?? For Math/Excel Experts

Goldfinger · 05-10-2002, 09:16 PM

?? for peep or any of the math guys here:

i am new to using excel and have started putting some #'s together, but i have a problem:

if i take 5 nba totals for example on a given team:

210
200
205
213
181

i am trying to find out this teams average pts. per game which is 201.8

but i see that the 181 skews my findings ---

what do i need to do in order to find a more accurate "gauge" of this teams average pts per game ???

i think standard deviation has something to do with it but am not that familiar with formal statistics -- can anyone break it down in simple terms and point me in the right direction ???

thank you

jlpblade · 05-10-2002, 09:23 PM

use the median, i.e. sort them and grab the one in the middle, or toss out high and low and avg what's left

AussieVamp2 · 05-10-2002, 09:38 PM

Jeff,

I think you will find it quite likely Excel has a Median function ;-)

jlpblade · 05-10-2002, 09:50 PM

AV:

Spreadsheets are for lightweights, programmable db's are for power users[img]i/expressions/face-icon-small-wink.gif[/img]

Jeff

KingOfTheSquares · 08-17-2002, 04:28 AM

jlpblade: Oh yeah? Well FORTRAN has more verified math subroutines than any language or environment. Why are you using it?

Spreadsheeting is programming for the common man. About 98% of anything you could want to do can be done in some fashion in Excel, Quattro, SuperCalc, etc, etc...even an old version of Visicalc.

AussieVamp2 · 08-17-2002, 08:02 AM

<< jlpblade: Oh yeah? Well FORTRAN has more verified math subroutines than any language or environment. Why are you using it?

Spreadsheeting is programming for the common man. About 98% of anything you could want to do can be done in some fashion in Excel, Quattro, SuperCalc, etc, etc...even an old version of Visicalc. >>

Not likely [img]i/expressions/face-icon-small-smile.gif[/img]

Although I will admit to not having used Fortran for awhile. If need something else, just grab a library somewhere and take it from that.

2006 · 08-22-2002, 01:38 AM

jpl,

what db do you recommend?

KingOfTheSquares · 08-22-2002, 01:49 AM

OK, Vamp. Name me some things that I can't do in Excel.

BTW, in FORTRAN there are well over 5000 subroutines that have been tested and verified (IMSL is one source) over the last 30 years or so, which BTW, comes with some of the latest compilers. Now, with all the wonderful code that a lot of today's "programmers" throw out there from scratch, I wonder how much of that spagetti has been tested?

Or are you going to tell me you write "bullet-proof" code?

jlpblade · 08-22-2002, 03:48 AM

KOTS:

I actually supported IMSL users at an aero outfit around '79-'80.

jlpblade · 08-22-2002, 03:53 AM

and my first spreadsheet was PerfectCalc in '83 that was bundled with my $7,000. pc. It did some business plans that turned out to be very successful, but I wouldn't want to use it now.

KingOfTheSquares · 08-22-2002, 12:48 PM

OK, blade: then you tell me what things you can't do in Excel...

You see, for the guy on the street, I think Excel can do most everything he/she needs. Sure you'll find power apps or languages that can work a problem larger/faster/"better" but for over 90% of common math problems, you can find a way to do it in Excel and usually do it faster. Because the big advantage is, you have a much shorter debugging cycle and thus can get to a problem solution in 1/10 the time for the bulk of most common problems. And it can be the way for the guy on the street to work almost every problem he can be interested in and for the tiny minority, you'll need a compiled language. Again, I'm talking math problems, not graphical/art/internet thingies, but things that I'd be interested in in the sports forecasting area...

...I don't think you can do it. All this "my language is better than your language" crap is simply the geek equivalent of my dad can kick your dad's a$$.

pokerjoe · 08-22-2002, 03:15 PM

Everything I've ever wanted to calculate for sports betting, I've been able to do with Excel, and without taking a course or even reading a book. And I write some damn long formulas, too. And, maybe most importantly, I have fun doing it. Learning a new language wouldn't be fun, and I wouldn't stick to it.

I think that anyone finding Excel inadequate for sports betting is falling victim to mis-precision and pseudo-sophistication.

AussieVamp2 · 08-22-2002, 04:11 PM

that's everything you ever wanted to do, though. [img]i/expressions/face-icon-small-smile.gif[/img]

The words Computer Group ring a bell? [img]i/expressions/face-icon-small-smile.gif[/img]

jlpblade · 08-22-2002, 06:59 PM

I wasn't trying to piss anyone off. My experience was that in the early days of pc software, dbs were the way to go vs spreadsheets. I just never went back.

jlpblade · 08-22-2002, 09:47 PM

and I didn't even mention APL [img]i/expressions/face-icon-small-smile.gif[/img]

KingOfTheSquares · 08-23-2002, 02:23 AM

Whoops, I just re-read my post and realized it sounds much more antigonistic than I meant. Apologies guys...

APL, "A Programming Language," yup, did that...all those stinkin' operators...wow, hadn't thought about that in years...you probably did ALGOL and SNOBOL, and what was that other one, with the ridgid column formatting...RPG? Shoot, I can't believe COBOL is still around...Y2K got a lot of those geezers jobs.

...geeze jlp, sounds like you're as old as me, you poor guy. I wouldn't even wish that on the Vamp...

jlpblade · 08-23-2002, 02:49 AM

KOTS: no problem. I just recently hit 45. For me, it was FORTRAN, APL, BASIC, PASCAL, some DOD language for a USMC project, and SQL lite.

Machiavelli · 08-24-2002, 06:59 AM

For my style of research, Excel does just about everything I need. Like someone else said, though, you sometimes have to "find a way" to trick it into doing what you want.

Machiavelli · 08-24-2002, 07:04 AM

Getting back to the orginal question.

Something like a median is probably what you want. You could average the "inner 50%" of the data points, dropping the higher 25% and the lower 25%, or something like that.

But those who were at my table at the MW Dinner at the Palm know that the complicated part starts after you're satisified with your average, then what do you do?? [img]i/expressions/face-icon-small-smile.gif[/img]