once and the other at the same horizon, it is not the product that decides the result, but quote volatility — and then you go back to steps 3 and 4 to close the missing data. The “Repeated measurement” column has the same job as in the other worksheets in this track: a number that does not reproduce on the second read is not a cost, it is an artifact.

When choice says “do not trade”

1. You do not know the funding rate or its accrual basis. A position with unknown holding cost is a position whose cost you do not control.
2. Spread in R exceeds the budget recorded in your plan. Thresholds of 0.25–0.30R are illustrative numbers; set your own threshold on your own sample and record it before comparing products.
3. The plan assumes holding longer than breakeven_days, and you still choose the more expensive side. That is an internal contradiction in the plan, not a matter of taste. Either shorten the horizon or change the product.
4. You cannot point to a liquid futures series or the CFD product has disappeared from the broker’s offering. A product that cannot be read or that may be withdrawn without notice is not the basis of a plan.

Common mistakes

• Comparing commission without funding. On the CFD side, holding cost grows with every day; omitting it inflates the product’s attractiveness.
• Comparing cost in USD without reference. The same amount at a different position size means a different decision — that is why R and percentage of position value count.
• Assuming that no commission means no cost. Cost is often in the spread, and with a short horizon, spread is often the largest component of the calculation.
• Comparing two brokers instead of two products. You are then changing two things at once and do not know what influenced the result.
• Ignoring session hours. A position opened outside the liquid window costs more on entry already.
• Mixing OTC and exchange rules in one playbook. The rule “I hold until the end of the week” does not mean the same thing on a continuous product and on a contract with an expiry date.
• Treating taxes as a year-end topic. It is a layer that changes the net result, and the questions must be asked of the broker before the first profit.

Measurable exercise

Take one underlying instrument to which you have access both as a CFD with your broker and as a futures contract. Choose a holding horizon — for example, twenty days — and perform steps 1–6 of the protocol. Record position value, both costs, both results in R and in percentages, and breakeven_days. At the end, add one sentence: whether your typical position holding time falls below or above this boundary. If you cannot estimate it, first calculate the average holding time from your journal — an answer without this number is an opinion.

Common questions

Are CFDs worse than futures?

Not in an absolute sense and not in this part of the course. CFDs are often more convenient to start with, because they require less capital and give a continuous product without roll. Futures are often clearer microstructurally and cheaper with longer holding. Legal and supervisory differences — paper ownership, compensation scheme — are covered in the lesson on brokerage accounts and CFDs. Here, the decision is resolved by calculation on your numbers.

Where to check funding rate and exchange fees?

In the instrument table and broker documents, such as the specification and the key information document. Funding rates and commissions change over time, so note them together with the read date. Do not base the calculation on a number memorized from a video.

Can I do this measurement on a demo account?

Yes, and that is the right place for the first measurement, because it does not risk capital. Remember, however, that demo is often more favorable than live — spread and slippage are often milder there. In the worksheet, record whether the number comes from demo or live; mixing both sources in one column will ruin the median.

Why calculate cost in R and in percentages of position value, instead of comparing rates alone?

Because the rate does not tell you whether the cost is large. The same commission amount is irrelevant on a large position and deadly on a small one, and the same spread in points means a different cost on markets with different point values. R references cost to the risk you take on the transaction, and percentage of position value — to the capital engaged in the exposure. Only together do they show whether the strategy's advantage survives after costs, and let you compare two products with different notional. Without this, you compare price lists, not impact on result. Confirm all rates, commissions, and fees with your broker, because they change together with the product and the market.