Best statistical software for handicapping/modelling

**Justin7** · 02-02-10, 11:40 AM

SPSS is a nice statistical package, but I don't think it is that good for modeling.

I prefer to manipulate databases in text formats, make smaller data sets, and run regressions on zunzun.com.

**brettd** · 02-02-10, 06:55 PM

thanks justin.

**mlb** · 02-03-10, 12:19 AM

interesting will have to check it out

**tltaylor89** · 02-03-10, 12:21 AM

Never tried zun zun

**brettd** · 02-03-10, 12:53 AM

Justin (or anyone else for that matter), is there any other software that you would recommended in being an assistance in handicapping/creating a model?

**Wrecktangle** · 02-04-10, 10:26 PM

Learn Excel. You can do 95%+ of what one can in a "real" programming language and the error trapping is better.

As yes, you can do regression too.

**brettd** · 02-04-10, 10:32 PM

I've been looking into this, and have been told that the statistical programming language of 'R' is ideal for modelling. The steep learning curve of its command line input in quite off putting though. However, have you had any experience with RExcel? It uses Excel as a user-friendly front end UI, with excel calling upon R, and then displaying results back in excel.

**MonkeyF0cker** · 02-05-10, 07:57 AM

I was about to recommend R, brettd. The command line is a benefit, not a hinderance really. I would recommend taking the time to learn it. It allows you to automate your tasks much easier than many other packages. It is also far more efficient than Excel. Effiiciency makes a huge difference when you're dealing with 140+ games on an NCAAB Saturday.

**roasthawg** · 02-05-10, 02:53 PM

Originally posted by Wrecktangle

Learn Excel. You can do 95%+ of what one can in a "real" programming language and the error trapping is better.

As yes, you can do regression too.

Only limited regressions still correct? Like 5 or so variables?

**roasthawg** · 02-05-10, 02:53 PM

Originally posted by Justin7

SPSS is a nice statistical package, but I don't think it is that good for modeling.

I prefer to manipulate databases in text formats, make smaller data sets, and run regressions on zunzun.com.

Why smaller data sets... will zunzun not take a lot?

**Sinister Cat** · 02-05-10, 03:42 PM

I use Easyreg. The interface is a little clunky, but's it's overall easy to use, and it's flexible and free.

**MonkeyF0cker** · 02-05-10, 03:43 PM

R is free too.

**roasthawg** · 02-05-10, 04:16 PM

Originally posted by Sinister Cat

I use Easyreg. The interface is a little clunky, but's it's overall easy to use, and it's flexible and free.

I have a question about this program... I looked at the site and found this:

However, the number of observations is stored in an integer variable, which in Visual Basic 5 is limited to a maximum value of 32767. Consequently, if you try to import a data set with more than 32767 observations you will get an error message, and EasyReg will crash.

Let's say I had 4 different stats for all 32 NFL teams... is that 128 "observations" or 32?

**Sinister Cat** · 02-05-10, 05:35 PM

Originally posted by roasthawg

I have a question about this program... I looked at the site and found this:

Let's say I had 4 different stats for all 32 NFL teams... is that 128 "observations" or 32?

"Observations" means the number of rows of data... so that would be 32 I guess.