The stock market is inherently volatile, and gaining even the slightest insight into its driving factors can lead to large reward. We looked to weather as an indirect contributor to the stock market. We performed a Markov Chain analysis on the last three years worth of weather data, correlated it with Dow Jones stock data, and found any signals we could (at a >58% threshold, with a reasonable number of instances >~50). The Markov chain takes into account the previous three days worth of data (we tested 2 and 4 as well, and decided on 3) in order to predict a change in the Dow Jones (trading price and volumes) the next day. We combined our analytical model with WeatherUndergrounds weather prediction to write a notification center app that lets the user know how the market may be poised to respond the next day. We tested the model we trained on the Dow Jones on the S&P500, and found small gains in all measures, but gains nonetheless! We integrated our app with the Bloomberg API to get its accuracy history, and allow you to check what the model would have predicted/how it would have faired on any date (with clear limitations). Clicking on the notification, your web browser pulls up the Bloomberg page for the Dow Jones and the WeatherUnderground page for New York, New York weather. As we see it, the app is more a proof-of-concept. You create a model that uses n-1 data points and access a prediction of data point n. If some criteria are met, the app makes a suggestion. The idea could be used in many fields, though we were interested in the data mining possibilities of finance and weather.