As a biostatistician, unfortunately, the probability math does not suggest he will get better barring some miracle. It says that we have enough evidence (ie 20% of likely total AB (450-500? max) at best have occured this year, and that is assuming he even gets those AB because he is so awful) to be 95% confident (with an 8% margin of error either way--which still stinks.) that this is exactly who he is now. Which is a real problem.
Here is a great calculator for estimating how much data you need to adjust your confidence for any numerical prediction you might make.
How exactly are you applying this data? Just as an example, using Manny Machado, the tool (if I'm using it properly) would suggest there's a 95% chance that Manny Machado's wOBA finishes between 433 and 509. Meaning he will almost surely finish higher than 2021 MVP Bryce Harper, while also providing MUCH better defense and thus putting up one of the best individual seasons in recent memory.
At +425 to win the NL MVP right now, if you believe in this tool with baseball as an application, you could make some serious dough over on the betting markets.
It's just a simple power calculator, which lets you know how much data you need to make inferences in samples from a normally distributed population. And as you know being a physician, when you put together something like a clinical trial you have to figure out how to power your study to tell you how much data you need to make inferences from your conclusions as well as given the universe of data you can possibly collect. For TG, If he gets 450 AB (the universe of AB for him) this year we have a presumed sample size of about 22% already, allowing to say with 95% confidence, that his aggregate numbers are + or - 8% of where they are going to be. That's all it says. He stinks at the plate and will likely continue to stink barring any "magical" changes, (PED? Glasses? A radical change in approach? etc) Re MM it just says that we have enough data on him this year to bet with high probability that he is not going to come much off these numbers give or take 10% etc. unless something drastic happens (injury? pitchers walk him every time? he gets fed up with the Padres and it affects him?)...etc...all highly unlikely just like it is highly unlikely that TG will figure it out much to my deep dismay
As a biostatistician, unfortunately, the probability math does not suggest he will get better barring some miracle. It says that we have enough evidence (ie 20% of likely total AB (450-500? max) at best have occured this year, and that is assuming he even gets those AB because he is so awful) to be 95% confident (with an 8% margin of error either way--which still stinks.) that this is exactly who he is now. Which is a real problem.
Here is a great calculator for estimating how much data you need to adjust your confidence for any numerical prediction you might make.
https://www.qualtrics.com/experience-management/research/determine-sample-size/
How exactly are you applying this data? Just as an example, using Manny Machado, the tool (if I'm using it properly) would suggest there's a 95% chance that Manny Machado's wOBA finishes between 433 and 509. Meaning he will almost surely finish higher than 2021 MVP Bryce Harper, while also providing MUCH better defense and thus putting up one of the best individual seasons in recent memory.
At +425 to win the NL MVP right now, if you believe in this tool with baseball as an application, you could make some serious dough over on the betting markets.
It's just a simple power calculator, which lets you know how much data you need to make inferences in samples from a normally distributed population. And as you know being a physician, when you put together something like a clinical trial you have to figure out how to power your study to tell you how much data you need to make inferences from your conclusions as well as given the universe of data you can possibly collect. For TG, If he gets 450 AB (the universe of AB for him) this year we have a presumed sample size of about 22% already, allowing to say with 95% confidence, that his aggregate numbers are + or - 8% of where they are going to be. That's all it says. He stinks at the plate and will likely continue to stink barring any "magical" changes, (PED? Glasses? A radical change in approach? etc) Re MM it just says that we have enough data on him this year to bet with high probability that he is not going to come much off these numbers give or take 10% etc. unless something drastic happens (injury? pitchers walk him every time? he gets fed up with the Padres and it affects him?)...etc...all highly unlikely just like it is highly unlikely that TG will figure it out much to my deep dismay
I also enjoy your deep insight into analytics!