Predictive Analytics and Baseball Sabermetrics
The dictionary defines “sabermetrics” as the application of statistical analysis to baseball records, especially in order to evaluate and compare the performance of individual players. Henry Chadwick invented the baseball boxscore in 1859 and for more than 120 years, the data collected in box scores drove all statistical analysis about baseball. The traditional metrics for hitters (batting average, home runs, runs batted in, and stolen bases) seemed to be adequate measures of productivity. Oddly enough, the baseball establishment did not analyze what actually led to runs scored, preferring to use runs batted in (RBIs) as the default measure of value, when many factors outside of the individual batter’s control weigh into a player’s RBI total. Statistics like RBIs only looked backwards and were not an indicator from year to year about a player’s absolute ability to produce runs.
In 1977 Bill James published the first edition of the Baseball Abstract, and fans began to realize that despite the enormous amount of historical information available, the statistics being compiled weren’t the essential information necessary to predict success. Anybody who read “Moneyball” by Michael Lewis is aware that the Oakland A’s were the first team to fully embrace the use of non-traditional statistics as a way to gain a competitive advantage. Concepts like fielding range, on base percentage plus slugging percentage (OPS), runs created, and pitches seen per at-bat were introduced and changed analysis and behavior in a game that hadn’t seen change in more than a century.
Change in predictive analytics for measuring reserve requirements has been slow also. Computer spreadsheets like Microsoft Excel (introduced in 1985) are the tool of choice for many large companies tracking trends and assessing future needs. While Excel is an excellent tool, it is akin to using batting average to understand offensive production rather than runs produced.
After is changing the game. We incorporate advanced techniques and predictive modeling methods to help companies manage their businesses more efficiently. Many large manufacturers struggle with complex data sets from a wide range of sources, and as a result they tend to be overly conservative in their reserves because they don’t have the right data to prove otherwise. We consolidate and interpret disparate data sources to feed our models and provide actionable recommendations to our partners.
The core of our modeling is our team of dedicated statisticians and analysts who are highly trained in specialized statistical techniques, graphical devices, and advanced reporting tools. The results are standard and customizable reports that give accurate predictive insights into your data, allowing you to manage your business more dynamically.
The benefits of right-sizing reserves are numerous, but the most obvious is the efficient use of capital. In the absence of reliable forecasts, companies tend to be overly conservative and maintain an excessive reserve, which ties up capital. The repercussions of being over-reserved affect cash-flow, pricing and profitability. But without reliable forecasts, even a conservative estimate can miss the mark, leading to insufficient reserves and forcing a negative balance sheet change, and again impacting profitability.
Another key area made easier with After’s tool-set is the analysis of “what-if” scenarios. After dramatically reduces the time required to determine how changes to key variables such as parts price increases or warranty coverage time intervals impact future expenses.
Additional benefits of our analysis often lead to sourcing parts differently and better vendor management. If you are still using Excel to manage your warranty reserves, you are missing out on advances in predictive analytics that can greatly benefit your operations. Let After analyze your reserve requirements and unlock hidden opportunities in your reserves.