Early access is open. Free sample datasets are live now. Download a sample

Data product · Early access

Continuous Futures Data with Back-Adjusted Rolls

One long price series per futures market, with the rolls handled the way you choose, so a ten-year ES or CL backtest doesn't trade fake gaps at every expiry.

Coverage

What you get

Item Detail
Product code FUT-CONTINUOUS
Coverage Continuous series for the futures roots we carry, front month (.C1) and deferred positions (.C2 and up)
Adjustments none, back_adjusted (Panama, additive) and ratio_adjusted, all three for every root
Roll rules volume, open_interest, first_notice, fixed_days_before_expiry
Granularity Daily on Starter; 1-minute and up on Pro
History Same depth as our futures bars for each root
Delivery REST and CSV or Parquet files, with a roll calendar
Update speed End of day

A single futures contract trades actively for a few weeks or months. Stitch contracts together raw and every roll shows up as a price jump that never happened in your account. Back-adjusting removes the jump. Ratio-adjusting keeps percentage returns correct. We give you all three and tell you exactly which contract sits under each bar.

Schema

Schema

Field Type Required Description
symbol string yes Continuous series ID: root + .C + position, e.g. ES.C1 (front), ES.C2 (second)
root string yes e.g. ES
adjustment string yes none, back_adjusted (Panama, additive), ratio_adjusted
roll_rule string yes volume, open_interest, first_notice, fixed_days_before_expiry
interval string yes 1m to 1d
timestamp / date string yes As in futures bars
open, high, low, close number yes Adjusted prices
volume number yes Volume of the underlying contract in that bar
open_interest number optional (daily) Open interest of the underlying contract
underlying_symbol string yes Contract used for this bar, e.g. ESZ6
unadjusted_close number optional Raw close of the underlying contract
cumulative_adjustment number optional Total offset or ratio applied so far
is_roll_date boolean optional True on the bar where the series switched contract

Example record (illustrative values):

{
  "symbol": "ES.C1",
  "root": "ES",
  "adjustment": "back_adjusted",
  "roll_rule": "volume",
  "interval": "1d",
  "date": "2026-11-18",
  "open": 6718.50,
  "high": 6761.25,
  "low": 6702.00,
  "close": 6742.25,
  "volume": 1384216,
  "open_interest": 2051376,
  "underlying_symbol": "ESZ6",
  "unadjusted_close": 6742.25,
  "cumulative_adjustment": 0.0,
  "is_roll_date": false
}

Sample data

Sample data

No free continuous futures sample yet. CME prices need a licence, so we can't publish a real one for free. A sample file is available on request: tell us the root, the adjustment and the roll rule, and we'll send it in this schema.

Use cases

Use cases

Trend-following backtests. Use ratio_adjusted close for returns and signals over many years. Ratio adjustment keeps percentage moves intact, which matters for a 20-year CL test where an additive offset could push old prices below zero.

Point-based P&L. For a system that trades a fixed number of ES points, use back_adjusted. Price differences stay in real points, so your stop distances and tick values still make sense.

Rebuilding real fills. underlying_symbol and unadjusted_close tell you which contract you would actually have held and at what price. When is_roll_date is true, book a roll trade in your simulator and charge the spread.

Checking the roll itself. Pull the same root with two roll_rule values and compare the results. If the backtest changes a lot, your edge may live in the roll, not the signal. For physically delivered contracts such as CL, the first_notice rule keeps you out of the delivery window.

Code

Code examples

Example: daily ES front month, back-adjusted, rolled on volume.

curl "https://api.marketdata.co/v1/futures/continuous?symbol=ES.C1&adjustment=back_adjusted&roll_rule=volume&interval=1d" \
  -H "Authorization: Bearer $MARKETDATA_API_KEY"

Example: Python, list the roll dates and the gap each one removed.

import os
import requests
import pandas as pd

resp = requests.get(
    "https://api.marketdata.co/v1/futures/continuous",
    params={"symbol": "CL.C1", "adjustment": "back_adjusted",
            "roll_rule": "first_notice", "interval": "1d"},
    headers={"Authorization": f"Bearer {os.environ['MARKETDATA_API_KEY']}"},
)
resp.raise_for_status()
df = pd.DataFrame(resp.json()["results"])
rolls = df[df["is_roll_date"]]
print(rolls[["date", "underlying_symbol", "cumulative_adjustment"]])

Pricing

Pricing

Plan Continuous futures
Starter, $29/mo Daily series in all three adjustments
Pro, $79/mo Adds intraday series and a custom roll rule parameter

Dataset: $29 per root for full daily history in all three adjustments, roll calendar included. A 100-root bundle is planned at $490. Elsewhere, monthly plans for continuous series run from $22.50 to $59.95, and one-off files from $50 to $330.

Side by side

How we compare

6 of the 50 providers we studied sell continuous futures, mostly as files or desktop software. Several give you one roll method or no adjustment at all. We compute every series from our own bars, publish all three adjustments with the underlying contract on each row, and list the price. A $29 file per root is yours to keep.

FAQ

Frequently asked questions

Which adjustment should I use?

Use ratio_adjusted for percentage returns and long backtests, back_adjusted when you think in points or ticks, and none when you want raw prices and plan to handle rolls yourself. All three come with every root.

When does the series roll?

That depends on roll_rule. volume rolls when the next contract trades more, open_interest when it holds more open positions, first_notice before the first notice day, and fixed_days_before_expiry a set number of days before the last trade date.

Can I see the real contract prices?

Yes. underlying_symbol names the contract behind each bar and unadjusted_close gives its raw close, so you can always map back to tradable prices.

What time zone are the bars in?

Intraday timestamp values are UTC. Daily bars use the exchange trade date, so a Globex session that opens Sunday evening counts toward Monday.

Is the data real-time?

No. Continuous series update at end of day. Real-time futures bars are a separate product on Business.

What formats can I get?

JSON from the API, and CSV or Parquet for datasets. Each dataset includes the roll calendar as a separate file.

Keep reading