Crude oil and the Wednesday EIA report

Micro Crude Oil (MCL) · 1-minute bars · 2016 to 2026 · Volatility

Worked in sample, failed out of sample: profit factor 1.28 before the split, 0.98 after.

Profit factor, out of sample: 0.98
Average trade, out of sample: −0.01R
Win rate, out of sample: 53%
Trades, 10 years: 516
Max drawdown: $4,314
Pass odds, median plan: 0.2%

Hypothetical results. These results are based on simulated or hypothetical performance results that have certain inherent limitations. Unlike the results shown in an actual performance record, these results do not represent actual trading. Also, because these trades have not actually been executed, these results may have under- or over-compensated for the impact, if any, of certain market factors, such as lack of liquidity. Simulated or hypothetical trading programs in general are also subject to the fact that they are designed with the benefit of hindsight. No representation is being made that any account will or is likely to achieve profits or losses similar to these being shown.

SharePost on X LinkedIn Reddit

Abstract

We backtested the EIA Wednesday breakout rules on Micro Crude Oil (MCL) 1-minute bars from 2016-01-11 to 2026-10-30: 516 trades, with commission and a tick of slippage on every market and stop fill. In sample (to 2023-08-03) it made +0.13R per trade with a profit factor of 1.28; out of sample (from 2023-08-04) −0.01R per trade and 0.98 over 150 trades. 0 of the 16 parameter settings we tried were profitable out of sample. Sized at $250 of risk a trade, simulated 50K evaluations passed 0% to 1% of the time across 36 tracked plans.

1Hypothesis

The government's weekly crude inventory report comes out at 10:30 ET on Wednesdays. Measure crude's range around it, then straddle the pre-report range.

If the report reliably moves crude, a break of the pre-report range on Wednesdays should follow through more than the same rule on other days.

2Data

Instrument: Micro Crude Oil (MCL), traded on the CL price series (CL.v.0: the most traded contract, rolled on volume). CME Globex via Databento.
Period: 2016-01-11 to 2026-10-30: 2,791 sessions, 3,787,721 one-minute bars, built into 3,787,721 1-minute bars around the clock.
Costs: $0.75 commission a contract each way and 1 tick ($1.00) of slippage on every market and stop fill. Limit orders fill when price touches them; market orders fill at the next bar's open and stops at their price, each with one tick of slippage; when a bar touches both the stop and the target, the stop counts.
Position size: About $250 of risk a trade: micro contracts, rounded down, at least one. R is a trade's net P&L divided by its risk at entry.
Out of sample: The last 30% of weekdays: 2023-08-04 on (846 days), after 1,974 in-sample days. The rules were fixed before either side was run.

3Method

Our reading of the publicly described rules; the creator may trade it differently. The rules, exactly as the code runs them:

The pre-report range is the high and low of the range_minutes (15) one-minute bars before 10:30 ET.
At 10:30 ET on Wednesdays (days "wed"; "other" runs every weekday but Wednesday, the control), a buy stop goes in one tick above the range and a sell stop one tick below it. When one fills, the other is cancelled. Unfilled orders expire at 10:45 ET.
Stop: the other side of the range.
Target: target_r (1) times the risk.
Exit: the stop, the target, or flat at 12:00 ET.
Size: about risk_usd dollars of risk a trade (micro contracts, rounded down, at least 1 and at most max_contracts).

Weeks with a Monday holiday move the report to Thursday; this version still trades Wednesday.

bot.py, the file the engine ran

1"""Crude oil around the EIA inventory report (Wednesdays, 10:30 ET), on Micro Crude Oil (MCL) 1-minute bars.2 3The study measures crude's range in the half hour after 10:30 ET by weekday; the rule trades a break4of the pre-report range.5 6Rules:7  1. The pre-report range is the high and low of the `range_minutes` (15) one-minute bars before 10:30 ET.8  2. At 10:30 ET on Wednesdays (`days` "wed"; "other" runs every weekday but Wednesday, the control), a9     buy stop goes in one tick above the range and a sell stop one tick below it. When one fills, the10     other is cancelled. Unfilled orders expire at 10:45 ET.11  3. Stop: the other side of the range.12  4. Target: `target_r` (1) times the risk.13  5. Exit: the stop, the target, or flat at 12:00 ET.14  6. Size: about `risk_usd` dollars of risk a trade (micro contracts, rounded down, at least 1 and at15     most `max_contracts`).16  Weeks with a Monday holiday move the report to Thursday; this version still trades Wednesday.17"""18import qy19 20 21class Bot(qy.Strategy):22    """Straddle the pre-report range when the EIA number comes out."""23 24    symbol = "MCL"25    timeframe = "1m"26    session = None27    params = {"range_minutes": 15, "target_r": 1.0, "days": "wed", "risk_usd": 250, "max_contracts": 50}28    max_position = 5029    entry_window = ("10:30", "10:45")30    flatten_time = "12:00"31    overnight = False32 33    def setup(self):34        self.hi = self.lo = None35 36    def _size(self, risk_points):37        per_contract = risk_points * self.point_value38        n = int(self.p.risk_usd // per_contract) if per_contract > 0 else 139        return max(1, min(self.p.max_contracts, n))40 41    def on_fill(self, fill):42        if fill.kind == "entry":43            self.cancel_all()44 45    def on_bar(self, bar):46        start = bar.dt.hour * 60 + bar.dt.minute47        first = 630 - self.p.range_minutes48        if start == first:49            self.hi, self.lo = bar.high, bar.low50        elif first < start < 630 and self.hi is not None:51            self.hi, self.lo = max(self.hi, bar.high), min(self.lo, bar.low)52        if start != 629 or self.hi is None:53            return54        hi, lo = self.hi, self.lo55        self.hi = self.lo = None56        wed = bar.date.weekday() == 257        if (self.p.days == "wed") != wed or bar.date.weekday() > 4 or not lo < bar.close < hi or not self.position.is_flat:58            return59        tick = self.tick_size60        qty = self._size(hi - lo + 2 * tick)61        f = (hi - lo) / bar.close * 10062        self.buy(qty, stop=hi + tick, stop_price=lo - tick, target_r=self.p.target_r, tag=f"EIA long | f={f:.4f}")63        self.sell(qty, stop=lo - tick, stop_price=hi + tick, target_r=self.p.target_r, tag=f"EIA short | f={f:.4f}")

4Results

Table 1. In sample (2016-01-11 to 2023-08-03) against out of sample (2023-08-04 to 2026-10-30), after costs.
Statistic	In sample	Out of sample	All
Trades	366	150	516
Net P&L	+$11,900	−$340	+$11,559
Profit factor	1.28	0.98	1.19
Average trade	+0.13R	−0.01R	+0.09R
95% interval of the average	0.01 to 0.25	−0.17 to 0.15	−0.01 to 0.18
t-statistic of trade R	2.10	−0.11	1.78
Win rate	55%	53%	54%
Average win / average loss	1.06	0.88	1.01
Max drawdown	$2,957	$4,314	$4,314
Sharpe ratio (daily, annualized)	1.81	−0.15	1.30
Longest losing streak	7	9	9
Long / short profit factor	0.92 / 2.15	0.82 / 1.15	0.90 / 1.74
Average holding time	13 min	22 min	16 min
Commission paid	$5,231	$1,958	$7,188

Cumulative P&LDrawdown from the highOut of sample from 2023-08-04

Drag across the chart to zoom into a period, double-click to reset.

Figure 1. Cumulative P&L after costs and its drawdown from the running high, 2016-01-11 to 2026-10-30, at about $250 of risk a trade. The shaded part is out of sample.

In sample (366 trades)Out of sample (150)Averages

Averages: +0.13R in sample, −0.01R out of sample.

Figure 2. The distribution of trade results in R, as a share of each side's trades. Bars are in sample; the line is out of sample.

Figure 3. Crude's average range from 10:30 to 11:00 ET by weekday, as % of price, with 95% intervals

In sampleOut of sampleQuintile mean, 95% intervalOLS fitDensity contours

Drag a box to zoom, double-click to reset.

Figure 4. Trade R against the pre-report range, log x-axis; the dashed line is the OLS fit (r = −0.04, p = 0.40, n = 516; Spearman ρ = 0.23). Out of sample alone: r = 0.03, p = 0.69. Squares are the means of five equal-count bins, with 95% intervals.

Table 2. The same rule with one change at a time, after costs.
Version	Trades	PF in sample	PF out of sample	Avg R out	Sharpe out	Max drawdown
Wednesdays (as tested)	516	1.28	0.98	−0.01R	−0.15	$4,314
Every weekday but Wednesday (the control)	1,823	0.82	0.83	−0.08R	−1.35	$30,741

5Robustness

We reran the strategy on a grid of 16 settings of its two main parameters, each on the full history with the same split. 16 were profitable in sample and 0 out of sample.

0.81

0.67

0.73

0.88

1.5

0.88

0.85

0.76

0.89

0.85

0.88

0.98

0.99

0.50

0.89

0.85

0.76

0.94

Pre-report range (minutes) across, target (R) up. Solid outline: the defaults we tested; dashed: the in-sample best.

Hover a cell for both sides of the split.

Pre-report range (minutes) 5.00 to 30.0 across · Target (R) 0.50 to 2.00 in depth · Profit factor, out of sample 0.67 to 0.99 as height and shade. Drag or use the arrow keys to turn it.

Drag to rotate; ctrl or ⌘ and scroll to zoom.

Figure 5. Profit factor across pre-report range (minutes) and target (R), as a map and as a surface of the same grid. Values above 4 are drawn at 4.

What would change it

Nothing on the grid helps. Its in-sample best, pre-report range (minutes) 10 and target (R) 0.5, had profit factor 1.41 in sample but 0.85 out of sample, against 0.98 for the defaults.

In-sample yearOut-of-sample year95% interval

Figure 6. Average trade by calendar year with its 95% interval: a year-by-year walk through the history. Lighter bars are in-sample years.

Figure 7. Average trade in three volatility regimes: terciles of the 20-session average of the daily range (% of the open), up to the day before the trade (full session). The cut-offs use the whole period, so this describes, it doesn't forecast.

Figure 8. Average trade by weekday and by the hour of entry (New York time), with 95% intervals.

Median25th to 75th percentile5th and 95thRuin: down $2,000

Figure 9. 2,000 simulated years of 252 sessions, drawn session by session from the whole backtest (days without trades included), with 48 paths drawn; red paths lost $2,000 at some point. Risk of ruin 14.0% [12.5%, 15.5%]; 72% of years ended up; median ending P&L +$1,007, 5th to 95th percentile −$2,017 to +$4,198.

6Prop-firm implications

The backtest's sessions run through the rules of 36 tracked 50K evaluations: 2,000 runs of up to 60 sessions each, with the drawdown and daily limits checked against each trade's intraday low. Intervals are Wilson 95% intervals of the simulation. A simulation of the past, not the chance that anyone passes.

Plan	Pass	Fail	Out of time	Median sessions
FundedNext Futures Flex Challenge 50K	1.2%	8%	91%	51
Alpha Futures Standard 50K Evaluation	0.2%	1%	99%	54.5
Alpha Futures Zero 50K Evaluation	0.2%	1%	99%	54.5
Apex Trader Funding 50K EOD Evaluation	0.2%	1%	99%	54.5
Apex Trader Funding 50K Intraday Trailing Drawdown Evaluation	0.2%	2%	98%	54.5
AquaFutures Beginner 50K Evaluation	0.2%	0.2%	100%	54.5
Bulenox 50K Qualification Account (Legacy, Option 1, trailing drawdown)	0.2%	0.3%	100%	54.5
Bulenox 50K Qualification Account (Legacy, Option 2, EOD drawdown)	0.2%	0.2%	100%	54.5
Earn2Trade Gauntlet Mini $50K	0.2%	1%	99%	54.5
Earn2Trade Trader Career Path TCP50	0.2%	1%	99%	54.5

Scale 0 to 10%.

Figure 10. Simulated pass odds by plan, best first, with each plan's fail and out-of-time shares and the median sessions to a pass.

7Limitations

Holiday weeks move the report to Thursday; the rule still trades Wednesday.
Fast markets after the release can slip more than one tick.
One market (MCL).
A backtest on the past, with fills simulated from one-minute bars; live fills, queues and outages differ.

8Conclusion

Worked in sample, failed out of sample: profit factor 1.28 before the split, 0.98 after. Nothing on the grid helps. Its in-sample best, pre-report range (minutes) 10 and target (R) 0.5, had profit factor 1.41 in sample but 0.85 out of sample, against 0.98 for the defaults. On this reading of the rules, with these costs, the historical record gives no reason to expect a profit.

9References

U.S. Energy Information Administration. Weekly Petroleum Status Report.