Adding Confidence to Your Trading Decisions

QUICK
SNAPSHOT
DAILY
CHANGES
DAYS OF
SUPPLY
Home Membership Research Trading Commentary Track Record Update Schedule Contacts
Log In

Forecasting Track Record

 

Market Variables

Month Dry Gas Production Imports Consumption Exports
EIA BGR Accuracy EIA BGR Accuracy EIA BGR Accuracy EIA BGR Accuracy
January, 2017 2,196,308 2,175,663 99.06% 291,628 275,853 94.59% 2,890,913 2,862,328 99.01% 271,849 265,494 97.66%
February, 2017 2,009,349 1,974,131 98.25% 254,479 242,705 95.37% 2,320,233 2,280,654 98.29% 255,021 248,384 97.40%
March, 2017 2,222,428 2,197,762 98.89% 280,534 265,938 94.80% 2,513,842 2,494,248 99.22% 271,816 261,778 96.31%
April, 2017 2,155,114 2,146,805 99.61% 238,248 244,075 97.61% 1,914,399 1,922,094 99.60% 247,228 235,285 95.17%
May, 2017 2,229,550 2,224,047 99.75% 244,099 251,113 97.21% 1,886,702 1,885,901 99.96% 253,841 255,288 99.43%
June, 2017 2,181,379 2,168,691 99.42% 239,776 234,838 97.94% 1,890,416 1,909,148 99.02% 253,305 244,993 96.72%
July, 2017 2,278,240 2,273,227 99.78% 250,481 249,065 99.43% 2,111,264 2,169,023 97.34% 248,420 257,647 96.42%

Last update: October 1, 2017

Comment: We tend to be very good at forecasting domestic variables (production and consumption), but not so good at predicting external variables (exports and imports). However, our track record in forecasting exports is much better than it is for imports.

Weekly Natural Gas Storage Report

Analyzed Data BGR Projection vs EIA Reported Figure** Trading Bias Accuracy***
Time Period Sample Size* 10+ 6-9 ≤5 Zero Total Bullish Bearish
December, 2015 - October, 2017 98 11% 26% 55% 8% 60% 56% 44%
January, 2017 - October, 2017 39 5% 26% 56% 13% 62% 38% 63%

* - number of reports;

** - simple difference in bcf, % share of observations;

*** - trading bias is calculated by subtracting BGR forecast from the consensus forecast (i.e., from market expectations as indicated in the Reuters survey); when BGR projection is below consensus, there is a bullish bias; when BGR projection is above consensus, there is a bearish bias; when there is a bullish bias and EIA then reports a figure that is below consensus, it is assumed that trading bias was predicted correctly. Same logic applies for situations when there is a bearish bias before the report and EIA releases a storage number that is above consensus. EIA revisions are excluded from trading bias calculations.    

Last update: October 10, 2017

Executed Trades

*total return excludes brokerage fees and holding cost; ** - NG_(Cash) is a CFD, provided by CMC Markets

Last update: October 16, 2017

Copyright 2015 - 2017 © Bluegold Research     [ Terms of Use ]   [ Contacts ]
Site by Smartum IT