World Series Analysis
Sept 21, 2020 4:18:42 GMT -8
Rays GM (Donavan), Cubs GM (Beau), and 4 more like this
Post by Rockies GM (Dan) on Sept 21, 2020 4:18:42 GMT -8
I was already thinking of doing this regardless of who was in the World Series. If I can somehow automate the more time consuming aspects of this, it would be cool to do it for at least one game per week next year. Donavan sent me a message saying he'd be interested in seeing what the simulation thought about the matchup, so I'm posting this.
This also kind of serves as a glimpse inside the inner workings of the simulation I created. I tweaked it so that instead of simulating the season 100,000 times, it just simulated this one matchup, with the results shown below. That's important to know looking at the box and whisker plots, that while there are numerous outliers leading to some staggering numbers (apparently, in 5 or so simulations, 0.00005% of the time, I accumulate over 25 HR), those are indeed far outliers. Far more important are the actual box and whiskers themselves.
I also only used players currently in the majors and not injured, with a few exceptions. It's very possible that the Rays might get Oviedo called up for a start tomorrow (but it's not a certainty), and they have some injured players that may return. I tended to err on the side of counting those players. Obviously, if they don't play, the numbers look a little different.
Also important to note is that OPS and WHIP are dependent on other simulated stats. So while the box and whisker charts here show the data indiscriminately, if the sim has either team with a particularly good week in HR and Average, the OPS will also be elevated, for example.
HITTING
PITCHING
I tend to have the edge on many offensive categories, whereas the Rays pitching gives them the best shot.
To answer Donavan's question, I first want to say that I don't put forth the following numbers as gospel. I'm tweaking the program to use it in ways it wasn't strictly intended for.
That being said, the simulation has the Rockies winning 72.98% of the time, whereas the Rays win 27.02%. Personally, I think it's even more of a tossup than those numbers portray, but I'm looking at the matchup subjectively.
Either way, good luck, Donavan. This is going to be a fun World Series.
This also kind of serves as a glimpse inside the inner workings of the simulation I created. I tweaked it so that instead of simulating the season 100,000 times, it just simulated this one matchup, with the results shown below. That's important to know looking at the box and whisker plots, that while there are numerous outliers leading to some staggering numbers (apparently, in 5 or so simulations, 0.00005% of the time, I accumulate over 25 HR), those are indeed far outliers. Far more important are the actual box and whiskers themselves.
I also only used players currently in the majors and not injured, with a few exceptions. It's very possible that the Rays might get Oviedo called up for a start tomorrow (but it's not a certainty), and they have some injured players that may return. I tended to err on the side of counting those players. Obviously, if they don't play, the numbers look a little different.
Also important to note is that OPS and WHIP are dependent on other simulated stats. So while the box and whisker charts here show the data indiscriminately, if the sim has either team with a particularly good week in HR and Average, the OPS will also be elevated, for example.
HITTING
PITCHING
I tend to have the edge on many offensive categories, whereas the Rays pitching gives them the best shot.
To answer Donavan's question, I first want to say that I don't put forth the following numbers as gospel. I'm tweaking the program to use it in ways it wasn't strictly intended for.
That being said, the simulation has the Rockies winning 72.98% of the time, whereas the Rays win 27.02%. Personally, I think it's even more of a tossup than those numbers portray, but I'm looking at the matchup subjectively.
Either way, good luck, Donavan. This is going to be a fun World Series.
