In part one of this post, we laid out the parameters of this project. Our mission is to create a data model that can help us generate solid, cash game winning lineups for daily fantasy football. Our strategy for winning cash games is to build a lineup that limits variability so we will be looking for players that consistently produce a high floor of fantasy points. In order to do this, we need to acquire the data that we will use to develop and train the model.
In this series of blog posts, I will walk you through a project that I am working on in my spare time to learn how to utilize data analytics, specifically, predictive analytics. Fantasy football, specifically, daily fantasy football, is a hobby that I am fairly passionate about and has a ton of statistics available publicly. This makes it easy to get plenty of data to develop and test the model, as well as many opportunities to put the model into practice. In this first blog post, I set the objectives and plan for the project.
If you are like me, you find documentation to be the bane of your existence. However, you also find it super helpful when six months down the road one of your end users asks you how a certain measure is calculated. Without documentation, we have to go through the code (whether it be T-SQL or SSIS packages) to see how the ETL job is performing the calculation. Documentation allows us to provide an answer faster and with more confidence than just by reviewing the code.
If you have participated in the SQL Server community, you probably have found that it is very helpful and very friendly. In fact, it might even have inspired you to give back to the community. The only problem is that you may not feel like an expert, and may think you have to be at the Brent Ozar, Jes Borland, or Pinal Dave level in order to give back to the community. If that is the case for you, then read on to discover 5 ways to give back to the SQL Server community without being a SQL Server expert yourself.
Would you look at this? I am actually writing a technical post! Today’s post is going to cover how I troubleshoot an issue that occurs when you have more than one row in staging that matches up to a row in the data warehouse. We’ll go over the error, the code that causes it, and how I fix it. Join us after the break to see more!
Last night (8/17) was the first meeting of the WausaPASS SQL Server user group. We had 15 people turn out for the event, including chapter representatives from FoxPASS and MadPASS. The sponsor for the meeting was Redgate (B | T) and we ended up with a double prize night thanks to PASS providing a jump drive of the PASS Summit 2015 sessions. Overall the session ran very smoothly with only a few minor technical glitches that were a result of me not reading directions more than anything else. We even successfully broadcasted the meeting via Skype for Business and had a user from Canada join on. So yeah, we are international now 🙂
An ETL framework is one of the perfect tools in developing a data warehouse. It helps you remain consistent as you build the system. Development time is decreased because you are not re-writing the same code over and over again. A good ETL framework can also include error trapping and performance monitoring in the system. In this series of blog posts, I will build a basic ETL framework that will build the various databases and tables in our data warehouse, create the routines to extract, transform, and load data, and even include a data quality engine to enforce business rules on the data.
Business intelligence has been listed near the top of the fastest growing and highest paid IT disciplines for several years now. As a result, new practitioners are entering the field on a daily basis. However, the field is so broad with so many tools and technologies it can be hard to know where to start learning. This post outlines my top ten skills that I think every BI pro should have.
Being the only data professional in a company with over 600 employees can seem like a nightmare to some people. But to others, it is an everlasting playground of joy. I fall somewhere in the middle. There are days where I absolutely love my job. There are also days where I have to use every bit of willpower I have to get out of bed and go to work. In this post, I will attempt to quantify my love/hate relationship with being the only data professional on staff at my organization.
You have a vendor database that you need to develop a report against, or load the data into the data warehouse. The problem is that you have no idea what tables relate to which tables because the vendor didn’t define any foreign key constraints. Your mission: reverse engineer the database schema. Now of course you could use 3rd party tools to explore the database, but let’s earn our whip and fedora and do it by hand. Read on to see how I explore databases with nothing more than just SSMS and maybe SQL Profiler for those pesky databases.