What are the OOTP Predictions for April 12 2019? You can find out what they are below. But, first, you can take a look at the results for yesterday.
I am happy to see that my Mets won, even though it cause the simulator to experience loss for predicting Atlanta. I know it is still a looooong season for the Mets. In the past, they would start unraveling right about now. I don't say this to wish on them, of course. It's just the way things happen with them, unfortunately. But, there's always hope!
Boston won. Wow! And the simulator predicted that Boston would win. They only won by one point, however. I know a win is a win. But, the simulator showed Boston out-scoring Toronto by about 40%.
I know that in playing 5,000 games, the law of large numbers would tend towards the 1.4 (40%) strength indicator. Still, you'd expect more of a run differential with this 40%.
As a reminder, the strength indicator (something I named) is the total runs scored divided by the total runs allowed.
It doesn't surprise me that San Diego won even though the simulator chose the Diamondbacks. The Padres have been a strong team thus far. They are first in their division. I know there are many more factors that go into the calculation but, the standings have to play some role in it, I would think.
Overall, the simulator predicted 7 out of 10 games correctly for yesterday. While this isn't bad, I can't help but wonder whether an AI engine should be doing a bit better. I get that AI is a broad field. But, the implication of calling something an AI engine sets up the scenario that the algorithm should learn somehow. In fact, I have stated in previous posts on this thread that I believed the simulator would learn being that it is an AI engine.
Only the developers of the software know this answer. For me, it's just a guess. I also realize that running the simulator 5,000 times brings the results closer to the expected values, whatever they happen to be. In other words, if Boston had a strength indicator of 1.4 and it were possible for them to play Toronto 5,000 games in a single day, Boston would be the likely winner. They only play a few games together for an entire season.
As an side experiment, I ran the simulator using the One Game Elimination option for 6 times for the Yankees vs. White Sox. With all six games the Yankees won. For the regular 5,000 games with these teams the strength indicator was 1.6 with the Yankee's predicted to win.
There are 13 games being played today. The Detroit Tigers vs. Minnesota Twins game is postponed, which would have made it 14 games. I did not run the simulator for these teams.
James is a data science writer who has several years' experience in writing and technology. He helps others who are trying to break into the technology field like data science. If this is something you've been trying to do, you've come to the right place. You'll find resources to help you accomplish this.