Support Board
Date/Time: Wed, 18 Jun 2025 11:54:31 +0000
Post From: Sierra Chart SPY Volume off by 30-50%
| [2022-07-30 11:15:44] |
| Techotomy - Posts: 104 |
|
This is a follow up regarding the thread that was locked revealing unreliable volume data for the SPY in Sierra Charts (see SC Gives Different 1-Min volume for SPY than NYSE's Website). I believe it’s critical to the success of Sierra Chart customers to receive accurate and reliable data when making important financial decisions and am obligated as a fellow trader to continue this conversation in the hopes of improving the quality of Sierra Charts software and data feeds. I understand that Sierra Charts is planning to implement a new consolidated tape data feed that may provide correct reliable data for the SPY. Given that other major trading platforms with consolidated feeds also have volume reliability data issues and that the out-of-trade filter you are using cuts volume counts by 30-50%, I have reasons to doubt that a new feed will simply resolve the issue. I discussed receiving incorrect volume data while on the phone with one of your support members on 07/22/22. He did not have a valid license available to verify the volume count on his side at the time. The volume Sierra Charts showed me for the opening one-minute bar was 428K, 30% off compared to NYSEs volume count of 618K (see first attachment). The volume I received from Sierra Charts for the opening one-minute bar on 7/29/22 was 386K, 50% off compared to NYSEs volume count of 721K (see second attachment). The volume counts from NYSE agree with the volume counts provided by your own data source for this volume information, BarCharts, within 99%. Please note that Thailand time is being used for NYSEs website in the attachments. If your out-of-range filter is reducing volume counts by more than a few percent, I seriously urge you to reconsider the validity of your filter. Outliers in statistical data generally require their values to be more than 3 standard deviations away from the population mean such that 99.7% of the data should be retained. Even if you used a cutoff of 2 standard deviations, your volume count should not be less than 95% compared to NYSEs. Seeing 30% and 50% differences in volume counts leads me to seriously question the validity of financial data you are providing to Sierra Chart customers. “The volume is not going to change on a one minute bar…” The volume can change on a one-minute bar. I have reported and confirmed incorrect real-time volume count that have retroactively changed with due evidence across other major trading platforms, including eSignal and TradingView, who admitted they have an issue and have at least stated they are trying to fix it. Having a BS and MS in computer science, I have a sufficient technical background to say that a loss in the data feed should not lead to malformed volume information if proper data integrity and validation techniques are in place (e.g., data parity and checksums). The checksums in the TCP packets alone should ensure that any incidence of malformed data is rare. “Unless we know exactly every single detail related to this and have the access to the direct information, we cannot trust what we are being told.” You locked the thread preventing me from providing further information. Are you saying that it is impossible for me to provide sufficient evidence? If you need direct information, you can simply verify the volume information I have provided today on your side. |
 SC_WrongSpyVolume.JPG / V - Attached On 2022-07-30 11:15:00 UTC - Size: 456.43 KB - 145 views SC_WrongSPYVolume2.png / V - Attached On 2022-07-30 11:15:14 UTC - Size: 349 KB - 116 views |