d2jsp
Log InRegister
d2jsp Forums > Off-Topic > Sports Coliseum > Baseball > Toronto Blue Jays Fan Thread
Prev19849859869879881469Next
Add Reply New Topic New Poll
Views: 160563Replies: 14682Track Topic
Member
Posts: 4,554
Joined: Dec 1 2008
Gold: 0.50
#9851
Jul 7 2013 06:49pm
Quote (Wine @ Jul 7 2013 07:23pm)
No. I have a brain and I use it to interpret statistical data, not like what you are doing and taking one game samples or looking at a player's batting average....this is the year 2013.
Every single time you post  here we all question whether you are a Nazi, Troll, or borderline retarded.

Please share how you determine a batters offensive worth and enlighten us with how we should do it too.


What qualifies you to be a statistician? Your degree in Hebrew Studies from York?


If I knew 100% how to determine a batter's offensive worth, I'd have a job in MLB. Fact is no one knows. Every hypothesis test I've done shows that AVG is a significant variable in a teams overall wins. I've never compared the two variables, but I have read studies which suggest OPS is slightly less significant than AVG. If you want to prove it wrong, run the tests yourself and post the results. I'd be interested to see.

If it's not clear that I am a neo-Nazi, maybe you're the one who's borderline retarded.
Member
Posts: 14,112
Joined: Jan 24 2008
Gold: 0.00
#9852
Jul 7 2013 06:58pm
Quote (Furio @ Jul 7 2013 08:49pm)
What qualifies you to be a statistician? Your degree in Hebrew Studies from York?


If I knew 100% how to determine a batter's offensive worth, I'd have a job in MLB. Fact is no one knows. Every hypothesis test I've done shows that AVG is a significant variable in a teams overall wins. I've never compared the two variables, but I have read studies which suggest OPS is slightly less significant than AVG. If you want to prove it wrong, run the tests yourself and post the results. I'd be interested to see.

If it's not clear that I am a neo-Nazi, maybe you're the one who's borderline retarded.


I would argue that BABIP and OPS in combination are better than average by a significant margin in predicting success in the long term....
Member
Posts: 66,487
Joined: Aug 18 2009
Gold: 1,003.55
#9853
Jul 7 2013 06:59pm
Member
Posts: 4,554
Joined: Dec 1 2008
Gold: 0.50
#9854
Jul 7 2013 07:05pm
Quote (xVitality @ Jul 7 2013 07:58pm)
I would argue that BABIP and OPS in combination are better than average by a significant margin in predicting success in the long term....


Then argue it with some data.
Member
Posts: 66,487
Joined: Aug 18 2009
Gold: 1,003.55
#9855
Jul 7 2013 07:08pm
Quote (Furio @ Jul 7 2013 09:05pm)
Then argue it with some data.


I'll take more power, extra base hits and a higher rate of getting on base via walks than hitting more singles
Member
Posts: 4,554
Joined: Dec 1 2008
Gold: 0.50
#9856
Jul 7 2013 07:11pm
Quote (Moofy @ Jul 7 2013 08:08pm)
I'll take more power, extra base hits and a higher rate of getting on base via walks than hitting more singles


I'm not going to argue with people who provide no supporting evidence to their claims. I'll run a multiple linear regression model tomorrow with season total runs as the dependent variable. I'll publish the results whether I'm right or wrong and put this shit to rest.
Member
Posts: 14,112
Joined: Jan 24 2008
Gold: 0.00
#9857
Jul 7 2013 07:15pm
Quote (Furio @ Jul 7 2013 09:11pm)
I'm not going to argue with people who provide no supporting evidence to their claims. I'll run a multiple linear regression model tomorrow with season total runs as the dependent variable. I'll publish the results whether I'm right or wrong and put this shit to rest.


What CI you going to use..? Cause the deviations could be HUGE.
Member
Posts: 4,554
Joined: Dec 1 2008
Gold: 0.50
#9858
Jul 7 2013 07:21pm
Quote (xVitality @ Jul 7 2013 08:15pm)
What CI you going to use..? Cause the deviations could be HUGE.


I'll do it in Gnumeric by default I believe it calculates 95, and 99. I'll add a column for 90 and we'll have all 3 industry standard levels.

It also lists R^2 and multiple R^2 so we'll have a fairly accurate idea of how much of the variables our model includes.

This post was edited by Furio on Jul 7 2013 07:23pm
Member
Posts: 50,348
Joined: Aug 12 2007
Gold: 80.00
Warn: 60%
#9859
Jul 7 2013 07:50pm
Quote (Wine @ Jul 7 2013 08:23pm)
No. I have a brain and I use it to interpret statistical data, not like what you are doing and taking one game samples or looking at a player's batting average....this is the year 2013.
Every single time you post  here we all question whether you are a Nazi, Troll, or borderline retarded.

Please share how you determine a batters offensive worth and enlighten us with how we should do it too.


Member
Posts: 4,554
Joined: Dec 1 2008
Gold: 0.50
#9860
Jul 8 2013 04:41pm
Ok... I was wrong. In fact OBP and SLG are extremely good indicators of offensive power. AVG is also a fairly good indicator but not as good. Note the OBP and SLG model has both a lower STDDEV and a higher multiple R^2.

Here are the results:

For AVG only (team stats):

Multiple R: 0.70431966926371
Standard Error: 45.0683411075515
Observations: 44

Intercept: -443.897652622573
Xavg: 4458.66130082263

Formula: Number of runs (season total) = -443.8977 + .AVG*4458.6613

For OBP and SLG:

Multiple R: 0.94696622926351
Standard Error: 20.6477629950155
Observations 44
Intercept -723.934529348815
Xobp: 2392.25822959176
Xslg: 1625.41885606391

Formula: Number of runs (season total) = -723.9345 + .OBP*2392.2582 + .SLG*1625.4189

Go Back To Baseball Topic List
Prev19849859869879881469Next
Add Reply New Topic New Poll

© 2003-2026 d2jsp  Contact