PDA

View Full Version : 31st oct Bendigo


Oaksnaf
31st October 2004, 01:53 PM
Race 5: #2 Metrol Lass

and thats it. There are a lot of dogs that just miss out on being a selection. but could only find one good one.

BJ
31st October 2004, 04:58 PM
Also my top rater but not a betting option for me. A bit worried about the #5. His name also helps sway me abit..... Bee Jay Brenic....

woof43
31st October 2004, 05:27 PM
Hi, just noticing that you use mps time adjustments, the problem with trying to even out the time from shorter distances to longer is that you will over compensate dogs racing over longer distances, giving them the impression of being able to run quicker then they actually can, the selections that you posted if you look at the avg distance raced by them is the highest in each race.

MPS adjustment systems selections will normally be at either end of the scale dogs with lowest avg distance raced or highest
Just food for thought

<font size=-1>[ This Message was edited by: woof43 on 2004-10-31 18:31 ]</font>

Oaksnaf
31st October 2004, 09:46 PM
yes woof i have had that problem. Usually i choose the common distance of the meeting.

BJ
1st November 2004, 01:06 PM
woof43, a very valid point. I try to compensate for this by overestimating the decline in speed over distance. Also I don't bet on anything where the average distance on previous races(3) is more than or less than 10% of the current race. Still learning and taking stats on this, I might indeed have to change this, maybe to within 5% or something.
Still looking at your other comments about standard deviation and so forth, but might take awhile to work out how to do all that in excel.

Oaksnaf
1st November 2004, 03:06 PM
standard deviation is probly the way. Quite a simplistic way to eliminate lost causes.

By the way. Race 5 #2 Metrol lass was unplaced.