Welcome to the North American Subaru Impreza Owners Club Saturday August 30, 2014
Home Forums WikiNASIOC Products Store Modifications Upgrade Garage
NASIOC
Here you can view your subscribed threads, work with private messages and edit your profile and preferences Home Registration is free! Visit the NASIOC Store NASIOC Rules Search Find other members Frequently Asked Questions Calendar Archive NASIOC Upgrade Garage Logout
Go Back   NASIOC > NASIOC General > Motorsports

Welcome to NASIOC - The world's largest online community for Subaru enthusiasts!
Welcome to the NASIOC.com Subaru forum.

You are currently viewing our forum as a guest, which gives you limited access to view most discussions and access our other features. By joining our community, free of charge, you will have access to post topics, communicate privately with other members (PM), respond to polls, upload content and access many other special features. Registration is free, fast and simple, so please join our community today!

If you have any problems with the registration process or your account login, please contact us.
* Registered users of the site do not see these ads.
Reply
 
Thread Tools Display Modes
Old 05-02-2010, 10:36 PM   #1
chibich
Scooby Specialist
 
Member#: 79134
Join Date: Jan 2005
Chapter/Region: MAIC
Location: missouri
Vehicle:
2004 STI
WRB

Default Formula One college senior final project, Need Help

background for those interested:
I am currently a senior at James Madison University and majoring in Finance. One of the classes that I'm taking right now is a 400 level finance course called Financial Modeling and risk analysis.The objective of the class is to be able to look at different situations (mainly dealing with financial returns on investment, valuing options, and etc) and be able to model the different variables that impact this situation(using excel and Oracle crystal ball) in order to predict the most likely outcome/ fair price. For those who haven't used crystal ball, it is an add on to excel used for modeling that allows the user to assign certain distributions to each variable, correlations that exist between certain variables, and then allows you to run simulations (monte carlo, latin hypercube, and etc) in order to get the range of possible outcomes.

Where I need help:
For an optional final project/ extra credit opportunity the professor has basically given us free reign to design a model based on anything we choose and the better the model the more extra credit we get (due friday and kind of getting a late start seeing that it is finals this week). My idea is to try and design a model that will predict the outcome of future formula one races (mainly using time series analysis). The problem is that I have only been watching formula one "religiously" for the past 2 seasons, and so I have some understanding of the main factors that can contribute to wins but I wanted to get some formula one veterans views on these variables.

The variables:
Since time is running out and I'll be busy studying for finals this week the plan right now is to only focus on the top probably 12-13 racers (or could just look at teams overall).
I am planning on looking at:
-past race results (over the past few seasons)
-Final time spreads on races
- drivers # of wins and weather conditions during those wins (most likely use dummy variables for wet vs dry)
- Motor Make and reliability of said motors and teams overall
- Maybe size of budget and years in formula 1 (running regression in order to get a r-square to see how wins are explained by these variables)
- I wasn't really planning on using length of pit stops because of the no-refueling ban, difference seems very minimal between teams (if I did though could assign a triangular distribution for mess ups/replacing noses and etc)
-qualifying position vs finishing position (assign a correlation between these variables)
- the probability of teams finishing (ex. virgin's gas tanks are too small this season and no news on them changing that)
-and maybe drivers past results at certain tracks.

Some of these variables may be harder to find and quantify than others and with the limited amount of time I may not be able to get all of them. I was wondering if other people could provide their insight into the top factors that they feel effect the results each race (maybe give top 5 or 6, and list in order of importance).

-thanks for future responses and sorry if this is wrong area for posting, just thought I would get best results here.
* Registered users of the site do not see these ads.

Last edited by chibich; 05-02-2010 at 11:51 PM.
chibich is offline   Reply With Quote
Old 05-03-2010, 12:18 AM   #2
OBShahn
Scooby Specialist
 
Member#: 117648
Join Date: Jun 2006
Chapter/Region: RMIC
Location: Denver, CO
Vehicle:
1999 Legacy Outback
Dark Blue

Default

You should also develop a weighting matrix based on past results to factor predicted finishing position related to qualifying position.

Some tracks are notoriously difficult to pass on so that matrix should give you a predicted likelihood of loosing a position relative to qualifying position. For those events also see, if possible, to develop a matrix to factor the average deviation from qualifying position to end of first lap position (a driver with a history of good starts will have a greater chance of winning from second on the grid than a driver with bad starts).

You should also see about factoring in the probability of the race being affected by weather. This would give weighting to your adverse conditions variables and then give you a possibility of factoring in number of race starts as well.
OBShahn is offline   Reply With Quote
Old 05-03-2010, 07:35 AM   #3
ptclaus98
Scooby Specialist
 
Member#: 87888
Join Date: May 2005
Chapter/Region: South East
Location: Chattanooga, TN
Vehicle:
1988 Toyota Supra
Boredom Personified

Default

Buy a subscription to Forix.
ptclaus98 is offline   Reply With Quote
Old 05-03-2010, 07:47 AM   #4
StuBeck
Scooby Guru
 
Member#: 19546
Join Date: Jun 2002
Chapter/Region: Tri-State
Location: Rochester, NY
Vehicle:
2009 Honda Fit

Default

Quote:
Originally Posted by ptclaus98 View Post
Buy a subscription to Forix.
This. Just throw the variables into the formula, and then use a random number generator when rain comes in. And not to be a pessimist, but I doubt you'll be able to get any useable formula out of this since every 10 years there is a massive regulation change that throws everything in a loop.
StuBeck is offline   Reply With Quote
Old 05-03-2010, 09:54 AM   #5
chibich
Scooby Specialist
 
Member#: 79134
Join Date: Jan 2005
Chapter/Region: MAIC
Location: missouri
Vehicle:
2004 STI
WRB

Default

Quote:
Originally Posted by ptclaus98 View Post
Buy a subscription to Forix.
Thanks, this will help out a lot.
chibich is offline   Reply With Quote
Old 05-03-2010, 09:56 AM   #6
chibich
Scooby Specialist
 
Member#: 79134
Join Date: Jan 2005
Chapter/Region: MAIC
Location: missouri
Vehicle:
2004 STI
WRB

Default

Quote:
Originally Posted by StuBeck View Post
This. Just throw the variables into the formula, and then use a random number generator when rain comes in. And not to be a pessimist, but I doubt you'll be able to get any useable formula out of this since every 10 years there is a massive regulation change that throws everything in a loop.
Yea, I don't think i'm going to get anything great out of this, but it is extra credit so all I really need is something that works and shows I put some deccent effort into it. Plus its something i'm interested in so don't really mind doing the work for it.
chibich is offline   Reply With Quote
Old 05-03-2010, 12:26 PM   #7
ptclaus98
Scooby Specialist
 
Member#: 87888
Join Date: May 2005
Chapter/Region: South East
Location: Chattanooga, TN
Vehicle:
1988 Toyota Supra
Boredom Personified

Default

Come back and show us your results.
ptclaus98 is offline   Reply With Quote
Old 05-04-2010, 04:51 PM   #8
OnTheGas
Scooby Specialist
 
Member#: 3807
Join Date: Jan 2001
Chapter/Region: SCIC
Location: Orange, CA, USA
Vehicle:
2002 WRX Sedan
Silver, tinted

Thumbs up

Quote:
Originally Posted by chibich View Post
The variables:
Since time is running out and I'll be busy studying for finals this week the plan right now is to only focus on the top probably 12-13 racers (or could just look at teams overall).
I am planning on looking at:
-past race results (over the past few seasons)
-Final time spreads on races
- drivers # of wins and weather conditions during those wins (most likely use dummy variables for wet vs dry)
- Motor Make and reliability of said motors and teams overall
- Maybe size of budget and years in formula 1 (running regression in order to get a r-square to see how wins are explained by these variables)
- I wasn't really planning on using length of pit stops because of the no-refueling ban, difference seems very minimal between teams (if I did though could assign a triangular distribution for mess ups/replacing noses and etc)
-qualifying position vs finishing position (assign a correlation between these variables)
- the probability of teams finishing (ex. virgin's gas tanks are too small this season and no news on them changing that)
-and maybe drivers past results at certain tracks.

Some of these variables may be harder to find and quantify than others and with the limited amount of time I may not be able to get all of them. I was wondering if other people could provide their insight into the top factors that they feel effect the results each race (maybe give top 5 or 6, and list in order of importance).
You're on the right track...
  • As Stu alluded to, it is worth while to break out the wet versus dry races. To your favor, I believe wet races may be "declared" in the official results. The results of wet races can be significantly different than dry races due to the abilities of the drivers, the characteristics of the chassis, and motor maps to adjust to the wet conditions. As a general rule of thumb, wet race successes don't necessarily translate in to dry race success, (although your study could show that supposition to be wrong).
  • I would include the motors as a primary factor.
  • I would include the chassis (aka the team). In prior years, the chassis & motor was a significant influence in the results. With the new rules this year (no refueling), the chassis & motor may be a little less significant, (driver is more significant), but it is too early to determine that (I believe anyway).
  • Use the starting grid position versus finishing position data.
  • It may be worthwhile to examine the difference between the average result of the both cars of a team versus the result of a driver against that average.
  • The starting grid position is not relevant except in regards to the finishing position at the end of the race.
  • The pit stops over the last few seasons were primarily dictated by refueling strategies, (which don't apply this year), and by opportunites to pass someone slower via clear air running on off-sequence stops (so to speak). That later rationality still applies, but was secondary in prior seasons. You are correct to downplay the pit-stop data, to my mind.
  • On possibility to explore, to make your work easier/faster, is to focus on finishing positions in the top ten (since this year one only earns points for a top ten finish). So that if a car fails to finish in the top ten (or top 12 or top 14, whatever), and normally the car finishes in the points (which was top 8 last few seasons), then assume that motor broke, or some other mechanical failure, or driver error, or team error of some sort. If you focus on the results in that manner, you're not digging around for the official reason at each race as to why someone was classified not running. Sometimes the official reason for failing to finish a race is not necessarily true (for instance, the old racing joke that the stated "electrical failure" was actually that the motor threw a rod out the crankcase and physically knocked the wiring loom off ).
Forix.com is a good idea, and they sell a monthly access for a few bucks, I believe.
In addition, Formula1.com is a resource, as are the official race results at FIA.com. Autosport.com has official results (as of the time of publication) at the end of their race reports, but I don't know if you have to be a subscriber to access their old race reports.
I like motorsport.com for their unfiltered team PR releases, and FIA reports, and that is free.

Hope that helps!
Best regards, Ken.
OnTheGas is offline   Reply With Quote
Old 05-04-2010, 07:36 PM   #9
chibich
Scooby Specialist
 
Member#: 79134
Join Date: Jan 2005
Chapter/Region: MAIC
Location: missouri
Vehicle:
2004 STI
WRB

Default

thanks, got the 3 month access to forix and have compiled some data from formula1.com. Good news is I went and talked to the teacher today about my idea and he is a big formula 1 fan which could significantly help the amount of credit I get, bad news is finals are kicking my ass and with an all nighter planned tonight for my 2 finals tomorrow and 2 more finals back to back thursday I haven't been able to spend much time on this. Even if I don't finish it by friday I'll still probably try and finish it just to see how well it works in a real life situation, even though my ability to model is a severely limiting factor.
chibich is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
my senior project. please help ScottScamehorn Private 'Wanted' Classifieds 7 01-25-2010 11:43 PM
welp, it finally happened...need help WildBlueWRX Factory 2.0L Turbo Powertrain 23 07-28-2008 10:17 PM
School...Final Project...Need Subaru and Mitsubishi mobilezen Texas Impreza Club Forum -- TXIC 39 11-10-2006 11:51 PM
Wideband is finally in!!!!! need help GregBolby Normally Aspirated with bolt-on Forced Induction Powertrain 4 03-16-2005 05:45 PM
Project!!! Need help turning wrenches from Boston Area guys Hitokiri New England Impreza Club Forum -- NESIC 45 07-25-2004 04:40 PM


All times are GMT -4. The time now is 07:52 AM.


Powered by vBulletin® Version 3.7.0
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Powered by Searchlight © 2014 Axivo Inc.
Copyright ©1999 - 2014, North American Subaru Impreza Owners Club, Inc.