The Lotus Cars Community banner

1 - 6 of 6 Posts

·
Registered
Joined
·
9,786 Posts
Discussion Starter #1
Data analysis is a very valuable tool... but care must be taken not to overlook sources of error or inaccuracy which could lead to misleading results and conclusions. For example, below is a series of my best sector times through the "Bus Stop" segment at Buttonwillow. Note that the fastest time (8.543 seconds) is nearly half a second faster than the next fastest time, after which they are all clustered together. 0.5 out of 9 seconds is a big jump... so it warrants further examination. If you note the max speed for that time, it is 208.61 mph... which is clearly impossible and would indicate a GPS error. Further examination made it clear that one end of that sector was misjudged by the GPS position data.

When calculating my theoretical best lap, I intentionally discarded that time, and went with the second best time. It would have been incorrect to use it.

This might be a good opportunity to check if a 2.280 second time amid a series of mid to high 3 second sectors times is erroneous as well. Sounds like it could add over a second to the theoretical best lap time... ;)
 

Attachments

·
Premium Member
Joined
·
4,697 Posts
Wowser, I knew you were fast but ....... :D
 

·
www.theapexinn.com
Joined
·
11,688 Posts
208 ! congrats ! :)
 

·
Registered
Joined
·
2,250 Posts
can you add statistical check and discard anything out side 1 standard deviation, and/or mark a 10% gain in red, 7% gain in yellow, -7% in orange and -10% with a smack on the knuckles for the driver..
 

·
Registered
Joined
·
9,786 Posts
Discussion Starter #5 (Edited)
can you add statistical check and discard anything out side 1 standard deviation, and/or mark a 10% gain in red, 7% gain in yellow, -7% in orange and -10% with a smack on the knuckles for the driver..
rotfl

I have other software that calculates the mean and standard deviation of say, speed or lateral Gs at any point on the track. It helps me determine where I'm being most consistent/inconsistent.

With any analysis software, it's possible for bad data to sneak in... I generally check max/min speeds, calculated engine power, distance traveled, etc. to check for anomalies... they do pop up from time to time.

It probably wouldn't turn out well if I was determined to reproduce that erroneous 8.5 second segment time... :eek:
 

·
Registered
Joined
·
175 Posts
Interesting. You can take the "mode" speed through corners and straights and come up with an ideal airfoil for down-force in corners to straightaway drag ratio based on the speeds on this particular track.
 
1 - 6 of 6 Posts
Top